docs: rewrite the shared-expenses section, which had gone false

Every warning in it was inverted by this week's work:

  - "settled is dead data, false on every row" -- there are now 1,266 settled
    splits across 657 pre-2026 transactions.
  - "do not fix getParticipantBalances to exclude settled splits" -- it now
    excludes them, via ACTIVE_OBLIGATION, and must.
  - "settlement cannot be attributed per trip" -- migration 0022 added
    split_payments.trip_id and it is attributed.
  - "splits exist from 2026-01-09 only" -- pre-2026 transactions are now split
    deliberately, to stop them inflating spend.

Replaced with what is actually true, including the rule that matters most: the
cutover DATE is the primary balance gate and the settled flag only refines it,
so pre-2026 expenses can be split freely.

.gitea/workflows/ci.yml

@@ -13,6 +13,15 @@ jobs:
           node-version: 22
       - name: Install
         run: npm ci
+      # The Prisma client is generated into src/generated/prisma, which is
+      # gitignored — so a fresh checkout has no client and every test that
+      # reaches src/lib/db.ts dies on "Cannot find package
+      # '@/generated/prisma/client'". Schema-only, so it needs no database.
+      # Without this the pipeline had been red on every run since at least
+      # ae0c34f, which is how the failure stayed invisible: it looked like
+      # the normal colour.
+      - name: Generate Prisma client
+        run: npx prisma generate
       # Advisory until the pre-existing lint debt is cleared (2026-07-19:
       # ~20 errors across budget/insights/shared pages) — then make blocking.
       - name: Lint (advisory)

.gitignore

@@ -45,3 +45,7 @@ next-env.d.ts
 
 # Raw statement exports — real financial data, never commit
 dump/
+
+# Python tooling for scripts/ (split_csv_match.py)
+.venv/
+__pycache__/

AGENTS.md

@@ -0,0 +1,34 @@
+# Repository Guidelines
+
+## Project Structure & Module Organization
+
+Application code lives in `src/`. Next.js App Router pages and API route handlers belong in `src/app/`; reusable UI components are in `src/components/`; database access, query functions, hooks, authentication, and domain helpers are in `src/lib/`. Tests are separated into `src/__tests__/unit/` and `src/__tests__/integration/`. PostgreSQL schema and numbered SQL migrations live under `prisma/`, static assets under `public/`, operational scripts under `scripts/`, and design notes under `docs/`.
+
+Keep data flow consistent: API routes call query functions in `src/lib/queries.ts`, which use `queryRaw()` from `src/lib/db.ts`; client components access APIs through TanStack Query hooks in `src/lib/hooks.ts`.
+
+## Build, Test, and Development Commands
+
+- `npm ci` installs the locked dependency set (Node 22 is used in CI).
+- `npm run dev` starts the local Next.js development server.
+- `npm run build` creates a production build; `npm start` serves it.
+- `npm run lint` runs the Next.js ESLint configuration. Existing lint debt makes CI lint advisory, but new code should pass.
+- `npm test` runs fast unit tests.
+- `npm run test:setup` prepares the PostgreSQL test database using `.env.test`.
+- `npm run test:integration` runs database-backed tests.
+- `npm run test:all` runs both test suites.
+
+## Coding Style & Naming Conventions
+
+Use strict TypeScript, two-space indentation, semicolons, and double quotes, matching existing files. Name React components and types in PascalCase, functions and variables in camelCase, and files/routes in kebab-case. Use the `@/` alias for imports from `src/`. Preserve owner scoping and prefer transaction overrides with `COALESCE` in financial queries. Every API route must authenticate before accessing data.
+
+## Testing Guidelines
+
+Vitest is the test framework. Name tests `*.test.ts` and place pure logic tests under `unit/`; put PostgreSQL-dependent behavior under `integration/`. Add regression coverage for query, rule, reconciliation, and category changes. No numeric coverage threshold is configured; focus on meaningful edge cases and run `npm run test:all` before submitting database-related changes.
+
+## Database, Security & Configuration
+
+Add schema changes as the next numbered `prisma/migrations/NNNN_description/migration.sql`. Never commit `.env`, `.env.test`, raw statements in `dump/`, or other financial data. Consult `CLAUDE.md` and relevant `docs/` notes before changing splits, settlements, loans, reconciliation, or statement accounting.
+
+## Commits & Pull Requests
+
+History follows concise Conventional Commit-style subjects such as `feat(rules): preview rule changes`, `fix(trips): ...`, and `docs: ...`. Keep commits focused. Pull requests should explain behavior and data-model impact, link related issues, list validation commands, and include screenshots for UI changes. Ensure unit tests and the production build pass; call out any known lint warnings or migration steps.

CLAUDE.md

@@ -111,29 +111,64 @@ it silently deletes the unlogged remainder from spend totals.
 
 ### Shared expenses and settlement — read before touching
 
-The model is under active redesign. See `docs/shared-expenses-design.md` for the
+Rebuilt 2026-07-28. `docs/shared-expenses-design.md` describes the live model;
-proposal and what is already decided. Three traps:
+it is no longer a proposal. Everything the old version of this section warned
+about has changed — if you are working from memory of it, re-read.
 
-**`transaction_splits.settled` is dead data.** It is `false` on every row. Its
+**The cutover date is the primary balance gate, not the `settled` flag.**
-only writer was `/api/splits/settle`, removed in `3f04cbd` because nothing called
+`ACTIVE_OBLIGATION` (`src/lib/analytics-sql.ts`) is
-it and one request could mark all of a participant's splits settled. Do not build
+`ts.settled = false AND t.transaction_date >= '2026-01-09'`. Nothing dated
-on this flag until settlement contexts exist.
+before the cutover can ever be owed, because carryover transaction **2348**
+(dated 2026-01-09, $1,093.22) already carries the whole pre-cutover balance as
+one figure. The bound is **inclusive** — 2348 is itself dated 2026-01-09, so an
+exclusive bound would drop the carryover and the entire pre-cutover balance.
 
-**`getParticipantBalances` computes `splits − payments` and is correct.** Do not
+Consequence: **pre-2026 transactions can be split freely.** A split on a 2024
-"fix" it to exclude settled splits — the payments that settled them are still
+grocery shop describes how the expense was shared — which is what stops it
-subtracted, so you would double-count. The two settlement models (running tab vs
+inflating spend — without asserting a debt. 657 pre-2026 transactions carry
-per-split flag) must not be mixed.
+1,266 such splits, imported from SplitMyExpenses and marked `settled`.
 
-**Settlement cannot be attributed per trip.** `split_payments` records only
+**Spend counts settled splits; owed does not.** `myShare`/`mySplitOf` must NOT
-from/to/amount/date. Any per-trip settled/unsettled figure is fabricated; the
+filter on `settled` — half a 2025 grocery shop was your expense whether or not
-trip view used to show one and always reported 100% unsettled. Trips show share
+the other half was repaid. Filtering it out re-inflates exactly the figures the
-only, and point at `/shared` for real balances.
+historical import exists to correct.
 
-Also: settlements already exist twice. Four of eight `split_payments` match an
+**Any split write path that deletes-and-recreates must carry `settled` across.**
-offset-account credit exactly on amount and date, with `linked_transaction_id`
+`POST /api/transactions/[id]/splits` did not, and silently converted discharged
-populated on only one. And Sonu's loan contributions (`…emi` in the offset
+obligations into live debt — $37,233.28 was exposed. Fixed in `6add958`.
-account, 39 rows, $37,980.24) are categorised `transfers`, indistinguishable from
+`rule-actions.ts` is safe only by the shape of its upsert
-ordinary internal transfers.
+(`ON CONFLICT DO UPDATE SET share_percent` never touches the flag).
+The rules-apply revert route restores it explicitly.
+
+**Settling up is recording a payment.** There is deliberately no "mark settled"
+action. `settled` marks obligations discharged *outside* this app; doing both
+would subtract the settlement twice.
+
+**Payments carry scope, and one transfer can carry several rows.**
+`split_payments.trip_id` (migration 0022) says which tab a payment settles;
+NULL is the ongoing household tab. There is no unique constraint on
+`linked_transaction_id`, so a grouped transfer is recorded as one row per scope
+that re-add to the transfer — that is how Sonu's $3,779.33 and $4,794.06 were
+allocated Europe-first with the remainder to household.
+
+**Trip owed must be owner-scoped; trip cost must not be.** The owed query
+applies `OWNER_SCOPE`; without it a debt between the *other two* participants
+reads as owed to you ($1,605.49 on Europe 2026). Trip *cost* deliberately counts
+every payer — a trip cost what the group put into it — which is why the stat card
+says "all payers, net of refunds". Do not "fix" the missing scoping there.
+
+**Duplicates are superseded, never deleted.** `transactions.superseded_by_id`
+(migration 0023); 31 rows / $42,040.68 from overlapping ANZ statements 107/142/143.
+Every child of `transactions` is `ON DELETE CASCADE`. The exclusion lives *inside*
+`EXCLUDE_RECONCILED_SOURCE`, so any query applying that fragment gets it free —
+and any query that does not still double-counts.
+
+**Refunds:** a *partial* refund is netted in SQL (`NET_SPEND_ROWS`/`SPEND_SIGNED`);
+a *cancelled* booking has both legs untagged from the trip by hand, because a
+trip never incurred a cost it cancelled.
+
+**Trips:** Europe 2026 (id 1, 19 Mar–12 Apr), Auckland 2026 (id 2),
+Europe — Sonu + Sunny (id 3, 12–28 Apr, created 2026-07-28 from tag 5).
 
 **Partial split coverage inside a category is usually correct, not a gap.** Only
 *shared* items are split. `utilities` sits at 69% yours because Globird, OVO, GWW
@@ -144,9 +179,12 @@ Billdu, Spotify and Patreon are not. `fees` and `charity` are 100% yours and
 correct. Check the merchants before concluding a rule was never applied — a
 category-level ratio that "looks wrong" usually is not.
 
-Splits exist in this app from **2026-01-09** only; earlier splits lived in
+**Still true, and still a caveat:** Sonu's loan contributions (`…emi` in the
-SplitMyExpenses. So a trailing-12-month per-person series splices six months of
+offset account, 39 rows, $37,980.24) are categorised `transfers`, indistinguishable
-gross onto six months of net. Use Feb–Jun 2026 for anything per-person.
+from ordinary internal transfers. The loan model below is unbuilt.
+
+**`order_reviews` is a table wired to nothing** — 0 rows, no API, no UI, no
+writes. The "never order from here again" capability does not exist.
 
 ### The shared loan
 

docs/shared-expenses-design.md

@@ -1,57 +1,159 @@
-# Shared expenses, settlement, and the shared loan — design proposal
+# Shared expenses and settlement
 
-Status: **proposal, nothing built**. Written 2026-07-26 for review.
+Status: **built and live**, as of 2026-07-28. The loan section at the end is
+still a proposal — nothing there is built.
 
-## Why this exists
+This replaces the 2026-07-26 proposal. That document described three problems
-
+and proposed a `settlement_contexts` table to solve them. The problems were
-Three questions have no answer in the current model:
+real; the table was not built, and the reasoning for not building it is
-
+recorded under [What was rejected](#what-was-rejected).
-1. Which splits does a settlement payment settle?
-2. Is the Europe trip settled, separately from the ongoing household tab?
-3. Whose expense is a $2,500 loan repayment when Sonu funds part of it?
-
-They look like three problems. They are one: **the app records money moving, and
-separately records who owes whom, and the two never meet.**
 
 ---
 
-## What is actually broken
+## The one rule
 
-### Two settlement models, neither finished
+**Spend and owed are different questions asked of the same table, and the line
+between them is the cutover date, refined by `transaction_splits.settled`.**
 
-| Model | Where | State |
+`ACTIVE_OBLIGATION` is `ts.settled = false AND t.transaction_date >=
+'2026-01-09'`. Nothing before the cutover can be owed, because carryover
+transaction 2348 already carries the entire pre-cutover balance as one figure —
+so a split on an older transaction describes only *how an expense was shared*.
+That is what makes splitting history safe, and it is why the flag is a
+refinement rather than the guard: any delete-and-recreate write path resets a
+boolean, and one did.
+
+|  | Counts settled splits? | Why |
 |---|---|---|
-| Running tab | `split_payments` (from, to, amount, date) | **in use** — 8 payments, $37,881.10 |
+| **Spend** (`myShare`, `mySplitOf`) | **Yes** | Half a 2025 grocery shop was my expense whether or not the other half was ever repaid. |
-| Per-split flag | `transaction_splits.settled` | **never used** — all 673 splits are `false` |
+| **Owed** (balances, trip figures) | **No** | A discharged obligation is not outstanding. |
 
-They are honoured inconsistently:
+Getting this backwards in either direction is the failure the model exists to
+prevent. Filtering settled rows out of spend would re-inflate exactly the
+figures that importing settled history exists to correct.
 
-- `getParticipantBalances` (the shared page) ignores `settled` entirely
+The predicate is `ACTIVE_OBLIGATION` in `src/lib/analytics-sql.ts`.
-- `/api/participants/[id]/balance` filters on `settled = false`
-- `getTripAnalytics` reports settled/unsettled **from the unused flag**
 
-The third is a live bug. Every trip shows 100% unsettled forever, even though
+## Two orthogonal axes
-Molina has paid $20,782.79 against $19,556.07 of splits and is square.
 
-### Settlements exist twice, unlinked
+`settled` and `trip_id` answer different questions and neither implies the
+other:
 
-Four of the eight recorded payments match an offset-account credit exactly:
+- **`transaction_splits.settled`** — *is this obligation still live?*
+- **`split_payments.trip_id`** — *which tab does this payment settle?*
+  NULL means the ongoing household tab.
 
-| Payment date | From | Amount | Offset transaction |
+A trip can be fully paid while the household tab runs a balance, and vice
-|---|---|---:|---|
+versa. Before migration 0022 there was one global pool and this could not be
-| 2026-02-02 | Molina | 7,500.00 | `Transfer from - MEGHALEE BOSE mummy Pa…` |
+expressed, so every trip reported 100% unsettled forever — including trips paid
-| 2026-04-12 | Sonu | 3,779.33 | `Transfer from - MEGHALEE BOSE transfer` |
+in full.
-| 2026-04-21 | Molina | 1,685.24 | `Transfer from - MEGHALEE BOSE mummy split` |
-| 2026-05-16 | Sonu | 4,794.06 | `Transfer from - MEGHALEE BOSE transfer` |
 
-The same money is a `split_payments` row *and* a `transactions` row.
+## How settling up actually works
-`split_payments.linked_transaction_id` exists but only 1 of 8 rows uses it. So a
-settlement is bookkeeping that happens to resemble a bank credit, rather than
-being that credit.
 
-### The shared loan is invisible
+**By recording a payment.** There is deliberately no "mark settled" action
+anywhere in the app.
 
-Sonu's contributions are already in the ledger and unrecognised:
+`settled` marks obligations discharged *outside* this app — the imported
+SplitMyExpenses history, whose repayments happened on a platform we no longer
+run and which therefore have no `split_payments` row here. A live obligation is
+settled by recording the payment, and the balance nets to zero on its own.
+
+Doing both would subtract the settlement twice: the splits leave the sum *and*
+the payment is deducted, driving the balance negative by the amount repaid.
+
+## What is built
+
+| Piece | Where | Note |
+|---|---|---|
+| `settled` as the single balance gate | `ACTIVE_OBLIGATION` | Applied in both arms of the balances UNION and in the trip owed query |
+| Payment scope | `split_payments.trip_id` (migration 0022) | Household payments do not settle a trip, and vice versa |
+| Owner-scoped owed | `OWNER_SCOPE` in the trip owed query | Without it, a debt between the *other two* participants was reported as owed to the owner — $1,605.49 on Europe 2026 |
+| Direction on screen | `/trips/[id]`, `/shared` | all square / owes you / ahead — you owe them |
+| Historical splits | `scripts/split_csv_match.py` | 1,242 rows across 657 transactions, all `settled` |
+| Duplicate suppression | `transactions.superseded_by_id` (migration 0023) | 31 rows, $42,040.68 |
+| Overlap detection | `STATEMENT_OVERLAPS` → statements page | Red badge; catches the cause rather than the symptom |
+
+## The historical import
+
+The five SplitMyExpenses CSVs are the record of how expenses were shared before
+this app existed. 676 of 1,536 shareable rows matched (44%), and 1,242 split
+rows were written as `settled = true`.
+
+**The deliverable is historical spend, not balances.** $35,259 left my spend —
+$13,088 in 2024 and $22,117 in 2025 — because a $200 grocery shop that was
+always half hers no longer reads as $200 of mine. Balances were byte-identical
+before and after, which is the assertion that mattered.
+
+Three things the matcher has to get right, each of which has bitten:
+
+1. **Date format is decided per file.** The household export writes D/M/YYYY and
+   the four trip exports write ISO; 474 rows parse validly under both readings.
+   Guessing per row silently swaps January and February for some rows and not
+   others.
+2. **A person's column is net balance impact, not their share.** The payer is
+   whoever is positive; the other's share is `|their negative| / cost`. So a
+   `+cost / -cost` row means the other party owes **100%** — not that the
+   expense was unshared, which is the reading that fakes an arrangement change.
+3. **Matching is one-to-one, best pair first.** The NZ trip has two identical
+   $10.16 Uber rows against three ledger rows; without this a ledger row is
+   claimed repeatedly while the second CSV row looks matched and is not.
+
+**The 44% is a coverage ceiling, not a matcher weakness.** The CSVs describe 678
+shared expenses in 2024; the ledger holds 591 rows for all of 2024, 3 to 72 a
+month, far less than a household actually spends. South Korea April 2024 matches
+4 of 158. Chasing a higher rate is chasing transactions that were never
+imported.
+
+### A reversed recommendation
+
+The 2026-07-26 proposal said, under *What I would not do*: "**Do not** restate
+history from the SplitMyExpenses CSVs… the value is low: those balances are
+settled and will not change."
+
+That was overturned on 2026-07-28, and it was wrong in an instructive way: it
+measured the value in *balances*, where it is indeed nil, and missed the value
+in *spend*, where it is $35,259. Importing as `settled` gets the second without
+touching the first. The "combining problem" it cited is real and is why the
+match rate is capped — but a partial restatement of spend beats none, and rows
+that cannot be matched simply keep their current treatment.
+
+## What was rejected
+
+**`settlement_contexts` as a table.** The need was real — a payment must say
+what it settles. But trips already exist and already carry membership on
+`transaction_overrides.trip_id`, so scope is a read of existing data rather
+than a new grouping key. One nullable column on `split_payments` expressed it.
+
+A general context table would have meant a new entity to create and maintain
+before a payment could be recorded, in a two-person household with two trips.
+
+**Deleting duplicate transactions.** Every child of `transactions` is
+`ON DELETE CASCADE`, and which member of a duplicate pair holds the curation is
+an accident of import order. Duplicates are superseded instead: the row stays,
+keeps its children, and points at the row that replaces it.
+
+**Reusing `reconciled_with_id` for duplicates.** Its predicate is scoped to
+`statement_id IS NULL` on purpose — a statement line pointing at something else
+is the survivor, not the duplicate. In the duplicate-import case both rows are
+statement lines, so that predicate can never hide either.
+
+## Scale note
+
+This is a home app for one user, occasionally two, and the second user consumes
+the splits view and little else. Reviews of this subsystem have repeatedly
+proposed enterprise-grade reconciliation, lineage and audit machinery; the
+*findings* are often right and the *sizing* is not. A one-column solution a
+person can hold in their head beats a correct-but-unmaintainable one here.
+
+---
+
+## Still a proposal: the shared loan
+
+**Nothing in this section is built.**
+
+Sonu's contributions are in the ledger and unrecognised. All are categorised
+`transfers` — correct for spend, but it makes a loan contribution and an expense
+settlement indistinguishable:
 
 | Pattern in offset credits | Rows | Total | Meaning |
 |---|---:|---:|---|
@@ -59,162 +161,15 @@ Sonu's contributions are already in the ledger and unrecognised:
 | `…mummy…` | 6 | $29,721.24 | Molina's money, forwarded by Sonu |
 | other Meghalee | 15 | $71,130.27 | Sonu's own settlements |
 
-All are categorised `transfers` — correct for spend purposes, but it means a
-loan contribution and an expense settlement are indistinguishable.
-
-Meanwhile the loan itself, over the 12 imported months:
-
-| | |
-|---|---:|
-| Principal repaid (`investment`, excluded from spend) | $63,500.00 |
-| Interest charged (`loan_interest`, the only part counted as spend) | $16,523.64 |
-
-At roughly $25,000/year of `emi` against ~$80,000 of annual repayments, Sonu
-funds about **31%** — of both the equity being built and the interest being paid.
-Today 100% of the interest counts as your spend and 100% of the equity as yours.
-
----
-
-## The core problem
-
-The model conflates two different things:
-
-- **Money movement** — a credit landed in the offset account
-- **Obligation** — someone owed someone else, and now owes less
-
-A settlement is both. A loan contribution is both. Right now movement lives in
-`transactions` and obligation lives in `transaction_splits` / `split_payments`,
-with nothing joining them. That is why a payment cannot say what it settles: it
-was never attached to anything in the first place.
-
----
-
-## Proposed model
-
-### 1. Settlement contexts
-
-Splits belong to something that is settled **as a unit**. Payments name which
-unit they settle. Balance is computed per context, not globally.
-
-| Context | Splits | Settled by | State |
-|---|---|---|---|
-| Household (default) | ongoing | periodic payments | running tab |
-| Europe 2026 | trip-bound | lump sum | closeable |
-| Pre-2026 (SplitMyExpenses) | historical | settled elsewhere | **born closed** |
-
-A closed context still contributes to analytics — you see your true share — but
-contributes nothing to what anyone owes.
-
-This answers all three opening questions, and it dissolves the `splitFrom` date
-cutoff: pre-2026 splits can be applied retroactively **because they are born
-into a closed context**, so they fix the analytics without creating debt. No date
-guard needed, no risk of resurrecting settled obligations.
-
-Mechanically: `settlement_contexts` table; `transaction_splits.context_id`;
-`split_payments.context_id`. `transaction_splits.settled` becomes derived
-("is my context closed?") or is dropped.
-
-### 2. Payments are transactions, not a side table
-
-A settlement is the offset-account credit. `split_payments` becomes a thin
-attribution layer over a real transaction rather than a parallel record of it:
-
-- Populate `linked_transaction_id` on all existing payments where a match exists
-- On ingestion, an incoming credit that looks like a settlement is *proposed* as
-  one for confirmation, rather than silently becoming `transfers`
-- A payment with no matching transaction (cash, or an account not imported)
-  stays as a manual row — the model must tolerate that
-
-### 3. The shared loan — a separate ledger
-
-Not a settlement context. The loan is a jointly funded asset with its own
-obligation, and mixing it with expense settlement would let a contribution
-accidentally settle a dinner.
-
-- `emi` credits are recognised as **contributions**, not generic transfers
-- A **contribution schedule** states what is owed per period (50% of the
-  repayment), independent of what was actually paid
-- The running difference is a **receivable** — currently $4,000.00
-
-The schedule matters: during Sonu's leave the obligation did not change, only the
-payment did. A percentage-of-actual model would silently redefine her share as
-30% and make the shortfall disappear.
-
----
-
-## Migration path
-
-### Phase 0 — DONE (2026-07-26, commit `3f04cbd`)
-
-Stop the trip view reporting a settlement breakdown it cannot compute.
-
-The original plan was "make `getTripAnalytics` and `getParticipantBalances`
-agree". **That plan was wrong** and reading the code before building is what
-caught it:
-
-- `getParticipantBalances` is *not* buggy. It computes `splits − payments`,
-  which is coherent. Excluding settled splits there while still subtracting the
-  payments that settled them would have double-counted and broken a working page.
-- The real defect was narrower: the trip view showed Settled/Unsettled from
-  `transaction_splits.settled`, which nothing sets. A correct per-trip figure is
-  not computable at all, because `split_payments` has no trip attribution.
-
-So the fix was **subtractive**: the trip view now shows each participant's share
-and points at Shared for what is actually owed.
-
-Also removed `/api/splits/settle` — unreachable from the UI but live on its URL,
-where one call with `participant_id` would mark every one of that person's splits
-settled, writing a flag nothing reads.
-
-`transaction_splits.settled` / `settled_at` still exist and are now pure dead
-data. Phase 1 either repurposes them ("is my context closed?") or drops them.
-
-### Phase 1 — settlement contexts
-
-Add contexts; put every existing split in "Household"; every payment likewise.
-Balance queries group by context. Touches `queries.ts` (both balance CTEs),
-`shared/page.tsx`, `trips/[id]/page.tsx`, `split-payments/route.ts`. ~1 day.
-
-### Phase 2 — link payments to transactions
-
-Backfill `linked_transaction_id` for the four exact matches; flag the other four
-for manual linking. On ingestion, propose a matching credit as a settlement
-rather than silently categorising it `transfers`. ~half a day.
-
-### Phase 3 — retroactive pre-2026 split
-
-Create the "Pre-2026" closed context. Apply household split rules into it via the
-rule preview (`/api/rules/[id]/matches`, built 2026-07-26) — fixes ~$97,627 of
-the trailing 12 months currently shown as 100% yours. Then delete the `splitFrom`
-cutoff entirely.
-
-**Validate the ratio first.** This assumes today's 50/50 held through 2025. The
-SplitMyExpenses CSVs should be used to *check* that assumption — not to
-reconcile, since transactions were sometimes combined and exact matching is
-impossible.
-
-### Phase 4 — loan ledger
-
-Contribution schedule, contributions recognised from `emi` credits, running
-receivable. Independent of contexts — the loan is a separate ledger. ~1–2 days.
-
----
-
-## Decisions taken (2026-07-26)
-
 ### The loan is separate from shared expenses
 
 Different obligations, different rhythms, different nature: one funds an asset,
-the other funds consumption. They do not share a settlement context and a
+the other funds consumption. They do not share a settlement scope, and a
-contribution is never a settlement.
+contribution must never settle a dinner.
 
-### The share is 50/50, fixed — with the shortfall tracked
+### The share is 50/50 fixed, with the shortfall tracked
 
-Not derived from actual payments, which fluctuate. Sonu's obligation is half the
+Not derived from actual payments, which fluctuate. Over 2025-07-01 → 2026-06-30:
-repayment; the difference between obligation and actual is a **receivable**, and
-it is the interesting number.
-
-Over 2025-07-01 → 2026-06-30:
 
 | | |
 |---|---:|
@@ -223,65 +178,69 @@ Over 2025-07-01 → 2026-06-30:
 | Actually contributed (26 payments) | $27,750.00 |
 | **Shortfall** | **$4,000.00** |
 
-She never missed a fortnight; the rate changed:
+She never missed a fortnight; the rate changed — $1,250 × 15 (Aug 2025–Feb
-
+2026, the correct 50%), $1,000 × 3 (Jul 2025, pre-adjustment), $750 × 8
-| Rate | Payments | Period |
+(Mar–Jun 2026, leave).
-|---|---:|---|
-| $1,250 | 15 | Aug 2025 – Feb 2026 (the correct 50%) |
-| $1,000 | 3 | Jul 2025 (pre-adjustment) |
-| $750 | 8 | Mar – Jun 2026 (leave) |
 
 So the model needs a **contribution schedule** (expected per period) alongside
-actual contributions, with the running difference as a tracked balance. A flat
+actual contributions, with the running difference as a tracked receivable. A
-percentage cannot express "obligation unchanged, payment temporarily reduced,
+flat percentage-of-actual cannot express "obligation unchanged, payment
-difference owed".
+temporarily reduced, difference owed" — it would silently redefine her share as
+30% and make the shortfall disappear.
 
 ### Interest: recommended as expense, pending final call
 
-The mechanics are as described — interest is debited to the loan and repayments
+Over 12 months $63,500 of cash left and debt fell by $44,127.36. The $16,523.64
-pay down the combined balance. Reconciles exactly:
+difference bought nothing and is not recoverable — an expense by definition.
+Excluding it leaves the balance sheet unable to reconcile cash out against
+equity gained, and understates annual cost by ~10%.
 
-    134:  31,000.00 − 8,553.27             = 22,446.73  = balance reduction
+The legitimate concern is that interest is non-discretionary. The answer is a
-    133:  32,500.00 − 7,970.37 − 2,849.00  = 21,680.63  = balance reduction
+fixed-commitments grouping alongside rent, insurance and utilities — a
+presentation change, not an exclusion.
 
-But mechanics are not the same as economics. Over 12 months $63,500 of cash left
+**Do not** model the loan as a recurring split: that would put $2,500 a
-and debt fell by $44,127.36. The $16,523.64 difference bought nothing and is not
+fortnight of principal into spend, the error migration 0014 exists to prevent.
-recoverable — that is an expense by definition. Excluding it leaves the balance
-sheet unable to reconcile cash out against equity gained, and understates annual
-cost by ~10%.
-
-The legitimate concern is that interest is **non-discretionary**. The answer to
-that is a fixed-commitments grouping alongside rent, insurance and utilities —
-a presentation change, not an exclusion.
-
-**Recommendation: keep `loan_interest` as spend, add a fixed/discretionary
-split.** Flagged rather than settled: it is a judgement about what "spend" means
-in your own reporting.
 
 ## Open questions
 
 1. **Does equity need tracking per person?** If Sonu accrues a share of the
    principal, that is a balance-sheet item the app has no concept of. Probably
-   belongs in the net-worth view rather than here.
+   belongs in a net-worth view rather than here.
-
+2. **Attribution of forwarded payments.** `mummy` in the description reliably
-4. **Attribution of forwarded payments.** `mummy` in the description reliably
    marks Molina's money in all six known cases, but it is a description match on
-   a free-text field. Acceptable as a *suggestion* requiring confirmation; not as
+   a free-text field. Acceptable as a *suggestion* requiring confirmation, not
-   an automatic rule.
+   as an automatic rule.
+3. **The solo leg.** A Qantas booking on 23 Apr (txn 2849, $1,366.40) is the
+   flight to Bangkok that begins a solo leg, and the Singapore spending
+   ($2,242.05, 40 rows, to 9 May) is solo — not shared. It has no trip record.
+   Worth one if trip *cost* is wanted for it; nothing about sharing depends on
+   it.
 
-5. **Retroactive split ratios.** Applying today's household rules to 2025
+## Resolved
-   assumes the arrangement has not changed. The SplitMyExpenses CSVs could give
-   real historical shares, but transactions were sometimes combined, so matching
-   is imperfect. Recommendation: use today's ratios, accept the approximation —
-   the goal is a truer analytics picture, not a restated ledger.
 
----
+- **Europe — Sonu + Sunny** (trip 3, 2026-04-12 → 2026-04-28, 124 rows,
+  $9,914.24). The leg after the group trip, previously marked only by tag 5 and
+  invisible to trip analytics. Includes one advance booking on 17 Mar
+  (Ticketmaster Nanterre) and the 12 Apr handover-day rows, which were already
+  held out of Europe 2026.
 
-## What I would not do
+- **Grouped payments, split by scope.** Payments are made grouped — one transfer
+  covers several tabs — and that needs no schema change, because
+  `split_payments` has no unique constraint on `linked_transaction_id`. So one
+  bank transfer carries one row per scope, and the rows re-add to the transfer.
 
-- **Do not** restate history from the SplitMyExpenses CSVs. The combining problem
+  Sonu's two "transfer" payments were allocated Europe-first, remainder to
-  makes exact reconciliation impossible, and the value is low: those balances are
+  household, chronologically so each settles what was outstanding when it was
-  settled and will not change.
+  made:
-- **Do not** make the loan a shared *expense*. It is a funded asset. Modelling it
+
-  as a recurring split would put $2,500 a fortnight of principal into spend,
+  | Transfer | Scope | Amount |
-  which is the error migration 0014 was written to prevent.
+  |---|---|---:|
+  | $3,779.33, 12 Apr (txn 4111) | Europe 2026 | 1,084.61 |
+  | | Europe — Sonu + Sunny | 2,694.72 |
+  | $4,794.06, 16 May (txn 4121) | Europe — Sonu + Sunny | 1,145.52 |
+  | | household | 3,648.54 |
+
+  Both Europe tabs now read $0.00 and her overall balance is unchanged at
+  $5,428.08 — allocation moves money between tabs, never between people. That
+  invariance is the check worth repeating on any future re-allocation.

docs/ui-information-architecture-review.md

@@ -0,0 +1,715 @@
+# UI and information architecture review
+
+**Date:** 2026-07-26  
+**Status:** Priority 0 implemented 2026-07-27 (see below). Priorities 1–4 remain
+proposals.
+
+## Implementation status — Priority 0 (2026-07-27)
+
+All six Priority 0 items landed, with three amendments found while verifying the
+proposals against the code:
+
+1. **Reconciled source rows** — the exclusion was missing from *all five*
+   analytics routes, not only `/monthly`. It is now one fragment
+   (`EXCLUDE_RECONCILED_SOURCE`) that `queries.ts` also imports, so the two
+   halves cannot drift apart again. Real effect: 48 rows, **$4,474.79** of
+   double-counted spend removed from every category total, mover, Pareto and
+   merchant ranking.
+2. **Spend pace** — now served by `/api/analytics/daily`, built from the same
+   fragments as the headline. Measured on live data, the old client-side series
+   ended July at **$4,747.31** against a headline of **$3,597.10** — a 32%
+   overstatement of the number directly above it.
+3. **Fees and interest** — bounded by an explicit period (default 12 months,
+   `months=0` for all time), with the range shown and selectable. The unbounded
+   figure was overstating the last 12 months by roughly **$2,700 of fees**.
+4. **Split-coverage warning** — *deliberately not implemented* (user decision,
+   2026-07-27).
+5. **Shared foreign currency** — amended. The obvious fix, reading `s.currency`,
+   would have mislabelled every order row as AUD, because an order receipt has
+   no statement and carries its own currency. Sourcing is now
+   `NATIVE_CURRENCY = COALESCE(s.currency, t.foreign_currency_code, 'AUD')`,
+   whose COALESCE order keeps two opposite denomination conventions apart. Note
+   this change is **latent on today's data**: no foreign transaction is
+   currently split, so nothing on Shared looks different yet.
+6. **Partial-month comparisons** — the hero average, the top movers and the pace
+   baseline now exclude the in-progress month, and compare through the same day
+   of the month when the selected month is the current one.
+
+Also fixed while in here, both found by checking rather than by proposal:
+
+- **Every analytics window was a day early.** `toISOString()` on a local-midnight
+  `Date` converts backwards through UTC in any timezone east of Greenwich. Now
+  `toDateStr()`. This was pre-existing in `/monthly` and `/merchants`.
+- **Rounding grain.** `/monthly` rounded per category and `/daily` per
+  category-day, so the pace chart ended the month a few cents off its own
+  headline. Both now carry 4dp and round once, at display.
+
+Guarded by `src/__tests__/integration/analytics-sql.test.ts`.
+
+The doc's characterisation of `REGULAR_CATEGORIES` (Insights section) is also
+slightly off: the set has 13 members including rent, utilities, insurance and
+subscriptions, not the 8 listed. The case for replacing it stands — a flat
+binary cannot express obligation — but that is the reason, not arbitrary
+membership. Note too that the proposed Fixed/Essential/Lifestyle model needs a
+commitment dimension that does not exist yet: `fees` cannot be split into
+avoidable versus known-annual, `subscriptions` cannot be split into contractual
+versus cancellable, and the contracted loan repayment is not in the spend stream
+at all (`SPEND_BASE` keeps only the interest portion). That is a data-model
+change, not an Insights rework.
+
+## Executive summary
+
+The July 19 UI refresh gave the app a cohesive and distinctive visual identity.
+The ink-and-copper palette, typography, financial number treatment, month spine,
+and transaction drill-downs are all strong foundations.
+
+The larger remaining issue is not appearance. It is information hierarchy.
+Analytics and Insights contain useful data, but they are reporting-heavy rather
+than decision-oriented. Shared communicates the immediate running balance, but
+the current settlement model prevents it from answering which expenses a payment
+settled, whether a trip is closed, or how the shared loan should be represented.
+
+The product should make four questions easy to answer:
+
+1. Am I financially okay?
+2. What changed and why?
+3. What needs my attention?
+4. Who owes what, and for which expenses?
+
+Today there is no single page that answers the first three. The app opens on
+Transactions and presents ten equally weighted navigation items.
+
+The recommended direction is:
+
+- Add an Overview as the default landing page.
+- Keep Analytics focused on historical exploration: **what happened?**
+- Rebuild Insights around decisions and attention: **what should I know or do?**
+- Rebuild Shared around settlement contexts: **who owes what, and why?**
+- Keep the shared loan as a separate ledger from shared consumption expenses.
+- Fix calculation and coverage inconsistencies before adding more visualisations.
+
+## Context reviewed
+
+This review covered:
+
+- The current Next.js pages and shared components.
+- Analytics SQL and API calculations.
+- Shared-expense balance and transaction queries.
+- `CLAUDE.md`.
+- `docs/shared-expenses-design.md`.
+- `docs/expense-baseline.md`.
+- Recent repository history.
+- Recent finance-app memories retrieved from OpenViking.
+
+The OpenViking history confirmed:
+
+- The July 19 redesign intentionally introduced the ink-and-copper theme,
+  Fraunces display type, month-spine navigation, top movers, category
+  sparklines, and heat-tinted ledger tables.
+- The user prefers a modern, high-fidelity interface and actionable analytics.
+- Later July 25–26 work changed the financial meaning under those screens:
+  split-aware personal spend, AUD-aware settlement, refund netting, loan
+  principal/interest separation, rule previews, and the proposed contextual
+  settlement model.
+- The preferred settlement model links payments to real transactions, separates
+  Household, Trip, and Historical contexts, and keeps the shared loan separate.
+
+## What already works
+
+### Visual system
+
+- The dark ink-and-copper theme is coherent and distinctive.
+- Serif headings and mono financial figures create useful hierarchy.
+- The copper accent is used consistently for selection and emphasis.
+- The design feels like one application rather than a collection of unrelated
+  pages.
+
+### Analytics interactions
+
+- The month spine is an effective year-at-a-glance navigation control.
+- “What changed” is more useful than a generic category chart.
+- Category sparklines make direction visible without creating a large
+  multi-series chart.
+- Category rows can be expanded into their transactions.
+- Inline recategorisation allows users to correct the data while investigating
+  it.
+
+### Shared workflow
+
+- “Owes you,” “you owe,” and “all square” communicate the immediate relationship
+  balance clearly.
+- Payment history is preserved rather than reducing settlement to a boolean.
+- Participant and tag filters support practical investigation.
+- Split transactions can be edited without returning to the main transaction
+  page.
+
+## App-wide information architecture
+
+### Current problem
+
+The app redirects `/` to `/transactions`. This makes the operational ledger the
+default product surface. Transactions are important, but they do not tell the
+user whether anything needs attention or what the current financial position
+means.
+
+The sidebar also gives equal weight to:
+
+- operational screens such as Reconcile;
+- analytical screens such as Analytics;
+- configuration screens such as Rules;
+- organisational screens such as Tags.
+
+This makes the product feel like a database administration interface even when
+the individual pages are well designed.
+
+The `/budget` route is labelled Analytics in navigation. This is a leftover from
+an older product concept and should become `/analytics`.
+
+### Recommended navigation
+
+Group navigation by intent:
+
+**Overview**
+
+- Overview
+
+**Money**
+
+- Transactions
+- Statements
+- Reconcile
+
+**Understand**
+
+- Analytics
+- Insights
+- Merchants
+
+**Shared**
+
+- Shared
+- Trips
+- Loan
+
+**Organise**
+
+- Tags
+- Rules
+
+Lower-frequency configuration items can be visually separated or collapsed.
+
+### Recommended Overview
+
+The default landing page should be a concise status and attention surface, not
+another full analytics dashboard.
+
+Suggested structure:
+
+1. **This month**
+   - Personal spend to date
+   - Expected baseline at this point in the month
+   - Income
+   - Net cash
+
+2. **Financial resilience**
+   - Realistic monthly baseline
+   - Cash coverage in months
+   - Redraw shown separately from cash
+
+3. **Needs attention**
+   - Uncategorised or `other` transactions
+   - Unreconciled transactions
+   - Statements failing balance assertions
+   - New or unusual recurring charges
+   - Shared expenses added since the last settlement
+
+4. **Shared**
+   - Current balances by person and context
+   - Loan contribution shortfall shown separately
+
+5. **Recent change**
+   - The two or three categories that explain the largest movement
+
+The Overview should link into Analytics, Insights, Shared, and Reconcile rather
+than reproduce their complete tables.
+
+## Analytics review
+
+### What the current page does
+
+The current Analytics page includes:
+
+- selected-month spend hero;
+- twelve-month month spine;
+- income, expenses, invested, and net-cash strip;
+- top category movers;
+- eight category sparkline cards;
+- spend-concentration Pareto chart;
+- cumulative spend pace;
+- expandable category table;
+- six-month heat-tinted category ledger.
+
+Each component is defensible in isolation. Together, they create too many
+competing summaries of the same category data.
+
+### What Analytics should answer
+
+Analytics should answer:
+
+> What happened during this period, how does it compare, and what explains the
+> difference?
+
+Recommended primary structure:
+
+1. Period and comparison controls.
+2. Personal spend, income, invested, and net cash.
+3. Explanation of the change versus the selected comparison.
+4. One main category/trend visualisation.
+5. Category breakdown with transaction drill-down.
+6. An optional Explore section for detailed tables.
+
+### Recommended removals and consolidation
+
+- Keep either category sparklines or the six-month ledger as the primary
+  category-trend representation, not both.
+- Move the Pareto chart behind an Explore section. It describes concentration
+  but rarely produces an immediate decision.
+- Retain “What changed,” but make each item clickable and explain which
+  transactions caused the movement.
+- Avoid comparing a partial current month with full prior months unless values
+  are projected or compared through the same day.
+- Add gross-versus-personal-share switching only if it is clearly labelled.
+  Personal share should remain the default.
+
+### Calculation and trust issues
+
+#### Reconciled source rows can be double-counted
+
+`/api/analytics/monthly` does not currently exclude manual source rows where
+`reconciled_with_id IS NOT NULL`. The baseline analysis identified 48
+double-counted rows.
+
+The analytics query should apply the same reconciled-row exclusion used by the
+main transaction queries.
+
+#### Spend pace compares unlike numbers
+
+The Analytics headline uses:
+
+- split-adjusted personal share;
+- fees and interest;
+- refund and credit netting;
+- loan interest rather than principal;
+- non-spend-category exclusions.
+
+The cumulative spend-pace chart uses only `transaction_type === "debit"` and
+adds gross `amount_aud ?? amount`. It does not use personal share and does not
+apply the same refund, fee, interest, or loan semantics.
+
+The chart can therefore disagree with the headline while both appear to
+represent “spend.” The cumulative series should be produced by the same
+server-side spend semantics as the monthly total.
+
+#### Split coverage changes mid-series
+
+Reliable in-app split data begins on 2026-01-09. A trailing twelve-month personal
+series currently combines older gross spending with newer split-adjusted
+spending.
+
+Until historical splits are restored:
+
+- default personal trend analysis to February–June 2026;
+- visibly mark periods with incomplete split coverage; or
+- offer gross-only twelve-month comparison separately.
+
+Do not present the mixed series as one comparable personal-spend trend.
+
+#### Comparison baseline is too naive
+
+The selected month is compared against the average of all other months with
+data. That average can include travel, annual fees, tax payments, incomplete
+current periods, and months with incompatible split coverage.
+
+Better comparison choices:
+
+- previous month;
+- same month last year;
+- median of comparable complete months;
+- recurring baseline;
+- user-selected comparison.
+
+### Data trust indicator
+
+Analytics should include a compact methodology and coverage indicator:
+
+> Personal share · refunds netted · investments excluded · split coverage
+> reliable from Feb 2026 · 12 transactions need classification
+
+This makes the meaning of the numbers inspectable without overwhelming the page.
+
+## Insights review
+
+### Current problem
+
+The current Insights page contains:
+
+- Regular versus occasional spending;
+- another monthly category breakdown;
+- recurring charges;
+- fees and interest.
+
+The monthly breakdown duplicates Analytics. The page does not yet surface the
+most decision-relevant findings already known from the data: sustainable monthly
+cost, liquidity, the loan-overpayment lever, data-quality weaknesses, or unusual
+changes requiring attention.
+
+### “Regular” is not the same as committed or essential
+
+`REGULAR_CATEGORIES` includes:
+
+- groceries;
+- dining;
+- transport;
+- health;
+- personal care;
+- government;
+- charity;
+- pets.
+
+These may recur, but they have very different flexibility and obligation.
+“Regular” describes transaction behaviour, not financial necessity.
+
+The current chart therefore cannot answer:
+
+- What is the minimum monthly cost?
+- What can be cut?
+- What is contractually committed?
+- What is lifestyle spending?
+- What is a one-off?
+
+### Recommended model
+
+Replace Regular versus Occasional with:
+
+1. **Fixed commitments**
+   - Contracted loan repayment
+   - Insurance
+   - Rates and registration
+   - Known annual fees
+   - Contractual subscriptions
+
+2. **Essential variable spending**
+   - Utilities
+   - Groceries
+   - Transport
+   - Health
+
+3. **Lifestyle and discretionary**
+   - Dining
+   - Shopping
+   - Entertainment
+   - Personal care
+
+4. **One-offs and travel**
+
+5. **Investments and transfers**
+   - Shown for cashflow context, excluded from spending
+
+This should support scenario views rather than claiming there is one true
+baseline.
+
+### Recommended Insights structure
+
+#### 1. Financial baseline
+
+Show the scenarios already established by the expense-baseline analysis:
+
+- Survival: contracted loan repayment and essentials only.
+- Realistic: contracted loan repayment plus ordinary dining and charity.
+- Status quo: current loan overpayment and normal life excluding travel.
+
+For each scenario show:
+
+- monthly amount;
+- six-month reserve;
+- twelve-month reserve.
+
+#### 2. Liquidity and resilience
+
+Show:
+
+- cash available;
+- redraw available separately;
+- months covered under each baseline;
+- a warning that redraw is lender-controlled and not equivalent to cash.
+
+#### 3. Biggest flexible levers
+
+Examples:
+
+- voluntary loan overpayment;
+- dining;
+- shopping;
+- subscriptions;
+- travel.
+
+The loan should always show both the contracted floor and actual repayment.
+
+#### 4. Attention and anomalies
+
+Examples:
+
+- a new recurring charge;
+- a charge larger than its prior range;
+- a category materially above baseline;
+- a fee increase;
+- an unexpected incoming credit categorised as spend;
+- a merchant still classified as `other`;
+- an investment incorrectly counted as spending.
+
+Each insight should link directly to the affected transactions.
+
+#### 5. Data-quality work queue
+
+The baseline analysis found that data quality is currently a larger blocker than
+visualisation:
+
+- `other` remains a large unresolved category;
+- Raiz, Vanguard Super, and moomoo need investment classification;
+- incoming `other` credits can make spending negative;
+- `government` conflates tax with rates and registration;
+- annual fees distort short-window monthly averages.
+
+Insights should make these visible as fixable tasks.
+
+### Recurring charges
+
+The current detector identifies merchants with regular transaction intervals.
+That does not necessarily mean a subscription or commitment. Weekly grocery
+shopping can look recurring.
+
+Recommended changes:
+
+- Rename the section **Recurring patterns** unless contractual charges can be
+  distinguished.
+- Show confidence and the basis for classification.
+- Show the next expected charge date.
+- Separate likely subscriptions from recurring merchants.
+- Allow dismissing or confirming a detected pattern.
+- Highlight price changes.
+- Collapse inactive patterns by default.
+
+The current eight-column table is also too wide for a primary page. Put secondary
+fields such as first seen, total paid, and count into an expandable detail row.
+
+### Fees and interest
+
+The current fees query aggregates statement summary values across all available
+statements without a date filter. The UI does not label the period, so the total
+looks like a current-period figure even though it is effectively lifetime to
+date.
+
+Recommended presentation:
+
+- Explicit date range.
+- Avoidable fees.
+- Known annual fees.
+- Credit-card interest.
+- Loan interest.
+- Change versus prior comparable period.
+- Drill-down transactions.
+
+Loan interest should remain spending, but appear under fixed or
+non-discretionary costs rather than being hidden.
+
+## Shared review
+
+### What the current page answers well
+
+The unfiltered balance cards correctly implement a running ledger:
+
+> splits minus payments
+
+This is coherent and should not be changed to exclude splits marked `settled`.
+The `transaction_splits.settled` field is dead data and must not be used for new
+UI claims.
+
+### What the current model cannot answer
+
+- Which split expenses did a payment settle?
+- Is a particular trip settled?
+- Can a trip be closed without closing Household?
+- Is an imported offset-account credit already represented by a manual payment?
+- What remains open inside one settlement context?
+- How should the shared-loan contribution shortfall be shown?
+
+The page should not imply answers that the data model cannot support.
+
+### Tag-filtered balance cards are semantically misleading
+
+When a tag filter is active, participant balance queries intentionally stop
+subtracting payments because payments are not attributable to a tag. The cards
+then show raw split totals for the tag.
+
+This behavior is explained in small text, but the card still says “owes you” or
+“you owe.” That looks like a real payable balance when it is not.
+
+When filtered, relabel the cards:
+
+> Split total in Europe 2026
+
+Do not show payment or settlement actions from that state.
+
+### Recommended settlement-context design
+
+Use explicit settlement contexts:
+
+- Household
+- Individual trips
+- Historical / Pre-2026
+- Closed contexts
+
+Recommended Shared navigation:
+
+- All
+- Household
+- Trips
+- Closed
+
+Within a context show:
+
+1. Net balance and direction.
+2. Expenses added since the last settlement.
+3. Payments attributed to that context.
+4. A chronological activity ledger combining expenses and payments.
+5. Context status: running, ready to settle, or closed.
+6. Settlement action.
+
+### Payments should link to transactions
+
+An offset-account credit and a manual `split_payments` row can represent the same
+money. The page should:
+
+- propose matching an imported credit to a settlement;
+- display the linked transaction;
+- prevent silent duplication;
+- allow a manual payment only when no matching transaction exists.
+
+“Record Payment” should become a context-aware settlement flow:
+
+1. Choose what is being settled.
+2. Match an existing incoming transaction where possible.
+3. Confirm amount and residual balance.
+4. Preserve an auditable history.
+
+### Shared transaction table
+
+The table currently shows raw `tx.amount` with a dollar sign and no currency
+indicator. Participant balances correctly convert to AUD.
+
+For foreign transactions, show:
+
+- the native amount and currency;
+- the AUD equivalent;
+- splits based on the AUD settlement amount.
+
+This prevents a visible mismatch between transaction rows and participant
+balances.
+
+### Shared loan
+
+The loan is not a shared-expense settlement context. It funds an asset rather
+than consumption, and a loan contribution must never settle a dinner or utility
+bill.
+
+Give it a separate page or clearly separated ledger showing:
+
+- expected contribution by period;
+- actual contribution;
+- running shortfall or receivable;
+- principal reduction;
+- interest expense;
+- contracted repayment;
+- actual repayment;
+- voluntary overpayment;
+- redraw movement.
+
+The partner obligation is a fixed 50% of the repayment schedule, not a percentage
+inferred from actual contributions.
+
+## Responsive and interaction improvements
+
+- Replace wide eight-column primary tables with compact rows and expandable
+  details.
+- Keep financial summaries readable at mobile widths without horizontal
+  scrolling.
+- Add explicit loading skeletons rather than only text.
+- Add error states for failed analytics requests.
+- Ensure chart meaning is not conveyed by colour alone.
+- Give interactive chart regions keyboard-accessible equivalents.
+- Confirm material deletions, including payment-history deletion.
+- Make expandable table rows use buttons with appropriate accessibility state.
+- Use consistent labels for personal share, gross amount, native currency, and
+  AUD equivalent.
+
+## Recommended implementation order
+
+### Priority 0 — metric integrity
+
+1. Exclude reconciled source rows from monthly analytics.
+2. Make spend pace use the same spend semantics as the headline.
+3. Add date ranges to fees and interest.
+4. Add split-coverage warnings to historical personal-share analysis.
+5. Fix Shared foreign-currency presentation.
+6. Avoid partial-month versus full-month comparisons.
+
+### Priority 1 — product hierarchy
+
+1. Add Overview and make it the default route.
+2. Group sidebar navigation by user intent.
+3. Rename `/budget` to `/analytics`.
+4. Add consistent methodology and coverage indicators.
+
+### Priority 2 — Analytics and Insights
+
+1. Simplify Analytics around period, comparison, change explanation, trend, and
+   drill-down.
+2. Remove the duplicate monthly breakdown from Insights.
+3. Add baseline scenarios and liquidity coverage.
+4. Add flexible-spending levers, anomalies, and a data-quality work queue.
+5. Rework recurring patterns and fees into decision-oriented summaries.
+
+### Priority 3 — Shared
+
+1. Add settlement contexts.
+2. Link payments to real transactions.
+3. Add context activity ledgers and closeable trip contexts.
+4. Introduce the separate loan contribution ledger.
+5. Backfill historical closed-context splits so long-range personal analytics
+   become comparable.
+
+### Priority 4 — polish
+
+1. Improve mobile layouts.
+2. Add accessibility semantics.
+3. Add richer loading, error, and empty states.
+4. Consolidate repeated card, table, filter, and page-header patterns into shared
+   components.
+
+## Proposed success criteria
+
+The redesign is successful when:
+
+- The first page explains current status and outstanding actions without opening
+  multiple screens.
+- Analytics can explain why one comparable period differs from another.
+- Insights identifies baseline cost, financial resilience, flexible levers, and
+  data-quality problems.
+- Every displayed total states or clearly implies its period and whether it is
+  gross or personal share.
+- Historical charts do not silently combine incompatible split coverage.
+- Shared can distinguish Household, Trip, and Historical balances.
+- A settlement can be traced to both the obligation it reduces and the real
+  transaction representing the payment.
+- Loan contributions cannot affect ordinary shared-expense balances.
+

prisma/migrations/0021_order_platform_provenance/migration.sql

@@ -0,0 +1,39 @@
+-- Order provenance: which platform the receipt came from, and the message it
+-- came from.
+--
+-- The parser has always known the platform (it has to, to read the template)
+-- and then threw it away. Without it a transaction reads "Order - Burger
+-- Corner" with no way to tell whether to look in DoorDash or Uber Eats for the
+-- detail, and no way to answer "how much of this is DoorDash?" at all.
+--
+-- `source_email_subject` / `source_email_from` already existed for the
+-- Paperless expense path and were simply never populated by order ingestion.
+
+ALTER TABLE expense_metadata ADD COLUMN IF NOT EXISTS platform text;
+
+COMMENT ON COLUMN expense_metadata.platform IS
+  'doordash | ubereats | uber — the receipt template the order was read from.';
+
+-- Backfill the 101 rows written by the 2026-07-27 backfill. DoorDash receipts
+-- carry no order id of their own, so ingestion synthesises `msg:<message-id>`;
+-- Uber receipts carry a real trip UUID. That is the only surviving
+-- discriminator, and it is exact.
+UPDATE expense_metadata
+   SET platform = CASE WHEN order_reference LIKE 'msg:%' THEN 'doordash' ELSE 'ubereats' END
+ WHERE platform IS NULL
+   AND source = 'email'
+   AND paperless_doc_id IS NULL   -- exclude the Paperless expense path
+   AND order_reference IS NOT NULL;
+
+-- Pick-up / delivery stops, as the receipt prints them. Uber puts these on
+-- every order under `Order details`; DoorDash prints no addresses at all, so
+-- this stays '[]' there. Same block a *trip* receipt uses for start and
+-- destination, so this column already fits rides when they come into scope.
+ALTER TABLE expense_metadata ADD COLUMN IF NOT EXISTS route jsonb NOT NULL DEFAULT '[]'::jsonb;
+
+COMMENT ON COLUMN expense_metadata.route IS
+  'Uber only: [{label, time, address}] — pick-up and delivery stops as printed.';
+
+CREATE INDEX IF NOT EXISTS idx_expense_metadata_platform
+    ON expense_metadata (platform)
+ WHERE platform IS NOT NULL;

prisma/migrations/0022_settlement_scope/migration.sql

@@ -0,0 +1,48 @@
+-- Settlement scope: which tab a payment settles.
+--
+-- `split_payments` has carried from/to/amount/date since it was written and
+-- nothing else. That is the whole reason a per-trip balance has never been
+-- computable — `getTripAnalytics` says so in a comment where the figure should
+-- be: "split_payments carries no trip attribution, so a payment cannot be
+-- assigned to a trip. Settlement is a property of the whole relationship."
+--
+-- It is also the reason the Shared page silently drops payments the moment a
+-- tag filter is applied (`getParticipantBalances`): with one global payments
+-- pool there is no honest way to show a filtered balance, so it showed gross
+-- splits under the same label instead. A tag is a view; a scope is a ledger.
+--
+-- The scope is a *trip*, not a new `settlement_contexts` table. `trips` already
+-- has owner_id, dates and an archived flag, and `transaction_overrides.trip_id`
+-- already decides which transactions belong to it. A second grouping beside it
+-- would be two unsynchronised scopes over the same rows — a trip could hold a
+-- mix of contexts and a context could span trips, with no invariant saying
+-- which one governs.
+--
+-- NULL means the ongoing household tab. That tab never closes, which is why
+-- this is nullable rather than defaulted to some "general" row: absence is the
+-- honest representation of "not attached to a trip", and it keeps every
+-- existing payment correct without a backfill.
+
+ALTER TABLE split_payments
+  ADD COLUMN IF NOT EXISTS trip_id integer REFERENCES trips(id) ON DELETE SET NULL;
+
+COMMENT ON COLUMN split_payments.trip_id IS
+  'The trip this payment settles. NULL = the ongoing household tab.';
+
+CREATE INDEX IF NOT EXISTS idx_split_payments_trip
+    ON split_payments (trip_id)
+ WHERE trip_id IS NOT NULL;
+
+-- `settled` answers a different question and the two must not be collapsed:
+-- trip_id is *which tab*, settled is *is this obligation still live*. A
+-- pre-2026 historical split is settled with no tab; a Europe split becomes
+-- settled when Europe's payment lands; a household split stays unsettled and
+-- open indefinitely.
+--
+-- Nothing writes `settled` today. The comment in queries.ts claims
+-- /api/splits/settle does — that route does not exist, and the column is false
+-- on all 1,279 rows, which is why every trip has always reported 100%
+-- unsettled including trips paid in full.
+
+COMMENT ON COLUMN transaction_splits.settled IS
+  'Obligation discharged. Excluded from owed figures; still counted in spend analytics.';

prisma/migrations/0023_supersede_duplicates/migration.sql

@@ -0,0 +1,37 @@
+-- A transaction imported twice cannot simply be deleted.
+--
+-- Every child of `transactions` is ON DELETE CASCADE — splits, tags, overrides,
+-- expense_metadata, order_reviews. Deleting a row said to be "the duplicate"
+-- therefore destroys whatever curation happens to sit on it, silently and
+-- unrecoverably. The curation is not reliably on the surviving side either: of
+-- the 31 known duplicate pairs, one carries splits and six carry overrides, and
+-- which member holds them is an accident of import order.
+--
+-- So a duplicate is superseded, never removed. The row stays, keeps its
+-- children, and points at the row that replaces it. Reversing a mistake is then
+-- one UPDATE rather than a restore from backup.
+--
+-- This is the statement-vs-statement case. `reconciled_with_id` already covers
+-- manual-vs-statement, and deliberately cannot be reused: the predicate that
+-- hides a reconciled row is scoped to `statement_id IS NULL`, because a
+-- statement line pointing at something else is the survivor, not the duplicate.
+-- Both of these rows are statement lines.
+
+ALTER TABLE transactions
+  ADD COLUMN IF NOT EXISTS superseded_by_id integer
+    REFERENCES transactions(id) ON DELETE SET NULL;
+
+COMMENT ON COLUMN transactions.superseded_by_id IS
+  'This row was imported twice; the named row is the one that counts. Excluded from every figure, kept for its children and its audit trail. NULL = live.';
+
+CREATE INDEX IF NOT EXISTS idx_transactions_superseded
+    ON transactions (superseded_by_id)
+ WHERE superseded_by_id IS NOT NULL;
+
+-- A row cannot supersede itself, and a survivor cannot itself be superseded
+-- (that would hide both members of the pair and lose the amount entirely).
+ALTER TABLE transactions
+  DROP CONSTRAINT IF EXISTS transactions_no_self_supersede;
+ALTER TABLE transactions
+  ADD CONSTRAINT transactions_no_self_supersede
+  CHECK (superseded_by_id IS NULL OR superseded_by_id <> id);

prisma/schema.prisma

@@ -18,6 +18,7 @@ model trips {
   archived    Boolean                 @default(false)
   created_at  DateTime                @default(now())
   overrides   transaction_overrides[]
+  payments    split_payments[]
 }
 
 model transaction_overrides {
@@ -75,9 +76,13 @@ model split_payments {
   payment_date          DateTime      @db.Date
   notes                 String?
   linked_transaction_id Int?
+  trip_id               Int?
   created_at            DateTime      @default(now())
   from_participant      participants  @relation("payments_from", fields: [from_participant_id], references: [id])
   to_participant        participants  @relation("payments_to", fields: [to_participant_id], references: [id])
+  trip                  trips?        @relation(fields: [trip_id], references: [id], onDelete: SetNull)
+
+  @@index([trip_id])
 }
 
 model tags {
@@ -178,11 +183,14 @@ model transactions {
   payment_method          String?       // card | cash | bank_transfer | other; NULL = unknown (migration 0016)
   owner_id                Int?
   reconciled_with_id      Int?
+  superseded_by_id        Int?
   principal_amount        Decimal?      @db.Decimal(12, 2)
   interest_amount         Decimal?      @db.Decimal(12, 2)
   statement               statements?   @relation(fields: [statement_id], references: [id], onDelete: Cascade)
   reconciled_with         transactions?  @relation("reconciled", fields: [reconciled_with_id], references: [id], onDelete: SetNull)
   reconciled_by           transactions[] @relation("reconciled")
+  superseded_by           transactions?  @relation("superseded", fields: [superseded_by_id], references: [id], onDelete: SetNull)
+  supersedes              transactions[] @relation("superseded")
   expense_metadata        expense_metadata?
   order_review            order_reviews?
 }

scripts/split_csv_match.py

@@ -0,0 +1,376 @@
+#!/usr/bin/env python3
+"""Match SplitMyExpenses CSV rows to transactions already in the ledger.
+
+Dry-run by default. It prints what it would do and writes nothing; `--write`
+is a separate step (task #9) and is deliberately not implemented here.
+
+Why this exists
+---------------
+The CSVs are the record of how expenses were actually shared before this app
+existed. Importing them is what makes historical *spend* correct: without a
+split row, a $200 grocery shop counts as $200 of my spending when half of it
+was never mine. The balances are already settled by carryover transaction 2348,
+so these splits are imported as `settled = true` and move no balance.
+
+Three things make the matching harder than "same date, same amount":
+
+1.  **Dates are ambiguous across files.** The household file writes D/M/YYYY;
+    the four trip files write ISO. 474 rows parse validly under both readings,
+    so the format is decided per file, from the file, and never guessed per row.
+
+2.  **The sign convention is not "who paid".** A person's column is their net
+    balance impact: positive means they are owed. So the payer is whoever is
+    positive, and the other person's share is |their negative| / cost. A row
+    reading +cost / -cost therefore means the other party owes 100% -- NOT that
+    the expense was unshared, which is the reading that would fake an
+    arrangement change.
+
+3.  **A settlement is not an expense.** Rows where one person hands the other
+    money must not become split transactions; they are already represented by
+    the carryover.
+
+Usage:
+    .venv/bin/python scripts/split_csv_match.py [--verbose] [--file NAME]
+"""
+
+from __future__ import annotations
+
+import argparse
+import csv
+import glob
+import os
+import re
+import sys
+from collections import Counter
+from dataclasses import dataclass, field
+from datetime import date, datetime, timedelta
+
+import psycopg2
+import psycopg2.extras
+
+DUMP_DIR = os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), "dump")
+
+# Read from the same place the app does rather than hardcoding a container IP,
+# which changes on every recreate.
+def db_url() -> str:
+    url = os.environ.get("DATABASE_URL")
+    if url:
+        return url
+    for envfile in (".env", ".env.test"):
+        path = os.path.join(os.path.dirname(DUMP_DIR), envfile)
+        if not os.path.exists(path):
+            continue
+        for line in open(path):
+            if line.startswith("DATABASE_URL"):
+                return line.split("=", 1)[1].strip().strip('"').strip("'")
+    sys.exit("No DATABASE_URL found (env, .env, or .env.test)")
+
+
+# The two people in these files. The CSV writes full names; the ledger uses
+# first names.
+CSV_ME = "Siddharth Bose"
+CSV_THEM = "Meghalee"
+PARTICIPANT_ME = 1
+PARTICIPANT_THEM = 4
+
+# A settlement transfers money; it is not a shared cost. These are the
+# descriptions SplitMyExpenses uses for them.
+SETTLEMENT_PAT = re.compile(
+    r"payment|settle|debts? remainder|reimburse|transfer to|paid back", re.I
+)
+
+
+@dataclass
+class CsvRow:
+    source: str
+    line: int
+    when: date
+    description: str
+    category: str
+    cost: float
+    currency: str
+    net_me: float
+    net_them: float
+
+    # Filled in by classify()
+    mode: str = ""
+    payer: int = 0            # participant id who paid
+    ower: int = 0             # participant id who owes
+    ower_share: float = 0.0   # 0-100
+
+    def __str__(self) -> str:
+        return f"{self.source}:{self.line} {self.when} {self.description[:38]!r} ${self.cost:.2f} [{self.mode}]"
+
+
+@dataclass
+class MatchReport:
+    rows: list = field(default_factory=list)
+    matched: list = field(default_factory=list)
+    ambiguous: list = field(default_factory=list)
+    unmatched: list = field(default_factory=list)
+    skipped: list = field(default_factory=list)
+
+
+def sniff_date_format(sample: list[str]) -> str:
+    """Decide ISO vs D/M/YYYY for a whole file.
+
+    Deciding per row is what produces a ledger where January and February are
+    silently swapped for some rows and not others. A file is written by one
+    exporter in one format, so the file is the unit of decision.
+    """
+    if not sample:
+        return "%Y-%m-%d"
+    slashes = sum(1 for s in sample if "/" in s)
+    return "%d/%m/%Y" if slashes > len(sample) / 2 else "%Y-%m-%d"
+
+
+def parse_rows(path: str) -> list[CsvRow]:
+    with open(path, newline="", encoding="utf-8-sig") as fh:
+        reader = list(csv.DictReader(fh))
+    if not reader:
+        return []
+
+    # Header names vary in quoting between exports.
+    def col(row: dict, *names: str):
+        for n in names:
+            for k in row:
+                if k.strip().strip('"') == n:
+                    return row[k]
+        return None
+
+    fmt = sniff_date_format([col(r, "Date") or "" for r in reader[:40]])
+    out: list[CsvRow] = []
+    for i, r in enumerate(reader, start=2):
+        raw_date = (col(r, "Date") or "").strip()
+        try:
+            when = datetime.strptime(raw_date, fmt).date()
+        except ValueError:
+            continue
+        try:
+            cost = float(col(r, "Cost") or 0)
+            net_me = float(col(r, CSV_ME) or 0)
+            net_them = float(col(r, CSV_THEM) or 0)
+        except ValueError:
+            continue
+        out.append(
+            CsvRow(
+                source=os.path.basename(path),
+                line=i,
+                when=when,
+                description=(col(r, "Description") or "").strip(),
+                category=(col(r, "Category") or "").strip(),
+                cost=cost,
+                currency=(col(r, "Currency") or "AUD").strip(),
+                net_me=net_me,
+                net_them=net_them,
+            )
+        )
+    return out
+
+
+def classify(row: CsvRow) -> CsvRow:
+    """Work out who paid and what share the other person owes.
+
+    A person's column is their net balance impact, not their share: positive
+    means they are owed money. So the payer is whoever is positive.
+    """
+    if row.cost == 0:
+        row.mode = "zero-cost"
+        return row
+
+    if SETTLEMENT_PAT.search(row.description):
+        row.mode = "settlement"
+        return row
+
+    # Both zero against a real cost: recorded but not shared.
+    if abs(row.net_me) < 0.005 and abs(row.net_them) < 0.005:
+        row.mode = "unshared"
+        return row
+
+    if row.net_me > 0:
+        row.payer, row.ower, owed = PARTICIPANT_ME, PARTICIPANT_THEM, abs(row.net_them)
+    else:
+        row.payer, row.ower, owed = PARTICIPANT_THEM, PARTICIPANT_ME, abs(row.net_me)
+
+    row.ower_share = round(owed / row.cost * 100, 2)
+    if abs(row.ower_share - 50) < 0.6:
+        row.mode = "50/50"
+    elif abs(row.ower_share - 100) < 0.6:
+        row.mode = "other-owes-all"
+    else:
+        row.mode = f"uneven-{row.ower_share:.0f}"
+    return row
+
+
+def norm(s: str) -> set[str]:
+    return {w for w in re.split(r"[^a-z0-9]+", (s or "").lower()) if len(w) > 2}
+
+
+def score(row: CsvRow, tx: dict) -> float:
+    """How well a ledger row matches a CSV row. Amount and date gate it;
+    description only ranks among survivors."""
+    days = abs((tx["transaction_date"] - row.when).days)
+    s = 100.0 - days * 4
+    overlap = norm(row.description) & (norm(tx["description"]) | norm(tx["merchant_normalized"]))
+    s += 12 * len(overlap)
+    return s
+
+
+def main() -> int:
+    ap = argparse.ArgumentParser()
+    ap.add_argument("--verbose", action="store_true")
+    ap.add_argument("--file", help="only this CSV (substring match)")
+    ap.add_argument("--window", type=int, default=5, help="date tolerance in days")
+    ap.add_argument(
+        "--write", action="store_true",
+        help="actually insert the splits (settled=true). Without this, nothing is written.",
+    )
+    args = ap.parse_args()
+
+    paths = sorted(glob.glob(os.path.join(DUMP_DIR, "*SplitMyExpenses*.csv")))
+    if args.file:
+        paths = [p for p in paths if args.file in os.path.basename(p)]
+    if not paths:
+        sys.exit("No SplitMyExpenses CSVs found in dump/")
+
+    conn = psycopg2.connect(db_url())
+    cur = conn.cursor(cursor_factory=psycopg2.extras.RealDictCursor)
+    # Superseded rows are duplicates; matching against them would attach a split
+    # to a row nothing else counts.
+    cur.execute(
+        """
+        SELECT t.id, t.transaction_date, COALESCE(t.description,'') AS description,
+               COALESCE(t.merchant_normalized,'') AS merchant_normalized,
+               COALESCE(t.amount_aud, t.amount)::float AS amount,
+               t.superseded_by_id,
+               EXISTS(SELECT 1 FROM transaction_splits x WHERE x.transaction_id=t.id) AS has_split
+        FROM transactions t
+        LEFT JOIN statements s ON s.id = t.statement_id
+        WHERE COALESCE(t.owner_id, s.owner_id) IN (%s, %s)
+          AND t.superseded_by_id IS NULL
+          AND NOT (t.statement_id IS NULL AND t.reconciled_with_id IS NOT NULL)
+          AND t.transaction_type IN ('debit','fee','interest')
+        """,
+        (PARTICIPANT_ME, PARTICIPANT_THEM),
+    )
+    txs = cur.fetchall()
+
+    # Index by rounded amount: the amount must agree, so it is the only cheap
+    # gate that never needs fuzzy comparison.
+    by_amount: dict[float, list[dict]] = {}
+    for t in txs:
+        by_amount.setdefault(round(t["amount"], 2), []).append(t)
+
+    rep = MatchReport()
+    modes: Counter = Counter()
+    candidates: list = []
+    for path in paths:
+        for row in parse_rows(path):
+            classify(row)
+            rep.rows.append(row)
+            modes[row.mode] += 1
+
+            if row.mode in ("settlement", "zero-cost", "unshared"):
+                rep.skipped.append(row)
+                continue
+            if row.currency != "AUD":
+                rep.skipped.append(row)
+                continue
+
+            cands = [
+                t for t in by_amount.get(round(row.cost, 2), [])
+                if abs((t["transaction_date"] - row.when).days) <= args.window
+            ]
+            if not cands:
+                rep.unmatched.append(row)
+            else:
+                candidates.append((row, cands))
+
+    # Assign one-to-one, best pair first.
+    #
+    # Without this a ledger row can be claimed by several CSV rows. That is not
+    # hypothetical: the NZ trip has two identical $10.16 Uber trips on one day
+    # and three matching ledger rows, and four PayMyPark rows in the same shape.
+    # Attaching a split twice is harmless (the unique key absorbs it) but it
+    # leaves the second CSV row silently unrepresented while looking matched,
+    # which is a lie in the report rather than a defect in the data.
+    scored = sorted(
+        ((score(row, t), row, t) for row, cands in candidates for t in cands),
+        key=lambda x: -x[0],
+    )
+    taken_tx: set[int] = set()
+    taken_row: set[int] = set()
+    for s, row, t in scored:
+        if id(row) in taken_row or t["id"] in taken_tx:
+            continue
+        taken_row.add(id(row))
+        taken_tx.add(t["id"])
+        rep.matched.append((row, t))
+    for row, cands in candidates:
+        if id(row) not in taken_row:
+            rep.ambiguous.append((row, cands))
+
+    total = len(rep.rows)
+    considered = total - len(rep.skipped)
+    print(f"CSV rows              {total}")
+    print(f"  skipped             {len(rep.skipped)}  (settlements, zero-cost, unshared, non-AUD)")
+    print(f"  considered          {considered}")
+    print(f"    matched           {len(rep.matched)}  ({len(rep.matched)/max(considered,1)*100:.1f}%)")
+    print(f"    ambiguous         {len(rep.ambiguous)}")
+    print(f"    unmatched         {len(rep.unmatched)}")
+    print()
+    print("Row modes:")
+    for m, n in modes.most_common():
+        print(f"  {m:<18} {n}")
+
+    already = sum(1 for _, t in rep.matched if t["has_split"])
+    print()
+    print(f"Of the matched, {already} already carry a split and would be left alone;")
+    print(f"{len(rep.matched) - already} would gain one.")
+
+    if args.write:
+        # Imported as settled: these obligations were discharged on a platform
+        # we no longer run, and the residual is already carried by transaction
+        # 2348. Writing them unsettled would re-open ~$40k of debts that were
+        # paid years ago. See ACTIVE_OBLIGATION in analytics-sql.ts -- settled
+        # rows stay in spend and leave every owed figure, which is exactly the
+        # point: this import exists to correct historical SPEND.
+        SETTLED_ON = "2026-01-09"  # the carryover's date
+        written = 0
+        for row, tx in rep.matched:
+            if tx["has_split"]:
+                continue
+            payer_share = round(100 - row.ower_share, 2)
+            pairs = [(row.ower, row.ower_share)]
+            if payer_share > 0:
+                pairs.append((row.payer, payer_share))
+            for pid, share in pairs:
+                cur.execute(
+                    """
+                    INSERT INTO transaction_splits
+                      (transaction_id, participant_id, share_percent, settled, settled_at)
+                    VALUES (%s, %s, %s, true, %s)
+                    ON CONFLICT (transaction_id, participant_id) DO NOTHING
+                    """,
+                    (tx["id"], pid, share, SETTLED_ON),
+                )
+                written += cur.rowcount
+        conn.commit()
+        print(f"\nWROTE {written} split rows (settled=true, settled_at={SETTLED_ON}).")
+
+    if args.verbose:
+        print("\n--- ambiguous ---")
+        for row, cands in rep.ambiguous[:40]:
+            print(f"  {row}")
+            for t in cands[:3]:
+                print(f"      -> #{t['id']} {t['transaction_date']} {t['description'][:44]!r}")
+        print("\n--- unmatched ---")
+        for row in rep.unmatched[:60]:
+            print(f"  {row}")
+
+    conn.close()
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

src/__tests__/fixtures/orders/real/ut-meta.json

@@ -0,0 +1,20 @@
+{
+  "ut-00": {
+    "messageId": "AQMkADAwATM0MDAAMS1hYzQyLTVjZDktMDACLTAwCgBGAAADS6kDcxBxKkKwbCK1wU1QlwcA7Ex_hlaTTU0AgHQ16optl_AAAAIBDAAAAOxMfoZWk01NAIB0NeqKbZfgAAkH8zVAAAAA",
+    "subject": "Your Sunday evening trip with Uber",
+    "receivedAt": "2026-06-21T10:11:16Z",
+    "sender": "noreply@uber.com"
+  },
+  "ut-01": {
+    "messageId": "AQMkADAwATM0MDAAMS1hYzQyLTVjZDktMDACLTAwCgBGAAADS6kDcxBxKkKwbCK1wU1QlwcA7Ex_hlaTTU0AgHQ16optl_AAAAIBDAAAAOxMfoZWk01NAIB0NeqKbZfgAAkH8zU-AAAA",
+    "subject": "Your Sunday morning trip with Uber",
+    "receivedAt": "2026-06-21T09:02:09Z",
+    "sender": "noreply@uber.com"
+  },
+  "ut-summary": {
+    "messageId": "AQMkADAwATM0MDAAMS1hYzQyLTVjZDktMDACLTAwCgBGAAADS6kDcxBxKkKwbCK1wU1QlwcA7Ex_hlaTTU0AgHQ16optl_AAAAIBDAAAAOxMfoZWk01NAIB0NeqKbZfgAAkH8zU3AAAA",
+    "subject": "Your Sunday morning trip with Uber",
+    "receivedAt": "2026-06-20T22:27:54Z",
+    "sender": "noreply@uber.com"
+  }
+}

src/__tests__/integration/analytics-sql.test.ts

@@ -0,0 +1,126 @@
+import { describe, it, expect } from "vitest";
+import { queryRaw, queryRow } from "../../lib/db";
+import {
+  EXCLUDE_RECONCILED_SOURCE,
+  NATIVE_CURRENCY,
+  AMOUNT_UNCONVERTED,
+} from "../../lib/analytics-sql";
+
+/**
+ * These fragments are the ones that drifted.
+ *
+ * The reconciled-row exclusion lived only in `queries.ts` for months while every
+ * analytics route counted the superseded manual rows as spend — 48 rows, $4,474
+ * of double count, invisible because the transaction list (which did exclude
+ * them) looked right. The currency expression has the same shape of risk: it is
+ * read by two call sites with two different denomination conventions.
+ *
+ * Assertions are per-row against known fixtures rather than against aggregate
+ * totals, so a change in unrelated data cannot mask a regression.
+ */
+
+async function scratchTxn(cols: string, vals: string, params: unknown[] = []) {
+  const row = await queryRow<{ id: number }>(
+    `INSERT INTO transactions (${cols}) VALUES (${vals}) RETURNING id`,
+    params
+  );
+  return row!.id;
+}
+
+/** Does this row survive the predicate? */
+async function passes(predicate: string, id: number): Promise<boolean> {
+  const rows = await queryRaw(
+    `SELECT t.id FROM transactions t
+       LEFT JOIN statements s ON s.id = t.statement_id
+      WHERE t.id = $1 AND (${predicate})`,
+    [id]
+  );
+  return rows.length === 1;
+}
+
+describe("EXCLUDE_RECONCILED_SOURCE", () => {
+  it("drops a manual row that a statement line has superseded", async () => {
+    const survivor = await scratchTxn(
+      "transaction_date, description, amount, transaction_type",
+      "'2026-03-01','Analytics fixture — survivor', 10.00, 'debit'"
+    );
+    const superseded = await scratchTxn(
+      "transaction_date, description, amount, transaction_type, reconciled_with_id",
+      "'2026-03-01','Analytics fixture — superseded', 10.00, 'debit', $1",
+      [survivor]
+    );
+
+    expect(await passes(EXCLUDE_RECONCILED_SOURCE, superseded)).toBe(false);
+  });
+
+  it("keeps an ordinary manual row that was never reconciled", async () => {
+    const id = await scratchTxn(
+      "transaction_date, description, amount, transaction_type",
+      "'2026-03-01','Analytics fixture — unreconciled', 10.00, 'debit'"
+    );
+    expect(await passes(EXCLUDE_RECONCILED_SOURCE, id)).toBe(true);
+  });
+
+  it("keeps a credits order row — nothing ever sets reconciled_with_id on one", async () => {
+    // The order slice records the card match in expense_metadata, not on the
+    // transaction, and needsCardMatch() holds these out of the reconcile queue.
+    // If that ever changes, this exclusion would start eating real spend.
+    const id = await scratchTxn(
+      "transaction_date, description, amount, transaction_type, payment_method",
+      "'2026-03-01','Order - Analytics fixture', 25.00, 'debit', 'credits'"
+    );
+    expect(await passes(EXCLUDE_RECONCILED_SOURCE, id)).toBe(true);
+  });
+});
+
+describe("NATIVE_CURRENCY", () => {
+  async function currencyOf(id: number) {
+    const row = await queryRow<{ ccy: string; unconverted: boolean }>(
+      `SELECT ${NATIVE_CURRENCY} AS ccy, ${AMOUNT_UNCONVERTED} AS unconverted
+         FROM transactions t LEFT JOIN statements s ON s.id = t.statement_id
+        WHERE t.id = $1`,
+      [id]
+    );
+    return row!;
+  }
+
+  it("reads a statement-less order row's own currency, not 'AUD'", async () => {
+    // The bug this guards: sourcing currency from s.currency alone labelled
+    // every foreign order row AUD, because an order has no statement.
+    const id = await scratchTxn(
+      "transaction_date, description, amount, transaction_type, payment_method, foreign_currency_amount, foreign_currency_code",
+      "'2026-03-01','Order - Foreign fixture', 3500.00, 'debit', 'credits', 3500.00, 'LKR'"
+    );
+    const { ccy, unconverted } = await currencyOf(id);
+    expect(ccy).toBe("LKR");
+    // No amount_aud: the ingest path refuses to assert an FX rate it lacks, so
+    // this row's AUD value is genuinely unknown and must be reported as such.
+    expect(unconverted).toBe(true);
+  });
+
+  it("prefers the statement's currency over the foreign-charge record", async () => {
+    // Opposite convention: on an AUD statement, `amount` is AUD and
+    // foreign_currency_code merely notes what was originally charged. Reading
+    // the foreign code here would mislabel an AUD row as USD.
+    const stmt = await queryRow<{ id: number }>(
+      `INSERT INTO statements (bank_name, account_number, billing_end_date, currency, filename)
+       VALUES ('Analytics Fixture Bank','0000','2026-03-31','AUD','analytics-fixture.pdf') RETURNING id`
+    );
+    const id = await scratchTxn(
+      "transaction_date, description, amount, amount_aud, transaction_type, statement_id, foreign_currency_amount, foreign_currency_code",
+      "'2026-03-01','Overseas purchase fixture', 45.00, 45.00, 'debit', $1, 30.00, 'USD'",
+      [stmt!.id]
+    );
+    const { ccy, unconverted } = await currencyOf(id);
+    expect(ccy).toBe("AUD");
+    expect(unconverted).toBe(false);
+  });
+
+  it("defaults a plain manual row to AUD", async () => {
+    const id = await scratchTxn(
+      "transaction_date, description, amount, transaction_type",
+      "'2026-03-01','Plain manual fixture', 12.00, 'debit'"
+    );
+    expect((await currencyOf(id)).ccy).toBe("AUD");
+  });
+});

src/__tests__/integration/helpers.ts

@@ -35,6 +35,7 @@ export async function resetDB(pool: Pool) {
       transactions,
       statements,
       tags,
+      trips,
       participants
     RESTART IDENTITY CASCADE
   `);
@@ -72,7 +73,10 @@ export async function insertTransaction(
      VALUES ($1, NULL, $2, $3, $4, $5, $6, 0) RETURNING id`,
     [
       ownerId,
-      overrides.transaction_date ?? "2024-06-15",
+      // Post-cutover by default: a split on an older transaction never counts
+      // towards a balance (ACTIVE_OBLIGATION), so a pre-cutover default would
+      // make every balance fixture silently read zero.
+      overrides.transaction_date ?? "2026-06-15",
       overrides.description ?? "Test transaction",
       overrides.amount ?? 100,
       overrides.transaction_type ?? "debit",

src/__tests__/integration/order-ingestion.test.ts

@@ -9,11 +9,11 @@ import {
   reconcilePendingOrders,
   parseOrderAmendment,
   applyOrderAmendment,
-  OrderParseError,
   NotAReceiptError,
   type MessageMeta,
 } from "../../lib/order-ingestion";
 import { EXCLUDE_NON_SPEND } from "../../lib/analytics-sql";
+import { bankLabel, needsCardMatch } from "../../lib/queries";
 
 /**
  * These run against REAL captured receipts, not synthetic fixtures. The earlier
@@ -142,6 +142,19 @@ describe("Order ingestion — invariants", () => {
   beforeEach(async () => {
     await queryRaw(`DELETE FROM expense_metadata WHERE source = 'email'`);
     await queryRaw(`DELETE FROM transactions WHERE description LIKE 'Order - %'`);
+    // The statement fixtures these tests insert survived into the next run, and
+    // reconcileCardLeg matched a leftover charge at ingest time — so an order
+    // meant to park "awaiting_card_statement" resolved immediately instead.
+    // That is the whole story behind the intermittent failure in "parks an
+    // unresolvable split": not a race, just fixtures that were never cleaned.
+    // Must happen BEFORE ingest, which is why cleaning up at the end of the
+    // test was not enough.
+    await queryRaw(
+      `DELETE FROM statements WHERE filename IN ('test-westpac-2026-03.pdf', 'panel-cba.pdf', 'panel-plain.pdf')`
+    );
+    await queryRaw(
+      `DELETE FROM transactions WHERE description IN ('DD *DOORDASH WOOLWORTHS MELBOURNE AUS', 'UBER *EATS ZURICH')`
+    );
   });
 
   it("I6: a credits order creates one transaction at face value", async () => {
@@ -164,7 +177,7 @@ describe("Order ingestion — invariants", () => {
     expect(b.skipped).toBe("already_ingested");
     expect(b.metadataId).toBe(a.metadataId);
     const n = await queryRow<{ c: string }>(
-      `SELECT count(*)::text c FROM transactions WHERE description = 'Order - Mad Mex'`
+      `SELECT count(*)::text c FROM transactions WHERE description = 'Order - Mad Mex (DoorDash)'`
     );
     expect(Number(n!.c)).toBe(1);
   });
@@ -181,14 +194,22 @@ describe("Order ingestion — invariants", () => {
     ).rejects.toThrow();
   });
 
-  it("I11: a [Family] order is imported, tagged, and excluded from spend", async () => {
+  it("I11: a [Family] order records provenance and creates no transaction", async () => {
+    // Requirement was "import them but tag so they're excluded from budgets".
+    // Correct mechanism: the CARD statement line is the transaction and carries
+    // the family tag. Creating a second, credits-flavoured row duplicated it.
     const p = parseOrderHTML(
       html("ue-04"),
       meta({ subject: "[Family] Your Tuesday evening order with Uber Eats", sender: "uber.com", receivedAt: "2026-07-07T10:08:00Z" })
     );
     const res = await processOrderIngestion(p);
-    expect(res.transactionId).not.toBeNull();
+    expect(res.transactionId).toBeNull();
-    expect(res.flags).toContain("family_payment_assumed_credits");
+    expect(res.flags).toContain("family_card_settled_no_transaction");
+
+    const meta_ = await queryRow<{ currency: string }>(
+      `SELECT currency FROM expense_metadata WHERE id = $1`, [res.metadataId]
+    );
+    expect(meta_!.currency).toBe("LKR");
   });
 
   it("a foreign-currency order records the original amount and code", async () => {
@@ -227,8 +248,11 @@ describe("Order ingestion — invariants", () => {
     );
 
     const out = await reconcilePendingOrders();
-    expect(out.resolved).toBeGreaterThanOrEqual(1);
+    // Deliberately not asserting global counts: reconcilePendingOrders() scans
-    expect(out.created).toBeGreaterThanOrEqual(1);
+    // every pending row in the database, so another test's leftovers change the
+    // totals. Assert on THIS order's outcome instead — that is what the test is
+    // actually about, and it does not depend on what else is in the table.
+    expect(out.examined).toBeGreaterThanOrEqual(1);
 
     const credits = await queryRow<{ amount: string }>(
       `SELECT t.amount::text FROM transactions t
@@ -240,9 +264,12 @@ describe("Order ingestion — invariants", () => {
   });
 
   it("reconciliation is idempotent — a second pass creates nothing", async () => {
-    const before = await queryRow<{ c: string }>(`SELECT count(*)::text c FROM transactions`);
+    // A first pass has already run above; a second must add nothing. Scoped to
+    // 'Order - %' rows so unrelated fixtures cannot move the number.
+    const q = `SELECT count(*)::text c FROM transactions WHERE description LIKE 'Order - %'`;
+    const before = await queryRow<{ c: string }>(q);
     const out = await reconcilePendingOrders();
-    const after = await queryRow<{ c: string }>(`SELECT count(*)::text c FROM transactions`);
+    const after = await queryRow<{ c: string }>(q);
     expect(out.created).toBe(0);
     expect(after!.c).toBe(before!.c);
   });
@@ -320,39 +347,162 @@ describe("Refund amendments", () => {
   });
 });
 
-describe("[Family] orders import rather than park", () => {
+describe("[Family] orders are card-settled, not credits", () => {
-  it("records a family order as credits and tags it", async () => {
+  it("creates provenance but NO transaction — the statement line is the transaction", async () => {
+    // Regression: these were assumed credits-funded because the receipt names
+    // the payer and no instrument. The card statement carries all four (CBA
+    // ...3893, exact LKR matches), so creating a transaction double-counted
+    // spend already recorded.
     const p = parseOrderHTML(
       html("ue-04"),
       meta({ subject: "[Family] Your Tuesday evening order with Uber Eats", sender: "uber.com", receivedAt: "2026-07-07T10:08:00Z" })
     );
-    expect(p.flags).toContain("family_payment_assumed_credits");
+    expect(p.flags).toContain("family_card_settled_no_transaction");
+    expect(p.payment.credits_amount).toBeNull();
 
     const res = await processOrderIngestion(p);
-    expect(res.transactionId).not.toBeNull();
+    expect(res.transactionId).toBeNull();
+    expect(res.metadataId).not.toBeNull();
+  });
+});
 
-    const tag = await queryRow<{ name: string }>(
+describe("owner scoping", () => {
-      `SELECT tg.name FROM transaction_tags tt JOIN tags tg ON tg.id = tt.tag_id
+  it("sets owner_id so the row is visible to the app", async () => {
-        WHERE tt.transaction_id = $1`,
+    // Regression: analytics scope on COALESCE(t.owner_id, s.owner_id). An
+    // ingested order has no statement, so a NULL owner_id made all 85
+    // backfilled rows invisible in every view while sitting in the table.
+    const p = parseOrderHTML(html("dd-01"), meta({ messageId: `owner-${Date.now()}` }));
+    const res = await processOrderIngestion(p);
+    const row = await queryRow<{ owner_id: number | null }>(
+      `SELECT owner_id FROM transactions WHERE id = $1`,
       [res.transactionId]
     );
-    expect(tag!.name).toBe("family");
+    expect(row!.owner_id).not.toBeNull();
 
-    // LKR is preserved, and amount_aud stays NULL — no FX rate is available.
+    const visible = await queryRaw(
-    const txn = await queryRow<{ foreign_currency_code: string; amount_aud: string | null }>(
+      `SELECT t.id FROM transactions t LEFT JOIN statements s ON s.id = t.statement_id
-      `SELECT foreign_currency_code, amount_aud::text FROM transactions WHERE id = $1`,
+        WHERE t.id = $1 AND COALESCE(t.owner_id, s.owner_id) = $2`,
+      [res.transactionId, row!.owner_id]
+    );
+    expect(visible).toHaveLength(1);
+  });
+});
+
+describe("how an ingested order presents in the app", () => {
+  it("names the platform in the description", async () => {
+    // "Order - Burger Corner" gives no way to know where to look for the
+    // detail, and the same restaurant can be on both platforms.
+    const p = parseOrderHTML(html("dd-01"), meta({ messageId: `desc-${Date.now()}` }));
+    const res = await processOrderIngestion(p);
+    const row = await queryRow<{ description: string }>(
+      `SELECT description FROM transactions WHERE id = $1`, [res.transactionId]
+    );
+    expect(row!.description).toMatch(/\(DoorDash\)$/);
+  });
+
+  it("reads as 'Gift Card', not 'Manual', and stays out of the reconcile queue", async () => {
+    // bank_name is derived — no statement means "Manual", which reads as
+    // "hand-entered, awaiting a card line". A credits order has no card line
+    // coming, ever; 81 of them sat in the queue waiting for one.
+    const p = parseOrderHTML(html("dd-01"), meta({ messageId: `bank-${Date.now()}` }));
+    const res = await processOrderIngestion(p);
+
+    const row = await queryRow<{ bank_name: string }>(
+      `SELECT ${bankLabel()} as bank_name
+         FROM transactions t LEFT JOIN statements s ON s.id = t.statement_id
+        WHERE t.id = $1`,
       [res.transactionId]
     );
-    expect(txn!.foreign_currency_code).toBe("LKR");
+    expect(row!.bank_name).toBe("Gift Card");
-    expect(txn!.amount_aud).toBeNull();
 
-    // And it must not reach spend.
+    const queued = await queryRaw(
-    const visible = await queryRaw(
       `SELECT t.id FROM transactions t
-        LEFT JOIN transaction_overrides o ON o.transaction_id = t.id
+        WHERE t.id = $1 AND t.statement_id IS NULL AND ${needsCardMatch("t")}`,
-       WHERE t.id = $1 AND (${EXCLUDE_NON_SPEND})`,
       [res.transactionId]
+    );
+    expect(queued).toHaveLength(0);
+  });
+
+  it("records the platform and the message it came from", async () => {
+    const p = parseOrderHTML(
+      html("ue-00"),
+      meta({ messageId: `prov-${Date.now()}`, subject: "Your Wednesday order with Uber Eats", sender: "uber.com" })
+    );
+    const res = await processOrderIngestion(p, {
+      messageId: `prov-${Date.now()}`,
+      subject: "Your Wednesday order with Uber Eats",
+      sender: "uber.com",
+    });
+    const row = await queryRow<{
+      platform: string; source_email_from: string; route: { label: string }[];
+    }>(
+      `SELECT platform, source_email_from, route FROM expense_metadata WHERE id = $1`,
+      [res.metadataId]
+    );
+    expect(row!.platform).toBe("ubereats");
+    expect(row!.source_email_from).toBe("uber.com");
+    expect(row!.route.map((r) => r.label)).toEqual(["Pick-up", "Delivery"]);
+  });
+});
+
+describe("receipt lookup for the transaction detail panel", () => {
+  // Mirrors /api/transactions/[id]/order — the panel resolves a receipt from
+  // either side, and a card-settled order only has the matched_transaction_id
+  // side, which is exactly where the detail would otherwise go missing.
+  const receiptFor = (txnId: number) =>
+    queryRow<{ platform: string; route: { label: string; address: string }[] }>(
+      `SELECT platform, route FROM expense_metadata
+        WHERE transaction_id = $1 OR matched_transaction_id = $1 LIMIT 1`,
+      [txnId]
+    );
+
+  it("finds the receipt for a credits order", async () => {
+    // ue-00 is Uber Cash — credits, so it creates a transaction. ue-09 names a
+    // payer with no instrument and correctly parks awaiting a card statement,
+    // which would leave nothing to look the receipt up by.
+    const p = parseOrderHTML(
+      html("ue-00"),
+      meta({ messageId: `panel-${Date.now()}`, subject: "Your Wednesday order with Uber Eats", sender: "uber.com" })
+    );
+    const res = await processOrderIngestion(p);
+    const r = await receiptFor(res.transactionId!);
+    expect(r!.platform).toBe("ubereats");
+    expect(r!.route.map((x) => x.label)).toEqual(["Pick-up", "Delivery"]);
+  });
+
+  it("finds it from the statement line for a card-settled order", async () => {
+    const st = await queryRow<{ id: number }>(
+      `INSERT INTO statements (bank_name, account_number, filename)
+       VALUES ('CBA', '5523504401723893', 'panel-cba.pdf') RETURNING id`
+    );
+    const card = await queryRow<{ id: number }>(
+      `INSERT INTO transactions (statement_id, transaction_date, description, amount, transaction_type)
+       VALUES ($1, '2026-04-07', 'UBER *EATS ZURICH', 51.23, 'debit') RETURNING id`,
+      [st!.id]
+    );
+    const m = await queryRow<{ id: number }>(
+      `INSERT INTO expense_metadata (source, order_reference, platform, route, matched_transaction_id)
+       VALUES ('email', $1, 'ubereats', '[{"label":"Pick-up","time":null,"address":"Ebikon"}]'::jsonb, $2)
+       RETURNING id`,
+      [`panel-card-${Date.now()}`, card!.id]
+    );
+    expect(m).not.toBeNull();
+
+    const r = await receiptFor(card!.id);
+    expect(r!.platform).toBe("ubereats");
+    expect(r!.route[0].address).toBe("Ebikon");
+  });
+
+  it("returns nothing for an ordinary transaction", async () => {
+    const st = await queryRow<{ id: number }>(
+      `INSERT INTO statements (bank_name, account_number, filename)
+       VALUES ('CBA', '1111', 'panel-plain.pdf') RETURNING id`
+    );
+    const t = await queryRow<{ id: number }>(
+      `INSERT INTO transactions (statement_id, transaction_date, description, amount, transaction_type)
+       VALUES ($1, '2026-04-07', 'COLES 1234', 12.00, 'debit') RETURNING id`,
+      [st!.id]
     );
-    expect(visible).toHaveLength(0);
+    expect(await receiptFor(t!.id)).toBeNull();
   });
 });

src/__tests__/integration/queries.test.ts

@@ -8,7 +8,7 @@ mockDbWithPool(pool);
 
 // Dynamic import AFTER the mock ensures getTransactions / getParticipantBalances
 // use the test pool rather than Prisma's singleton.
-const { getTransactions, getParticipantBalances } = await import("@/lib/queries");
+const { getTransactions, getParticipantBalances, getTripAnalytics, getTripById, getStatements } = await import("@/lib/queries");
 
 beforeEach(async () => {
   await resetDB(pool);
@@ -253,3 +253,471 @@ describe("getParticipantBalances", () => {
     expect(bobBalance!.unsettled_count).toBe(2);
   });
 });
+
+describe("getTransactions — order provenance for the description sub-line", () => {
+  it("carries the route and platform of an order-derived row", async () => {
+    // Five rows all reading "Order - Uber Trip" are indistinguishable in the
+    // list; where the trip went is the only thing that separates them, and it
+    // was already stored.
+    const { ownerId } = await seedParticipants(pool);
+    const txId = await insertTransaction(pool, ownerId, { description: "Order - Uber Trip" });
+    await pool.query(
+      `INSERT INTO expense_metadata (source, order_reference, platform, route, transaction_id)
+       VALUES ('email', $1, 'uber',
+         '[{"label":"Pick-up","time":"7:32 pm","address":"Terminal 2, Melbourne Airport (MEL), Tullamarine VIC 3045, Australia"},
+           {"label":"Drop-off","time":"8:10 pm","address":"19 Lady Penrhyn Dr, Wyndham Vale VIC 3024, Australia"}]'::jsonb,
+         $2)`,
+      [`route-${Date.now()}`, txId]
+    );
+
+    const { data } = await getTransactions(ownerId, { limit: 50, offset: 0 });
+    const row = data.find((r) => r.id === txId)!;
+    expect(row.order_platform).toBe("uber");
+    expect(row.order_route).toHaveLength(2);
+    expect(row.order_route![0].address).toContain("Melbourne Airport");
+  });
+
+  it("resolves from the statement line for a card-settled order", async () => {
+    // A card-settled order creates no transaction of its own (I5) — the
+    // receipt points at the statement line through matched_transaction_id.
+    const { ownerId } = await seedParticipants(pool);
+    const txId = await insertTransaction(pool, ownerId, { description: "UBER *TRIP AUCKLAND" });
+    await pool.query(
+      `INSERT INTO expense_metadata (source, order_reference, platform, route, matched_transaction_id)
+       VALUES ('email', $1, 'uber',
+         '[{"label":"Pick-up","time":null,"address":"64 Federal Street, Auckland 1010, NZ"},
+           {"label":"Drop-off","time":null,"address":"International Terminal, Auckland 2022, New Zealand"}]'::jsonb,
+         $2)`,
+      [`route-card-${Date.now()}`, txId]
+    );
+
+    const { data } = await getTransactions(ownerId, { limit: 50, offset: 0 });
+    const row = data.find((r) => r.id === txId)!;
+    expect(row.order_route).toHaveLength(2);
+  });
+
+  it("leaves an ordinary transaction with no route", async () => {
+    const { ownerId } = await seedParticipants(pool);
+    const txId = await insertTransaction(pool, ownerId, { description: "COLES 1234" });
+    const { data } = await getTransactions(ownerId, { limit: 50, offset: 0 });
+    const row = data.find((r) => r.id === txId)!;
+    expect(row.order_route).toBeNull();
+    expect(row.order_platform).toBeNull();
+  });
+});
+
+// ── settlement scope: settled + split_payments.trip_id (migration 0022) ───────
+
+describe("getParticipantBalances — settled", () => {
+  it("excludes a settled split from what is owed", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const txId = await insertTransaction(pool, ownerId, { amount: 100 });
+    await pool.query(
+      `INSERT INTO transaction_splits (transaction_id, participant_id, share_percent, settled)
+       VALUES ($1, $2, 50, true)`,
+      [txId, otherId]
+    );
+
+    const balances = await getParticipantBalances(ownerId);
+    const bob = balances.find((b) => b.id === otherId);
+    expect(Number(bob!.total_owed)).toBeCloseTo(0);
+  });
+
+  it("still counts an unsettled split alongside a settled one", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const settledTx = await insertTransaction(pool, ownerId, { amount: 100 });
+    const liveTx = await insertTransaction(pool, ownerId, { amount: 40 });
+    await pool.query(
+      `INSERT INTO transaction_splits (transaction_id, participant_id, share_percent, settled)
+       VALUES ($1, $2, 50, true), ($3, $2, 50, false)`,
+      [settledTx, otherId, liveTx]
+    );
+
+    const balances = await getParticipantBalances(ownerId);
+    const bob = balances.find((b) => b.id === otherId);
+    // Only the live split counts: 50% of 40.
+    expect(Number(bob!.total_owed)).toBeCloseTo(20);
+  });
+});
+
+describe("getTripAnalytics — per-trip settlement", () => {
+  async function seedTrip(ownerId: number, otherId: number) {
+    const trip = await pool.query(
+      `INSERT INTO trips (owner_id, name, start_date, end_date)
+       VALUES ($1, 'Test Trip', '2026-03-01', '2026-03-10') RETURNING id`,
+      [ownerId]
+    );
+    const tripId = trip.rows[0].id as number;
+    const txId = await insertTransaction(pool, ownerId, { amount: 200, category: "travel" });
+    await pool.query(
+      `INSERT INTO transaction_overrides (transaction_id, trip_id) VALUES ($1, $2)`,
+      [txId, tripId]
+    );
+    await pool.query(
+      `INSERT INTO transaction_splits (transaction_id, participant_id, share_percent)
+       VALUES ($1, $2, 50)`,
+      [txId, otherId]
+    );
+    return tripId;
+  }
+
+  it("reports the gross share before any payment", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const tripId = await seedTrip(ownerId, otherId);
+
+    const { participant_splits } = await getTripAnalytics(tripId, ownerId);
+    const bob = participant_splits.find((r) => r.participant_id === otherId);
+    expect(Number(bob!.owed)).toBeCloseTo(100);
+  });
+
+  it("nets off a payment scoped to that trip", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const tripId = await seedTrip(ownerId, otherId);
+    await pool.query(
+      `INSERT INTO split_payments (from_participant_id, to_participant_id, amount, payment_date, trip_id)
+       VALUES ($1, $2, 60, '2026-03-15', $3)`,
+      [otherId, ownerId, tripId]
+    );
+
+    const { participant_splits } = await getTripAnalytics(tripId, ownerId);
+    const bob = participant_splits.find((r) => r.participant_id === otherId);
+    expect(Number(bob!.owed)).toBeCloseTo(40);
+  });
+
+  // The point of the whole scope column: settling the household tab must not
+  // make a trip look paid. Before trip_id existed there was one global pool and
+  // this distinction could not be expressed.
+  it("ignores a household payment when reporting the trip", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const tripId = await seedTrip(ownerId, otherId);
+    await pool.query(
+      `INSERT INTO split_payments (from_participant_id, to_participant_id, amount, payment_date, trip_id)
+       VALUES ($1, $2, 60, '2026-03-15', NULL)`,
+      [otherId, ownerId]
+    );
+
+    const { participant_splits } = await getTripAnalytics(tripId, ownerId);
+    const bob = participant_splits.find((r) => r.participant_id === otherId);
+    expect(Number(bob!.owed)).toBeCloseTo(100);
+  });
+
+  it("drops a settled split from the trip figure", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const tripId = await seedTrip(ownerId, otherId);
+    await pool.query(`UPDATE transaction_splits SET settled = true WHERE participant_id = $1`, [otherId]);
+
+    const { participant_splits } = await getTripAnalytics(tripId, ownerId);
+    const bob = participant_splits.find((r) => r.participant_id === otherId);
+    expect(bob === undefined || Number(bob.owed) === 0).toBe(true);
+  });
+});
+
+describe("getTripAnalytics — owner scoping", () => {
+  it("ignores a trip expense someone else paid for", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const trip = await pool.query(
+      `INSERT INTO trips (owner_id, name) VALUES ($1, 'Owner Scope Trip') RETURNING id`,
+      [ownerId]
+    );
+    const tripId = trip.rows[0].id as number;
+
+    // Bob paid this one. Alice's share of it is a debt Alice owes Bob — it is
+    // not something Bob owes Alice, so it must not appear on Alice's trip view.
+    const bobPaid = await insertTransaction(pool, otherId, { amount: 500, category: "travel" });
+    await pool.query(`INSERT INTO transaction_overrides (transaction_id, trip_id) VALUES ($1, $2)`, [bobPaid, tripId]);
+    await pool.query(
+      `INSERT INTO transaction_splits (transaction_id, participant_id, share_percent)
+       VALUES ($1, $2, 50), ($1, $3, 50)`,
+      [bobPaid, ownerId, otherId]
+    );
+
+    const { participant_splits } = await getTripAnalytics(tripId, ownerId);
+    const bob = participant_splits.find((r) => r.participant_id === otherId);
+    expect(bob === undefined || Number(bob.owed) === 0).toBe(true);
+  });
+
+  it("ignores a payment settled between the other two participants", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const third = await pool.query(
+      `INSERT INTO participants (name, email) VALUES ('Carol', 'carol@example.com') RETURNING id`
+    );
+    const carolId = third.rows[0].id as number;
+    const trip = await pool.query(
+      `INSERT INTO trips (owner_id, name) VALUES ($1, 'Third Party Trip') RETURNING id`,
+      [ownerId]
+    );
+    const tripId = trip.rows[0].id as number;
+
+    const txId = await insertTransaction(pool, ownerId, { amount: 300, category: "travel" });
+    await pool.query(`INSERT INTO transaction_overrides (transaction_id, trip_id) VALUES ($1, $2)`, [txId, tripId]);
+    await pool.query(
+      `INSERT INTO transaction_splits (transaction_id, participant_id, share_percent) VALUES ($1, $2, 50)`,
+      [txId, carolId]
+    );
+    // Carol pays Bob, not the owner. Carol still owes the owner $150.
+    await pool.query(
+      `INSERT INTO split_payments (from_participant_id, to_participant_id, amount, payment_date, trip_id)
+       VALUES ($1, $2, 150, '2026-03-15', $3)`,
+      [carolId, otherId, tripId]
+    );
+
+    const { participant_splits } = await getTripAnalytics(tripId, ownerId);
+    const carol = participant_splits.find((r) => r.participant_id === carolId);
+    expect(Number(carol!.owed)).toBeCloseTo(150);
+  });
+});
+
+// A refunded trip expense must not still read as trip cost. These queries
+// filtered on debit/fee/interest, so a refund was dropped entirely and the
+// original purchase stood at full value.
+describe("getTripAnalytics — refunds reduce trip cost", () => {
+  async function seedTripWithRefund(ownerId: number) {
+    const trip = await pool.query(
+      `INSERT INTO trips (owner_id, name, start_date, end_date)
+       VALUES ($1, 'Refund Trip', '2026-03-01', '2026-03-10') RETURNING id`,
+      [ownerId]
+    );
+    const tripId = trip.rows[0].id as number;
+
+    const spend = await insertTransaction(pool, ownerId, {
+      amount: 200, category: "travel", description: "Hotel booking", transaction_date: "2026-03-02",
+    });
+    const refund = await insertTransaction(pool, ownerId, {
+      amount: 50, category: "travel", description: "Hotel partial refund",
+      transaction_type: "refund", transaction_date: "2026-03-05",
+    });
+    await pool.query(
+      `INSERT INTO transaction_overrides (transaction_id, trip_id) VALUES ($1, $3), ($2, $3)`,
+      [spend, refund, tripId]
+    );
+    return tripId;
+  }
+
+  it("nets the refund out of the category total", async () => {
+    const { ownerId } = await seedParticipants(pool);
+    const tripId = await seedTripWithRefund(ownerId);
+
+    const { category_breakdown } = await getTripAnalytics(tripId, ownerId);
+    const travel = category_breakdown.find((c) => c.category === "travel");
+    expect(Number(travel!.amount)).toBeCloseTo(150);
+  });
+
+  it("nets the refund out of the trip's headline total_spend", async () => {
+    const { ownerId } = await seedParticipants(pool);
+    const tripId = await seedTripWithRefund(ownerId);
+
+    const trip = await getTripById(tripId, ownerId);
+    expect(Number(trip!.total_spend)).toBeCloseTo(150);
+  });
+
+  it("shows the refund as a negative on its own day", async () => {
+    const { ownerId } = await seedParticipants(pool);
+    const tripId = await seedTripWithRefund(ownerId);
+
+    const { daily_spend } = await getTripAnalytics(tripId, ownerId);
+    const refundDay = daily_spend.find((d) => d.date === "2026-03-05");
+    expect(Number(refundDay!.amount)).toBeCloseTo(-50);
+  });
+
+  // The owed side must be untouched: a refund carries no split, and the owed
+  // query deliberately excludes credits. Netting cost must not move a balance.
+  it("leaves what the other participant owes unchanged", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const tripId = await seedTripWithRefund(ownerId);
+    const rows = await pool.query(
+      `SELECT transaction_id FROM transaction_overrides WHERE trip_id = $1 ORDER BY transaction_id`,
+      [tripId]
+    );
+    await pool.query(
+      `INSERT INTO transaction_splits (transaction_id, participant_id, share_percent) VALUES ($1, $2, 50)`,
+      [rows.rows[0].transaction_id, otherId]
+    );
+
+    const { participant_splits } = await getTripAnalytics(tripId, ownerId);
+    const bob = participant_splits.find((r) => r.participant_id === otherId);
+    expect(Number(bob!.owed)).toBeCloseTo(100);
+  });
+});
+
+// An account cannot be billed twice for the same day. The boundary handling is
+// the whole difficulty: these statements are issued back-to-back with one
+// period ending the day the next begins, so naive inclusive ranges flag every
+// consecutive pair.
+describe("getStatements — overlapping billing periods", () => {
+  async function addStatement(
+    ownerId: number, account: string, start: string | null, end: string | null
+  ): Promise<number> {
+    const r = await pool.query(
+      `INSERT INTO statements (filename, bank_name, account_number, owner_id,
+                               billing_start_date, billing_end_date)
+       VALUES ($1, 'ANZ', $2, $3, $4, $5) RETURNING id`,
+      [`stmt-${account}-${start}.pdf`, account, ownerId, start, end]
+    );
+    return r.rows[0].id as number;
+  }
+
+  it("does not flag statements that merely touch at a boundary", async () => {
+    const { ownerId } = await seedParticipants(pool);
+    await addStatement(ownerId, "4085-56264", "2025-05-16", "2025-11-14");
+    await addStatement(ownerId, "4085-56264", "2025-11-14", "2026-05-15");
+
+    const rows = await getStatements(ownerId);
+    expect(rows.every((r) => r.overlaps.length === 0)).toBe(true);
+  });
+
+  it("flags a genuine overlap on both statements, with the day count", async () => {
+    const { ownerId } = await seedParticipants(pool);
+    const a = await addStatement(ownerId, "4085-56264", "2025-11-12", "2026-03-12");
+    const b = await addStatement(ownerId, "4085-56264", "2025-11-14", "2026-05-15");
+
+    const rows = await getStatements(ownerId);
+    const rowA = rows.find((r) => r.id === a)!;
+    const rowB = rows.find((r) => r.id === b)!;
+    expect(rowA.overlaps).toEqual([{ id: b, days: 118 }]);
+    expect(rowB.overlaps).toEqual([{ id: a, days: 118 }]);
+  });
+
+  // The real duplicate got in because the existing key compared raw text and
+  // ANZ wrote the same account both ways.
+  it("matches the same account written with and without punctuation", async () => {
+    const { ownerId } = await seedParticipants(pool);
+    const a = await addStatement(ownerId, "408556264", "2025-11-12", "2026-03-12");
+    const b = await addStatement(ownerId, "4085-56264", "2025-11-14", "2026-05-15");
+
+    const rows = await getStatements(ownerId);
+    expect(rows.find((r) => r.id === a)!.overlaps).toEqual([{ id: b, days: 118 }]);
+  });
+
+  it("ignores a different account billing the same days", async () => {
+    const { ownerId } = await seedParticipants(pool);
+    await addStatement(ownerId, "4085-56264", "2025-11-12", "2026-03-12");
+    await addStatement(ownerId, "9999-11111", "2025-11-12", "2026-03-12");
+
+    const rows = await getStatements(ownerId);
+    expect(rows.every((r) => r.overlaps.length === 0)).toBe(true);
+  });
+
+  // NULL is unbounded to daterange, which would make an undated statement
+  // overlap the entire history.
+  it("does not treat an undated statement as overlapping everything", async () => {
+    const { ownerId } = await seedParticipants(pool);
+    await addStatement(ownerId, "4085-56264", "2025-11-12", "2026-03-12");
+    await addStatement(ownerId, "4085-56264", null, null);
+
+    const rows = await getStatements(ownerId);
+    expect(rows.every((r) => r.overlaps.length === 0)).toBe(true);
+  });
+});
+
+// A statement imported twice puts every transaction in the overlap in the
+// ledger twice. The duplicate is superseded rather than deleted, because every
+// child of `transactions` cascades on delete.
+describe("superseded duplicates are excluded but kept", () => {
+  it("hides a superseded row from the transaction list", async () => {
+    const { ownerId } = await seedParticipants(pool);
+    const keep = await insertTransaction(pool, ownerId, { description: "RAIZ INVESTMENT", amount: 1500 });
+    const dup = await insertTransaction(pool, ownerId, { description: "RAIZ INVESTMENT", amount: 1500 });
+    await pool.query(`UPDATE transactions SET superseded_by_id = $1 WHERE id = $2`, [keep, dup]);
+
+    const { data, total } = await getTransactions(ownerId, { limit: 50, offset: 0 });
+    expect(total).toBe(1);
+    expect(data.map((t) => t.id)).toEqual([keep]);
+  });
+
+  it("keeps the superseded row and its children in the database", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const keep = await insertTransaction(pool, ownerId, { amount: 100 });
+    const dup = await insertTransaction(pool, ownerId, { amount: 100 });
+    await pool.query(
+      `INSERT INTO transaction_splits (transaction_id, participant_id, share_percent) VALUES ($1, $2, 50)`,
+      [dup, otherId]
+    );
+    await pool.query(`UPDATE transactions SET superseded_by_id = $1 WHERE id = $2`, [keep, dup]);
+
+    const rows = await pool.query(`SELECT superseded_by_id FROM transactions WHERE id = $1`, [dup]);
+    expect(rows.rows[0].superseded_by_id).toBe(keep);
+    const kids = await pool.query(`SELECT count(*)::int AS n FROM transaction_splits WHERE transaction_id = $1`, [dup]);
+    expect(kids.rows[0].n).toBe(1);
+  });
+
+  // The point of excluding it: a split on a duplicate must not be owed twice.
+  it("does not count a superseded row towards what someone owes", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const keep = await insertTransaction(pool, ownerId, { amount: 100 });
+    const dup = await insertTransaction(pool, ownerId, { amount: 100 });
+    for (const id of [keep, dup]) {
+      await pool.query(
+        `INSERT INTO transaction_splits (transaction_id, participant_id, share_percent) VALUES ($1, $2, 50)`,
+        [id, otherId]
+      );
+    }
+    await pool.query(`UPDATE transactions SET superseded_by_id = $1 WHERE id = $2`, [keep, dup]);
+
+    const balances = await getParticipantBalances(ownerId);
+    const bob = balances.find((b) => b.id === otherId);
+    expect(Number(bob!.total_owed)).toBeCloseTo(50);
+  });
+
+  it("refuses to let a row supersede itself", async () => {
+    const { ownerId } = await seedParticipants(pool);
+    const id = await insertTransaction(pool, ownerId);
+    await expect(
+      pool.query(`UPDATE transactions SET superseded_by_id = $1 WHERE id = $1`, [id])
+    ).rejects.toThrow();
+  });
+});
+
+// Nothing before the cutover can be owed: carryover transaction 2348 already
+// carries the entire pre-cutover balance as one figure. Splits on older
+// transactions exist to describe how an expense was shared -- which keeps it
+// out of spend -- without asserting a debt.
+describe("the split cutover gates every balance", () => {
+  it("ignores a split on a transaction before the cutover", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const txId = await insertTransaction(pool, ownerId, {
+      amount: 100, transaction_date: "2026-01-08",
+    });
+    await pool.query(
+      `INSERT INTO transaction_splits (transaction_id, participant_id, share_percent) VALUES ($1, $2, 50)`,
+      [txId, otherId]
+    );
+
+    const balances = await getParticipantBalances(ownerId);
+    const bob = balances.find((b) => b.id === otherId);
+    expect(Number(bob?.total_owed ?? 0)).toBeCloseTo(0);
+  });
+
+  // Inclusive: transaction 2348, which carries the whole pre-cutover balance,
+  // is itself dated 2026-01-09. An exclusive bound would drop it.
+  it("counts a split dated exactly on the cutover", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const txId = await insertTransaction(pool, ownerId, {
+      amount: 100, transaction_date: "2026-01-09",
+    });
+    await pool.query(
+      `INSERT INTO transaction_splits (transaction_id, participant_id, share_percent) VALUES ($1, $2, 50)`,
+      [txId, otherId]
+    );
+
+    const balances = await getParticipantBalances(ownerId);
+    const bob = balances.find((b) => b.id === otherId);
+    expect(Number(bob!.total_owed)).toBeCloseTo(50);
+  });
+
+  // The point of the date guard: it does not depend on `settled` surviving.
+  it("still ignores a pre-cutover split whose settled flag was lost", async () => {
+    const { ownerId, otherId } = await seedParticipants(pool);
+    const txId = await insertTransaction(pool, ownerId, {
+      amount: 200, transaction_date: "2025-06-01",
+    });
+    await pool.query(
+      `INSERT INTO transaction_splits (transaction_id, participant_id, share_percent, settled)
+       VALUES ($1, $2, 50, false)`,
+      [txId, otherId]
+    );
+
+    const balances = await getParticipantBalances(ownerId);
+    const bob = balances.find((b) => b.id === otherId);
+    expect(Number(bob?.total_owed ?? 0)).toBeCloseTo(0);
+  });
+});

src/__tests__/unit/order-ingestion.test.ts

@@ -5,8 +5,8 @@ import {
   parseOrderHTML,
   validateOrderTotals,
   resolveCategory,
-  OrderParseError,
   NotAReceiptError,
+  orderDescription,
   type MessageMeta,
   type ParsedOrder,
 } from "../../lib/order-ingestion";
@@ -71,6 +71,7 @@ describe("validateOrderTotals", () => {
       service_fee: null, tip: null, discounts: null, total_charged: 10,
     },
     line_items: [],
+    route: [],
     is_family: false,
     flags: [],
     ...over,
@@ -164,3 +165,158 @@ describe("order_reference anchoring", () => {
     expect(p.flags).toContain("order_uuid_ambiguous");
   });
 });
+
+describe("mixed Uber payment (issuer-named card leg)", () => {
+  it("captures both legs when the card is labelled by issuer, not brand", () => {
+    // Real receipt: Uber Cash $1.17 + Westpac ••••8032 $15.33 = $16.50.
+    // A brand allowlist (Visa|MasterCard|Amex) misses "Westpac" and drops the
+    // card half, leaving payments that do not account for the total.
+    const p = parseOrderHTML(
+      readFileSync(resolve(dir, "ue-mixed.html"), "utf-8"),
+      meta({ subject: "Your Friday morning order with Uber Eats", sender: "uber.com", receivedAt: "2026-01-09T09:26:44Z" })
+    );
+    expect(p.totals.total_charged).toBeCloseTo(16.50, 2);
+    expect(p.payment.credits_amount).toBeCloseTo(1.17, 2);
+    expect(p.payment.card_amount).toBeCloseTo(15.33, 2);
+    expect(p.payment.card_last4).toBe("8032");
+    expect(validateOrderTotals(p).ok).toBe(true);
+  });
+});
+
+describe("Uber route (pick-up / delivery)", () => {
+  const uber = (f: string, subject = "Your Wednesday order with Uber Eats") =>
+    parseOrderHTML(html(f), meta({ subject, sender: "uber.com" }));
+
+  it("reads both stops with their times, as printed", () => {
+    const p = uber("ue-00");
+    expect(p.route).toEqual([
+      { label: "Pick-up", time: "1:20 pm", address: "197 Watton St, Werribee VIC 3030, Australia" },
+      { label: "Delivery", time: "1:40 pm", address: "19 Lady Penrhyn Dr, Wyndham Vale VIC 3024, Australia" },
+    ]);
+  });
+
+  it("de-duplicates the block Uber renders twice", () => {
+    // The receipt emits the whole address section a second time for narrow
+    // screens. Without de-duplication every trip has four stops.
+    expect(uber("ue-00").route).toHaveLength(2);
+    expect(uber("ue-26").route).toHaveLength(2);
+  });
+
+  it("keeps the receipt's own wording rather than normalising it", () => {
+    // Uber is not internally consistent: "Pick-up" on some receipts,
+    // "Pickup" on others. Inventing a canonical spelling would hide that a
+    // template changed.
+    expect(uber("ue-mixed", "Your Friday morning order with Uber Eats").route[0].label).toBe("Pickup");
+  });
+
+  it("works on an international receipt", () => {
+    const p = uber("ue-26");
+    expect(p.route[1].address).toContain("Luzern, Switzerland");
+  });
+
+  it("DoorDash has no route — its receipts carry no addresses", () => {
+    expect(parseOrderHTML(html("dd-01"), meta()).route).toEqual([]);
+  });
+});
+
+describe("Uber line items", () => {
+  it("itemises a grocery order, binding qty/title/amount by item id", () => {
+    const p = parseOrderHTML(
+      html("ue-09"),
+      meta({ subject: "Your Sunday evening order with Uber Eats", sender: "uber.com" })
+    );
+    expect(p.line_items).toHaveLength(5);
+    expect(p.line_items[0]).toMatchObject({
+      qty: 1,
+      description: "Highland Brewing MILK FULL CREAM U H T 900ML",
+      amount: 440,
+    });
+    // A sold-out item prints 0.00 and is kept: it is why the total is lower
+    // than what was ordered, and dropping it makes the receipt unexplainable.
+    expect(p.line_items.map((i) => i.amount)).toContain(0);
+  });
+
+  it("a restaurant order legitimately has none", () => {
+    // Uber itemises groceries only; a restaurant receipt states a total and
+    // nothing else. Empty here is the receipt, not a parse failure — so it
+    // must not raise no_line_items_parsed either.
+    const p = parseOrderHTML(
+      html("ue-00"),
+      meta({ subject: "Your Wednesday order with Uber Eats", sender: "uber.com" })
+    );
+    expect(p.line_items).toEqual([]);
+    expect(p.flags).not.toContain("no_line_items_parsed");
+  });
+});
+
+/**
+ * Uber trips. Captured 2026-06 via a dry-run against the real mailbox after the
+ * user pointed out that only *overseas* rides go on a card — local rides are
+ * paid with credits, which puts them in the same class as delivery orders.
+ */
+describe("Uber trips", () => {
+  const utMeta: Record<string, MessageMeta> = JSON.parse(
+    readFileSync(resolve(dir, "ut-meta.json"), "utf-8")
+  );
+  const trip = (f: string) => parseOrderHTML(html(f), utMeta[f]);
+
+  it("a local trip is credits-funded", () => {
+    const p = trip("ut-00");
+    expect(p.platform).toBe("uber");
+    expect(p.currency).toBe("AUD");
+    expect(p.totals.total_charged).toBeCloseTo(84.78, 2);
+    expect(p.payment.credits_amount).toBeCloseTo(84.78, 2);
+    expect(p.payment.card_last4).toBeNull();
+    expect(validateOrderTotals(p).ok).toBe(true);
+  });
+
+  it("an overseas trip is card-settled", () => {
+    const p = trip("ut-01");
+    expect(p.currency).toBe("NZD");
+    expect(p.totals.total_charged).toBeCloseTo(55.51, 2);
+    expect(p.payment.card_amount).toBeCloseTo(55.51, 2);
+    expect(p.payment.card_last4).toBe("3893");
+  });
+
+  it("labels the two ends of a trip, which the receipt does not", () => {
+    // Delivery receipts write "1:20 pm - Pick-up"; trip receipts print the time
+    // alone. The naive split put the time in `label` and left `time` null.
+    const p = trip("ut-00");
+    expect(p.route).toEqual([
+      {
+        label: "Pick-up",
+        time: "7:32 pm",
+        address: "Terminal 2, Melbourne Airport (MEL), Tullamarine VIC 3045, Australia",
+      },
+      {
+        label: "Drop-off",
+        time: "8:10 pm",
+        address: "19 Lady Penrhyn Dr, Wyndham Vale VIC 3024, Australia",
+      },
+    ]);
+  });
+
+  it("rejects the charge summary Uber sends before the receipt", () => {
+    // Uber sends two mails per trip with the same subject and the same total.
+    // The first says "This is not a payment receipt" and carries no
+    // tripReference, so order_reference would fall back to msg:<id> and I7
+    // could not dedupe it — every trip would be recorded twice.
+    expect(() => trip("ut-summary")).toThrow(NotAReceiptError);
+  });
+});
+
+describe("orderDescription", () => {
+  it("names the platform", () => {
+    expect(orderDescription("doordash", "Mad Mex")).toBe("Order - Mad Mex (DoorDash)");
+    expect(orderDescription("ubereats", "Coles (Wyndham Vale)")).toBe(
+      "Order - Coles (Wyndham Vale) (Uber Eats)"
+    );
+  });
+
+  it("does not restate a platform the merchant already names", () => {
+    // A trip's merchant is literally "Uber Trip"; "(Uber)" after it says
+    // nothing. What identifies a trip is its addresses, and those live in the
+    // Order details panel.
+    expect(orderDescription("uber", "Uber Trip")).toBe("Order - Uber Trip");
+  });
+});

src/app/api/analytics/daily/route.ts

@@ -0,0 +1,87 @@
+import { NextRequest, NextResponse } from "next/server";
+import { getCurrentUser } from "@/lib/auth";
+import { queryRaw } from "@/lib/db";
+import {
+  OWNER_SCOPE,
+  STATEMENTS_JOIN,
+  EXCLUDE_NON_SPEND,
+  EXCLUDE_RECONCILED_SOURCE,
+  EFFECTIVE_CATEGORY,
+  NET_SPEND_ROWS,
+  SPEND_SIGNED,
+  mySplitOf,
+  toDateStr,
+} from "@/lib/analytics-sql";
+
+/**
+ * Daily net spend, by month, day-of-month and category.
+ *
+ * This exists so the spend-pace chart stops computing its own totals. It used to
+ * sum gross `amount_aud ?? amount` over `transaction_type = 'debit'` in the
+ * browser, which meant it ignored personal share, refunds, fees, interest, and
+ * itemised loan repayments — every rule the headline applies. The two numbers
+ * could disagree while both were labelled "spend", and the chart's own baseline
+ * line was drawn from the split-adjusted monthly totals, so the two series in
+ * one chart were on different bases.
+ *
+ * Day-of-month granularity is also what lets the page compare a partial current
+ * month against prior months *through the same day*, instead of against their
+ * full-month totals — which always made a month in progress look thrifty.
+ *
+ * Same fragments as /api/analytics/monthly. If that route's semantics change,
+ * this one changes with it.
+ */
+export async function GET(req: NextRequest) {
+  const user = await getCurrentUser(req);
+  if (!user) return NextResponse.json({ error: "Unauthorized" }, { status: 403 });
+
+  const { searchParams } = new URL(req.url);
+  const monthCount = Math.min(Math.max(Number(searchParams.get("months") || "12"), 1), 24);
+
+  const now = new Date();
+  const endDate = new Date(now.getFullYear(), now.getMonth() + 1, 1);
+  const startDate = new Date(now.getFullYear(), now.getMonth() - monthCount + 1, 1);
+
+  const rows = await queryRaw<{ month: string; day: number; category: string; spent: string }>(
+    `SELECT
+      TO_CHAR(DATE_TRUNC('month', t.transaction_date::date), 'YYYY-MM') as month,
+      EXTRACT(DAY FROM t.transaction_date::date)::int as day,
+      ${EFFECTIVE_CATEGORY} as category,
+      -- 4dp, not 2. This is grouped finer than /monthly (by day as well as
+      -- category), so rounding each bucket to cents and summing accumulates a
+      -- different error than rounding per category does — the pace chart ended
+      -- the month a few cents off the headline it sits under. Round once, at
+      -- display time.
+      SUM(${mySplitOf(SPEND_SIGNED)})::numeric(14,4) as spent
+    FROM transactions t
+    LEFT JOIN transaction_overrides o ON o.transaction_id = t.id
+    LEFT JOIN transaction_splits ts ON ts.transaction_id = t.id AND ts.participant_id = $1
+    ${STATEMENTS_JOIN}
+    WHERE ${OWNER_SCOPE} = $1
+      AND ${NET_SPEND_ROWS}
+      AND ${EXCLUDE_NON_SPEND}
+      AND ${EXCLUDE_RECONCILED_SOURCE}
+      AND t.transaction_date >= $2
+      AND t.transaction_date < $3
+    GROUP BY 1, 2, 3
+    ORDER BY 1, 2`,
+    [user.id, toDateStr(startDate), toDateStr(endDate)]
+  );
+
+  // Sparse by design — a day with no spend has no entry, and the client treats
+  // a missing day as zero. Emitting 31 zeroes per month per category would
+  // dominate the payload.
+  const daily: Record<string, Record<number, number>> = {};
+  const byCategory: Record<string, Record<string, Record<number, number>>> = {};
+
+  for (const r of rows) {
+    const spent = Number(r.spent);
+    const m = (daily[r.month] ??= {});
+    m[r.day] = (m[r.day] ?? 0) + spent;
+
+    const c = ((byCategory[r.month] ??= {})[r.category] ??= {});
+    c[r.day] = (c[r.day] ?? 0) + spent;
+  }
+
+  return NextResponse.json({ daily, byCategory });
+}

src/app/api/analytics/fees/route.ts

@@ -1,12 +1,35 @@
 import { NextRequest, NextResponse } from "next/server";
 import { getCurrentUser } from "@/lib/auth";
 import { queryRaw } from "@/lib/db";
-import { OWNER_SCOPE, STATEMENTS_JOIN, mySplitOf } from "@/lib/analytics-sql";
+import { OWNER_SCOPE, STATEMENTS_JOIN, EXCLUDE_RECONCILED_SOURCE, mySplitOf, toDateStr } from "@/lib/analytics-sql";
 
+/**
+ * Fees and interest over an explicit window.
+ *
+ * This used to aggregate every statement ever imported with no date filter, and
+ * the UI printed the result with no period label — so a lifetime-to-date total
+ * read as a current-period one, and grew forever. `months=0` asks for all time
+ * deliberately, which is a different claim from asking for it by accident.
+ */
 export async function GET(req: NextRequest) {
   const user = await getCurrentUser(req);
   if (!user) return NextResponse.json({ error: "Unauthorized" }, { status: 403 });
 
+  const { searchParams } = new URL(req.url);
+  const monthsParam = Number(searchParams.get("months") ?? "12");
+  const months = Number.isFinite(monthsParam) ? Math.min(Math.max(monthsParam, 0), 120) : 12;
+  const allTime = months === 0;
+
+  const now = new Date();
+  const from = new Date(now.getFullYear(), now.getMonth() - months + 1, 1);
+  const fromStr = toDateStr(from);
+  const toStr = toDateStr(new Date(now.getFullYear(), now.getMonth() + 1, 1));
+
+  // A statement is dated by the period it covers, not by when it was imported.
+  const stmtWindow = allTime ? "" : `AND billing_end_date >= $2 AND billing_end_date < $3`;
+  const txnWindow = allTime ? "" : `AND t.transaction_date >= $2 AND t.transaction_date < $3`;
+  const windowParams = allTime ? [] : [fromStr, toStr];
+
   // Statement-level fees and interest (aggregated by Gemini from the PDF)
   const stmtRows = await queryRaw<{
     bank_name: string;
@@ -19,10 +42,11 @@ export async function GET(req: NextRequest) {
       SUM(COALESCE(interest_charged, 0))::numeric(12,2) AS interest
     FROM statements
     WHERE owner_id = $1
+    ${stmtWindow}
     GROUP BY bank_name
     HAVING SUM(COALESCE(fees_charged, 0)) + SUM(COALESCE(interest_charged, 0)) > 0
     ORDER BY (SUM(COALESCE(fees_charged, 0)) + SUM(COALESCE(interest_charged, 0))) DESC`,
-    [user.id]
+    [user.id, ...windowParams]
   );
 
   // Transaction-level fee and interest line items (split-adjusted)
@@ -49,8 +73,10 @@ export async function GET(req: NextRequest) {
     ${STATEMENTS_JOIN}
     WHERE ${OWNER_SCOPE} = $1
       AND t.transaction_type IN ('fee', 'interest')
+      AND ${EXCLUDE_RECONCILED_SOURCE}
+      ${txnWindow}
     ORDER BY t.transaction_date DESC`,
-    [user.id]
+    [user.id, ...windowParams]
   );
 
   const by_bank = stmtRows.map((r) => ({
@@ -69,5 +95,13 @@ export async function GET(req: NextRequest) {
   const total_fees = by_bank.reduce((s, r) => s + r.fees, 0);
   const total_interest = by_bank.reduce((s, r) => s + r.interest, 0);
 
-  return NextResponse.json({ by_bank, transactions, total_fees, total_interest });
+  return NextResponse.json({
+    by_bank,
+    transactions,
+    total_fees,
+    total_interest,
+    // The period is part of the answer — the client must be able to say what
+    // window these totals cover rather than implying "now".
+    period: { months, from: allTime ? null : fromStr, to: allTime ? null : toStr, all_time: allTime },
+  });
 }

src/app/api/analytics/merchants/[merchant]/route.ts

@@ -1,7 +1,8 @@
 import { NextRequest, NextResponse } from "next/server";
 import { getCurrentUser } from "@/lib/auth";
 import { queryRaw } from "@/lib/db";
-import { OWNER_SCOPE, STATEMENTS_JOIN, EFFECTIVE_CATEGORY, mySplitOf } from "@/lib/analytics-sql";
+import { OWNER_SCOPE, STATEMENTS_JOIN, EXCLUDE_RECONCILED_SOURCE, EFFECTIVE_CATEGORY, mySplitOf } from "@/lib/analytics-sql";
+import { bankLabel } from "@/lib/queries";
 
 const MY_AMOUNT = mySplitOf(`COALESCE(t.amount_aud, t.amount)`);
 
@@ -39,7 +40,7 @@ export async function GET(
       END::numeric(10,2) as my_amount,
       t.transaction_type,
       ${EFFECTIVE_CATEGORY} as category,
-      COALESCE(s.bank_name, 'Manual') as bank_name,
+      ${bankLabel()} as bank_name,
       t.statement_id
     FROM transactions t
     ${STATEMENTS_JOIN}
@@ -48,6 +49,7 @@ export async function GET(
     WHERE ${OWNER_SCOPE} = $1
       AND t.transaction_type IN ('debit', 'fee', 'interest', 'refund', 'credit')
       AND COALESCE(o.merchant_normalized, t.merchant_normalized, t.merchant_name, t.description) = $2
+      AND ${EXCLUDE_RECONCILED_SOURCE}
     ORDER BY t.transaction_date DESC
     LIMIT 500
   `, [user.id, decoded]);

src/app/api/analytics/merchants/route.ts

@@ -1,7 +1,7 @@
 import { NextRequest, NextResponse } from "next/server";
 import { getCurrentUser } from "@/lib/auth";
 import { queryRaw } from "@/lib/db";
-import { OWNER_SCOPE, STATEMENTS_JOIN, EXCLUDE_NON_SPEND, EFFECTIVE_CATEGORY, mySplitOf } from "@/lib/analytics-sql";
+import { OWNER_SCOPE, STATEMENTS_JOIN, EXCLUDE_NON_SPEND, EXCLUDE_RECONCILED_SOURCE, EFFECTIVE_CATEGORY, mySplitOf, toDateStr } from "@/lib/analytics-sql";
 
 // Split-adjusted amount helper (positive for spend, negative for refunds)
 const MY_AMOUNT = mySplitOf(`COALESCE(t.amount_aud, t.amount)`);
@@ -21,7 +21,7 @@ export async function GET(req: NextRequest) {
 
   const cutoff = new Date();
   cutoff.setMonth(cutoff.getMonth() - months);
-  const fromDate = cutoff.toISOString().slice(0, 10);
+  const fromDate = toDateStr(cutoff);
 
   // Merchant aggregates — net spend (debits + fees - refunds/credits)
   const rows = await queryRaw<{
@@ -69,6 +69,7 @@ export async function GET(req: NextRequest) {
       AND t.transaction_type IN ('debit', 'fee', 'interest', 'refund', 'credit')
       AND t.transaction_date >= $2
       AND ${EXCLUDE_NON_SPEND}
+      AND ${EXCLUDE_RECONCILED_SOURCE}
     GROUP BY 1
     HAVING SUM(${SPEND_EXPR}) > 0
     ORDER BY net_spend DESC
@@ -95,6 +96,7 @@ export async function GET(req: NextRequest) {
         AND t.transaction_date >= $2
         AND COALESCE(o.merchant_normalized, t.merchant_normalized, t.merchant_name, t.description) = ANY($3)
         AND ${EXCLUDE_NON_SPEND}
+        AND ${EXCLUDE_RECONCILED_SOURCE}
       GROUP BY 1, 2
       ORDER BY 1, 2
     `, [user.id, fromDate, topMerchants]);

src/app/api/analytics/monthly/route.ts

@@ -6,9 +6,11 @@ import {
   STATEMENTS_JOIN,
   EFFECTIVE_CATEGORY,
   EXCLUDE_NON_SPEND,
+  EXCLUDE_RECONCILED_SOURCE,
   NET_SPEND_ROWS,
   SPEND_SIGNED,
   mySplitOf,
+  toDateStr,
 } from "@/lib/analytics-sql";
 
 export async function GET(req: NextRequest) {
@@ -22,8 +24,8 @@ export async function GET(req: NextRequest) {
   const endDate = new Date(now.getFullYear(), now.getMonth() + 1, 1);
   const startDate = new Date(now.getFullYear(), now.getMonth() - monthCount + 1, 1);
 
-  const startStr = startDate.toISOString().slice(0, 10);
+  const startStr = toDateStr(startDate);
-  const endStr = endDate.toISOString().slice(0, 10);
+  const endStr = toDateStr(endDate);
 
   // Expenses: debits excluding transfers and investments, split-adjusted
   const spendRows = await queryRaw<{
@@ -35,7 +37,9 @@ export async function GET(req: NextRequest) {
     `SELECT
       TO_CHAR(DATE_TRUNC('month', t.transaction_date::date), 'YYYY-MM') as month,
       ${EFFECTIVE_CATEGORY} as category,
-      SUM(${mySplitOf(SPEND_SIGNED)})::numeric(12,2) as total_spent,
+      -- 4dp so the month total is summed from unrounded parts; every consumer
+      -- rounds for display. See the note in /api/analytics/daily.
+      SUM(${mySplitOf(SPEND_SIGNED)})::numeric(14,4) as total_spent,
       COUNT(*)::int as transaction_count
     FROM transactions t
     LEFT JOIN transaction_overrides o ON o.transaction_id = t.id
@@ -44,6 +48,7 @@ export async function GET(req: NextRequest) {
     WHERE ${OWNER_SCOPE} = $1
       AND ${NET_SPEND_ROWS}
       AND ${EXCLUDE_NON_SPEND}
+      AND ${EXCLUDE_RECONCILED_SOURCE}
       AND t.transaction_date >= $2
       AND t.transaction_date < $3
     GROUP BY 1, 2
@@ -67,6 +72,7 @@ export async function GET(req: NextRequest) {
     WHERE ${OWNER_SCOPE} = $1
       AND t.transaction_type IN ('credit', 'payment')
       AND ${EFFECTIVE_CATEGORY} = 'income'
+      AND ${EXCLUDE_RECONCILED_SOURCE}
       AND t.transaction_date >= $2
       AND t.transaction_date < $3
     GROUP BY 1
@@ -89,6 +95,7 @@ export async function GET(req: NextRequest) {
     ${STATEMENTS_JOIN}
     WHERE ${OWNER_SCOPE} = $1
       AND ${EFFECTIVE_CATEGORY} = 'investment'
+      AND ${EXCLUDE_RECONCILED_SOURCE}
       AND t.transaction_date >= $2
       AND t.transaction_date < $3
     GROUP BY 1

src/app/api/analytics/subscriptions/route.ts

@@ -1,7 +1,7 @@
 import { NextRequest, NextResponse } from "next/server";
 import { getCurrentUser } from "@/lib/auth";
 import { queryRaw } from "@/lib/db";
-import { OWNER_SCOPE, STATEMENTS_JOIN, EXCLUDE_NON_SPEND, EFFECTIVE_CATEGORY, mySplitOf } from "@/lib/analytics-sql";
+import { OWNER_SCOPE, STATEMENTS_JOIN, EXCLUDE_NON_SPEND, EXCLUDE_RECONCILED_SOURCE, EFFECTIVE_CATEGORY, mySplitOf } from "@/lib/analytics-sql";
 
 export async function GET(req: NextRequest) {
   const user = await getCurrentUser(req);
@@ -31,6 +31,7 @@ export async function GET(req: NextRequest) {
       WHERE ${OWNER_SCOPE} = $1
         AND t.transaction_type IN ('debit', 'fee')
         AND ${EXCLUDE_NON_SPEND}
+        AND ${EXCLUDE_RECONCILED_SOURCE}
         AND COALESCE(o.merchant_normalized, t.merchant_normalized, t.merchant_name) IS NOT NULL
     ),
     merchant_with_lag AS (

src/app/api/orders/ingest/route.ts

@@ -72,7 +72,11 @@ export async function POST(req: NextRequest) {
 
     if (dryRun) return NextResponse.json({ kind: "order", order });
 
-    const result = await processOrderIngestion(order, { messageId: meta.messageId });
+    const result = await processOrderIngestion(order, {
+      messageId: meta.messageId,
+      subject: meta.subject,
+      sender: meta.sender,
+    });
     return NextResponse.json({
       kind: "order",
       order_reference: order.order_reference,

src/app/api/participants/[id]/balance/route.ts

@@ -1,35 +0,0 @@
-import { NextRequest, NextResponse } from "next/server";
-import { queryRaw } from "@/lib/db";
-import { getCurrentUser } from "@/lib/auth";
-
-interface BalanceRow {
-  participant_id: number;
-  name: string;
-  total_owed: number;
-  transaction_count: number;
-}
-
-export async function GET(
-  req: NextRequest,
-  { params }: { params: Promise<{ id: string }> }
-) {
-  const user = await getCurrentUser(req);
-  if (!user) return NextResponse.json({ error: "Unauthorized" }, { status: 403 });
-  const { id } = await params;
-
-  const rows = await queryRaw<BalanceRow>(
-    `SELECT ts.participant_id, p.name,
-       SUM(COALESCE(t.amount_aud, t.amount) * ts.share_percent / 100)::numeric(12,2) as total_owed,
-       COUNT(*)::int as transaction_count
-     FROM transaction_splits ts
-     JOIN transactions t ON t.id = ts.transaction_id
-     JOIN participants p ON p.id = ts.participant_id
-     WHERE ts.participant_id = $1 AND ts.settled = false
-     GROUP BY ts.participant_id, p.name`,
-    [Number(id)]
-  );
-
-  return NextResponse.json(
-    rows[0] ?? { participant_id: Number(id), total_owed: 0, transaction_count: 0 }
-  );
-}

src/app/api/transactions/[id]/order/route.ts

@@ -0,0 +1,44 @@
+import { NextRequest, NextResponse } from "next/server";
+import { queryRow } from "@/lib/db";
+import { getCurrentUser } from "@/lib/auth";
+import { canAccessTransactions } from "@/lib/queries";
+
+/**
+ * Order provenance for one transaction.
+ *
+ * `expense_metadata` has held the itemised receipt since ingestion started and
+ * nothing in the UI ever read it — a transaction that came from a DoorDash or
+ * Uber Eats receipt showed a merchant and an amount, with the item list and the
+ * delivery addresses sitting unread in the row behind it (user, 2026-07-27).
+ *
+ * Read-only. The receipt is a record of what a provider sent; editing it here
+ * would make provenance mean nothing.
+ */
+export async function GET(
+  req: NextRequest,
+  { params }: { params: Promise<{ id: string }> }
+) {
+  const user = await getCurrentUser(req);
+  if (!user) return NextResponse.json({ error: "Unauthorized" }, { status: 403 });
+  const { id } = await params;
+  if (!(await canAccessTransactions(user.id, [Number(id)]))) {
+    return NextResponse.json({ error: "Forbidden" }, { status: 403 });
+  }
+
+  const row = await queryRow(
+    `SELECT platform, order_reference, line_items, route, subtotal, amount,
+            currency, card_last4, flags, source_email_subject, transaction_date
+       FROM expense_metadata
+      -- A card-settled order creates no transaction of its own (I5): the
+      -- statement line is the transaction, and the receipt points at it
+      -- through matched_transaction_id. Both directions have to resolve or the
+      -- detail is missing on exactly the orders that were paid by card.
+      WHERE transaction_id = $1 OR matched_transaction_id = $1
+      LIMIT 1`,
+    [Number(id)]
+  );
+
+  // Not an order — most transactions aren't. Null, not 404: the caller is
+  // asking "is there a receipt behind this?", and "no" is a normal answer.
+  return NextResponse.json(row ?? null);
+}

src/app/api/transactions/[id]/splits/route.ts

@@ -66,18 +66,47 @@ export async function POST(
     );
   }
 
+  // Carry `settled` across the rewrite.
+  //
+  // This replaces every split rather than editing in place, so without this the
+  // recreated rows take the column default of false — silently converting a
+  // discharged historical obligation into a live debt. That is not theoretical:
+  // 657 pre-2026 transactions carry settled splits imported from
+  // SplitMyExpenses, $37,233.28 of balance that the carryover (transaction
+  // 2348) already accounts for. Editing one would double-count its share, and
+  // nothing on screen would say so.
+  //
+  // Changing someone's percentage does not re-open the obligation — it was
+  // settled outside this app and stays settled. A participant added who was not
+  // there before is a genuinely new obligation and correctly starts unsettled.
+  const previous = await queryRaw<{
+    participant_id: number;
+    settled: boolean;
+    settled_at: string | null;
+  }>(
+    `SELECT participant_id, settled, settled_at
+       FROM transaction_splits WHERE transaction_id = $1`,
+    [transactionId]
+  );
+  const settledBefore = new Map(
+    previous.map((p) => [p.participant_id, { settled: p.settled, settled_at: p.settled_at }])
+  );
+
   // Replace all splits for this transaction atomically
   await prisma.$transaction([
     prisma.transaction_splits.deleteMany({ where: { transaction_id: transactionId } }),
-    ...splits.map((s) =>
+    ...splits.map((s) => {
-      prisma.transaction_splits.create({
+      const before = settledBefore.get(s.participant_id);
+      return prisma.transaction_splits.create({
         data: {
           transaction_id: transactionId,
           participant_id: s.participant_id,
           share_percent: s.share_percent,
+          settled: before?.settled ?? false,
+          settled_at: before?.settled_at ? new Date(before.settled_at) : null,
         },
-      })
+      });
-    ),
+    }),
   ]);
 
   const result = await queryRaw<SplitRow>(

src/app/budget/page.tsx

@@ -16,7 +16,7 @@ import {
   ReferenceLine,
 } from "recharts";
 import { useQueryClient } from "@tanstack/react-query";
-import { useMonthlyAnalytics, useTransactions, useUpdateTransaction } from "@/lib/hooks";
+import { useMonthlyAnalytics, useDailySpend, useTransactions, useUpdateTransaction } from "@/lib/hooks";
 import { formatCategory, CATEGORIES } from "@/lib/categories";
 import { CATEGORY_COLORS, CHART, TOOLTIP_STYLE } from "@/lib/category-colors";
 
@@ -37,6 +37,26 @@ function formatShortMonth(m: string): string {
   const [year, month] = m.split("-");
   return new Date(Number(year), Number(month) - 1, 1).toLocaleString("default", { month: "short" });
 }
+function daysInMonthOf(m: string): number {
+  const [year, month] = m.split("-").map(Number);
+  return new Date(year, month, 0).getDate();
+}
+/**
+ * How much of `month` has actually happened. A month in progress is only
+ * complete up to today; every earlier month is complete.
+ */
+function elapsedDays(m: string): number {
+  return m === currentMonthStr() ? new Date().getDate() : daysInMonthOf(m);
+}
+/** Spend in `month` from day 1 through `throughDay` inclusive. */
+function spendThrough(days: Record<number, number> | undefined, throughDay: number): number {
+  if (!days) return 0;
+  let sum = 0;
+  for (const [d, v] of Object.entries(days)) {
+    if (Number(d) <= throughDay) sum += v;
+  }
+  return sum;
+}
 function fmt(n: number): string { return `$${Math.round(n).toLocaleString()}`; }
 function fmtExact(n: number): string { return `$${n.toFixed(2)}`; }
 function fmtSigned(n: number): string { return `${n >= 0 ? "+" : "−"}$${Math.abs(n) >= 100 ? Math.round(Math.abs(n)).toLocaleString() : Math.abs(n).toFixed(0)}`; }
@@ -195,12 +215,14 @@ export default function AnalyticsPage() {
     // eslint-disable-next-line react-hooks/exhaustive-deps
   }, [months]);
 
-  // Cumulative chart: fetch this month's transactions
+  // Day-of-month spend, split-adjusted server-side by the same rules as the
-  const smFrom = `${selectedMonth}-01`;
+  // headline. Also what makes the comparisons below like-for-like.
-  const [smYear, smMonth] = selectedMonth.split("-").map(Number);
+  const { data: dailyData } = useDailySpend(12);
-  const smNextDate = new Date(smYear, smMonth, 1);
+
-  const smTo = `${smNextDate.getFullYear()}-${String(smNextDate.getMonth() + 1).padStart(2, "0")}-01`;
+  // A month in progress is only comparable to prior months through the same day.
-  const { data: monthTxData } = useTransactions({ from: smFrom, to: smTo, limit: 1000 });
+  // Comparing 27 days of July against a full June always flattered July.
+  const compareDay = elapsedDays(selectedMonth);
+  const selectedIsPartial = selectedMonth === currentMonthStr();
 
   // Category rows for selected month
   const categoryRows = useMemo(() => {
@@ -226,22 +248,27 @@ export default function AnalyticsPage() {
       .slice(0, 8);
   }, [analytics, months, selectedMonth]);
 
-  // Top movers vs previous month
+  // Top movers vs previous month, compared through the same day of the month so
+  // a month in progress is not measured against a complete one.
   const movers = useMemo(() => {
     if (!analytics) return [];
     const pm = prevMonth(selectedMonth);
     if (!months.includes(pm)) return [];
-    return analytics.rows
+
-      .map((r) => ({
+    const catsNow = dailyData?.byCategory?.[selectedMonth] ?? {};
-        category: r.category,
+    const catsBefore = dailyData?.byCategory?.[pm] ?? {};
-        delta: (r.spent[selectedMonth] || 0) - (r.spent[pm] || 0),
+    const categories = new Set([...Object.keys(catsNow), ...Object.keys(catsBefore)]);
-        now: r.spent[selectedMonth] || 0,
+
-        before: r.spent[pm] || 0,
+    return Array.from(categories)
-      }))
+      .map((category) => {
+        const now = spendThrough(catsNow[category], compareDay);
+        const before = spendThrough(catsBefore[category], compareDay);
+        return { category, delta: now - before, now, before };
+      })
       .filter((r) => Math.abs(r.delta) >= 1)
       .sort((a, b) => Math.abs(b.delta) - Math.abs(a.delta))
       .slice(0, 6);
-  }, [analytics, months, selectedMonth]);
+  }, [analytics, dailyData, months, selectedMonth, compareDay]);
 
   // Pareto chart data
   const paretoData = useMemo(() => {
@@ -258,39 +285,40 @@ export default function AnalyticsPage() {
     });
   }, [categoryRows]);
 
-  // Cumulative spend chart data
+  // Cumulative spend chart data.
+  //
+  // Both series now come from the same server-side spend definition as the
+  // headline. The typical line is also a real averaged curve rather than the
+  // month total spread evenly — spending is lumpy (rent on the 1st, a shop on
+  // the weekend), so a straight line made ordinary months look erratic.
   const cumulativeData = useMemo(() => {
-    const daysInMonth = new Date(smYear, smMonth, 0).getDate();
+    const daysInMonth = daysInMonthOf(selectedMonth);
-    const isCurrentMonth = selectedMonth === currentMonthStr();
+    const lastDay = elapsedDays(selectedMonth);
-    const today = new Date();
-    const lastDay = isCurrentMonth ? today.getDate() : daysInMonth;
 
-    const daily: Record<number, number> = {};
+    const daily = dailyData?.daily?.[selectedMonth] ?? {};
-    (monthTxData?.data ?? [])
+    // Only complete months with data form the baseline. A month still in
-      .filter((tx) => tx.transaction_type === "debit" &&
+    // progress has no spend recorded past today, so including it would pull the
-              !["transfers", "investment"].includes(tx.effective_category) &&
+    // typical curve down by however much of it has not happened yet.
-              !tx.tags?.some((t: any) => (typeof t === "string" ? t : t.name) === "family"))
+    const priorMonths = (analytics?.months ?? []).filter(
-      .forEach((tx) => {
+      (m) => m !== selectedMonth && m !== currentMonthStr() && (analytics?.totals[m]?.spent || 0) > 0
-        const day = new Date(tx.transaction_date).getDate();
+    );
-        daily[day] = (daily[day] || 0) + Number(tx.amount_aud ?? tx.amount);
-      });
-
-    const priorMonths = analytics?.months.filter((m) => m !== selectedMonth) ?? [];
-    const priorAvg = priorMonths.length > 0
-      ? priorMonths.reduce((s, m) => s + (analytics?.totals[m]?.spent || 0), 0) / priorMonths.length
-      : 0;
 
     let cum = 0;
     return Array.from({ length: daysInMonth }, (_, i) => {
       const day = i + 1;
       if (day <= lastDay) cum += daily[day] || 0;
+
+      const typical = priorMonths.length
+        ? priorMonths.reduce((s, m) => s + spendThrough(dailyData?.daily?.[m], day), 0) / priorMonths.length
+        : 0;
+
       return {
         day,
         actual: day <= lastDay ? Math.round(cum * 100) / 100 : null,
-        typical: Math.round((priorAvg * day / daysInMonth) * 100) / 100,
+        typical: Math.round(typical * 100) / 100,
       };
     });
-  }, [monthTxData, analytics, selectedMonth, smYear, smMonth]);
+  }, [dailyData, analytics, selectedMonth]);
 
   if (isLoading || !analytics) {
     return (
@@ -305,16 +333,30 @@ export default function AnalyticsPage() {
   const hasIncome = months.some((m) => (analytics.totals[m]?.income || 0) > 0);
   const hasInvestments = months.some((m) => (analytics.totals[m]?.investments || 0) > 0);
 
-  // Hero delta vs the average of the other months that have data
+  // Hero delta vs the average of the other *complete* months that have data.
-  const otherMonths = months.filter((m) => m !== selectedMonth && (analytics.totals[m]?.spent || 0) > 0);
+  //
+  // Two partial-month traps here. The month in progress never belongs in the
+  // baseline, because most of it has not happened. And when the month in
+  // progress is the one selected, its running total has to be measured against
+  // the same slice of each prior month, not against their full totals.
+  const otherMonths = months.filter(
+    (m) => m !== selectedMonth && m !== currentMonthStr() && (analytics.totals[m]?.spent || 0) > 0
+  );
+  const comparableSpend = selectedIsPartial
+    ? spendThrough(dailyData?.daily?.[selectedMonth], compareDay)
+    : totals.spent;
   const avgSpend = otherMonths.length
-    ? otherMonths.reduce((s, m) => s + (analytics.totals[m]?.spent || 0), 0) / otherMonths.length
+    ? otherMonths.reduce(
+        (s, m) => s + (selectedIsPartial ? spendThrough(dailyData?.daily?.[m], compareDay) : analytics.totals[m]?.spent || 0),
+        0
+      ) / otherMonths.length
     : 0;
-  const avgDeltaPct = avgSpend > 0 ? Math.round(((totals.spent - avgSpend) / avgSpend) * 100) : 0;
+  const avgDeltaPct = avgSpend > 0 ? Math.round(((comparableSpend - avgSpend) / avgSpend) * 100) : 0;
+  const throughQualifier = selectedIsPartial ? ` through day ${compareDay}` : "";
   const heroSentence =
     avgSpend === 0 ? "" :
-    Math.abs(avgDeltaPct) <= 3 ? `in line with your ${otherMonths.length}-month average` :
+    Math.abs(avgDeltaPct) <= 3 ? `in line with your ${otherMonths.length}-month average${throughQualifier}` :
-    `${Math.abs(avgDeltaPct)}% ${avgDeltaPct > 0 ? "above" : "below"} your ${otherMonths.length}-month average of ${fmt(avgSpend)}`;
+    `${Math.abs(avgDeltaPct)}% ${avgDeltaPct > 0 ? "above" : "below"} your ${otherMonths.length}-month average of ${fmt(avgSpend)}${throughQualifier}`;
 
   const pareto80idx = paretoData.findIndex((r) => r.cumulative >= 80);
   const tableMonths = analytics.months.slice(0, 6); // newest-first, last 6
@@ -373,7 +415,10 @@ export default function AnalyticsPage() {
       {movers.length > 0 && (
         <div className="bg-zinc-900 border border-zinc-800 rounded-xl p-4">
           <h3 className="text-sm font-medium mb-1">What changed</h3>
-          <p className="text-xs text-zinc-500 mb-4">Biggest category moves vs {formatShortMonth(prevMonth(selectedMonth))}</p>
+          <p className="text-xs text-zinc-500 mb-4">
+            Biggest category moves vs {formatShortMonth(prevMonth(selectedMonth))}
+            {selectedIsPartial && `, both through day ${compareDay}`}
+          </p>
           <div className="grid sm:grid-cols-2 gap-x-8 gap-y-2.5">
             {movers.map((m) => (
               <div key={m.category} className="flex items-center gap-3">

src/app/insights/page.tsx

@@ -10,6 +10,18 @@ import { CHART, TOOLTIP_STYLE } from "@/lib/category-colors";
 
 const SPEND_TYPES = new Set(["debit", "fee", "interest"]);
 
+/**
+ * The API returns an exclusive upper bound (first day of the month after the
+ * window). Showing that date verbatim would claim a month the totals exclude,
+ * so `exclusive` steps back a day for display.
+ */
+function formatPeriodBound(iso: string | null, exclusive = false): string {
+  if (!iso) return "—";
+  const d = new Date(`${iso}T00:00:00`);
+  if (exclusive) d.setDate(d.getDate() - 1);
+  return d.toLocaleDateString("default", { month: "short", year: "numeric" });
+}
+
 function fmt(n: number) {
   return new Intl.NumberFormat("en-AU", { style: "currency", currency: "AUD", maximumFractionDigits: 0 }).format(n);
 }
@@ -41,10 +53,13 @@ const FREQ_LABEL: Record<string, string> = {
 };
 
 // ─── Section wrapper ────────────────────────────────────────────────
-function Section({ title, children }: { title: string; children: React.ReactNode }) {
+function Section({ title, aside, children }: { title: string; aside?: React.ReactNode; children: React.ReactNode }) {
   return (
     <div className="mb-8">
-      <h3 className="text-base font-semibold text-zinc-200 mb-3">{title}</h3>
+      <div className="flex flex-wrap items-baseline justify-between gap-x-4 gap-y-1 mb-3">
+        <h3 className="text-base font-semibold text-zinc-200">{title}</h3>
+        {aside}
+      </div>
       {children}
     </div>
   );
@@ -298,7 +313,8 @@ export default function InsightsPage() {
   const { data: analytics } = useMonthlyAnalytics(12);
   const { data: analytics6 } = useMonthlyAnalytics(6);
   const { data: subData } = useSubscriptions();
-  const { data: feesData } = useFees();
+  const [feeMonths, setFeeMonths] = useState(12);
+  const { data: feesData } = useFees(feeMonths);
 
   // Build regular/occasional chart data
   const chartData = useMemo(() => {
@@ -431,11 +447,38 @@ export default function InsightsPage() {
       </Section>
 
       {/* ── 4. Fees & Interest ── */}
-      <Section title="Fees & interest">
+      <Section
+        title="Fees & interest"
+        aside={
+          <div className="flex items-center gap-3">
+            {feesData?.period && (
+              <span className="text-xs text-zinc-500 tabular-nums">
+                {feesData.period.all_time
+                  ? "All time"
+                  : `${formatPeriodBound(feesData.period.from)} – ${formatPeriodBound(feesData.period.to, true)}`}
+              </span>
+            )}
+            <select
+              value={feeMonths}
+              onChange={(e) => setFeeMonths(Number(e.target.value))}
+              className="bg-zinc-900 border border-zinc-800 rounded px-2 py-1 text-xs text-zinc-300 focus:outline-none focus:border-indigo-500 cursor-pointer"
+              aria-label="Fees period"
+            >
+              <option value={3}>Last 3 months</option>
+              <option value={6}>Last 6 months</option>
+              <option value={12}>Last 12 months</option>
+              <option value={24}>Last 24 months</option>
+              <option value={0}>All time</option>
+            </select>
+          </div>
+        }
+      >
         {!feesData ? (
           <p className="text-zinc-500 text-sm">Loading...</p>
         ) : feesData.by_bank.length === 0 && feesData.transactions.length === 0 ? (
-          <p className="text-zinc-500 text-sm">No fees or interest recorded across your statements.</p>
+          <p className="text-zinc-500 text-sm">
+            No fees or interest recorded {feesData.period?.all_time ? "on any statement" : "in this period"}.
+          </p>
         ) : (
           <div className="space-y-4">
             {feesData.by_bank.length > 0 && (

src/app/shared/page.tsx

@@ -22,8 +22,12 @@ function formatDate(d: string) {
 
 const SPEND_TYPES = new Set(["debit", "fee", "interest"]);
 
-function formatAmount(n: number, type?: string) {
+function formatAmount(n: number, type?: string, currency?: string) {
-  const formatted = `$${Number(n).toFixed(2)}`;
+  // A bare "$" on a non-AUD row was the visible half of the problem: the row
+  // read as dollars while the participant balances converted to AUD, so the
+  // two disagreed on screen with nothing to explain why.
+  const value = Number(n).toFixed(2);
+  const formatted = !currency || currency === "AUD" ? `$${value}` : `${currency} ${value}`;
   return type && !SPEND_TYPES.has(type) ? `+${formatted}` : formatted;
 }
 
@@ -298,7 +302,17 @@ export default function SharedPage() {
     return <span className="ml-0.5">{sortDir === "desc" ? "↓" : "↑"}</span>;
   }
   const { data: balances = [], isLoading: balLoading } = useParticipantBalances(realTagIds);
+  const { data: allTags = [] } = useTags();
   const { data: me } = useCurrentUser();
+
+  // Names the tag scope when one is active. Non-empty means the cards below are
+  // split totals rather than payable balances.
+  const tagScopeLabel =
+    realTagIds.length === 0
+      ? null
+      : realTagIds.length === 1
+        ? (allTags.find((t) => String(t.id) === realTagIds[0])?.name ?? "this tag")
+        : `${realTagIds.length} tags`;
   const [addingParticipant, setAddingParticipant] = useState(false);
   const [paymentModal, setPaymentModal] = useState<{ id: number; name: string; balance: number } | null>(null);
   const [showHistory, setShowHistory] = useState<number | null>(null);
@@ -351,23 +365,37 @@ export default function SharedPage() {
                   <div>
                     <p className="font-medium">{b.name}</p>
                     <p className="text-xs text-zinc-500">
-                      {settled ? "all square" : theyOweMe ? `owes you` : "you owe"}
+                      {/* With a tag filter on, payments are deliberately not
+                          subtracted — so this is a split total, not a payable
+                          balance, and must not claim to be one. */}
+                      {tagScopeLabel
+                        ? `split total in ${tagScopeLabel}`
+                        : settled ? "all square" : theyOweMe ? "owes you" : "you owe"}
                     </p>
                   </div>
                   <div className="text-right">
-                    <p className={`text-lg font-semibold ${settled ? "text-zinc-500" : theyOweMe ? "text-amber-400" : "text-blue-400"}`}>
+                    <p className={`text-lg font-semibold ${tagScopeLabel ? "text-zinc-300" : settled ? "text-zinc-500" : theyOweMe ? "text-amber-400" : "text-blue-400"}`}>
                       ${net.toFixed(2)}
                     </p>
+                    {b.unconverted_count > 0 && (
+                      <p className="text-[11px] text-amber-500/80 mt-0.5">
+                        approx · {b.unconverted_count} unconverted
+                      </p>
+                    )}
                   </div>
                 </div>
 
                 <div className="flex gap-2">
+                  {/* Settling against a tag-scoped total would record a payment
+                      for a figure that never was the debt. */}
+                  {!tagScopeLabel && (
                     <button
                       onClick={() => setPaymentModal({ id: b.id, name: b.name, balance: b.total_owed })}
                       className="flex-1 py-1.5 text-xs font-medium bg-zinc-800 hover:bg-zinc-700 text-zinc-300 rounded-lg"
                     >
                       Record Payment
                     </button>
+                  )}
                   <button
                     onClick={() => setShowHistory(showHistory === b.id ? null : b.id)}
                     className={`px-3 py-1.5 text-xs rounded-lg ${showHistory === b.id ? "bg-zinc-700 text-white" : "bg-zinc-800 text-zinc-500 hover:text-zinc-300"}`}
@@ -440,7 +468,16 @@ export default function SharedPage() {
                       )}
                     </td>
                     <td className={`px-4 py-3 text-right font-medium tabular-nums ${SPEND_TYPES.has(tx.transaction_type) ? "" : "text-green-400"}`}>
-                      {formatAmount(tx.amount, tx.transaction_type)}
+                      {formatAmount(tx.amount, tx.transaction_type, tx.currency)}
+                      {tx.currency !== "AUD" && (
+                        // Splits settle on the AUD figure, so show it next to the
+                        // native one rather than leaving the two to differ silently.
+                        <span className="block text-xs font-normal text-zinc-500">
+                          {tx.amount_unconverted
+                            ? "AUD value unknown"
+                            : `≈ ${formatAmount(Number(tx.amount_aud), tx.transaction_type, "AUD")} AUD`}
+                        </span>
+                      )}
                     </td>
                     <td className="px-4 py-3">
                       <div className="flex flex-wrap gap-1">

src/app/statements/page.tsx

@@ -206,6 +206,21 @@ export default function StatementsPage() {
                     </td>
                     <td className="px-4 py-3 text-zinc-400 whitespace-nowrap">
                       {formatPeriod(s.billing_start_date, s.billing_end_date)}
+                      {/* An account cannot be billed twice for the same day, so
+                          an overlap means these transactions are in the ledger
+                          twice. Red rather than amber: the balance warning above
+                          means a statement doesn't add up, this means the data is
+                          double-counted everywhere it is summed. */}
+                      {s.overlaps?.length > 0 && (
+                        <div
+                          className="text-[10px] text-red-400 mt-0.5"
+                          title={`Billing period overlaps statement ${s.overlaps
+                            .map((o) => `#${o.id} by ${o.days} day${o.days === 1 ? "" : "s"}`)
+                            .join(", ")}. The overlapping transactions are likely imported twice.`}
+                        >
+                          ⚠ overlaps #{s.overlaps.map((o) => o.id).join(", #")}
+                        </div>
+                      )}
                     </td>
                     <td className="px-4 py-3 text-zinc-400 whitespace-nowrap">
                       {formatDate(s.payment_due_date ?? s.billing_end_date)}

src/app/transactions/page.tsx

@@ -1,6 +1,6 @@
 "use client";
 
-import { useState, useCallback, useRef, useEffect, Suspense } from "react";
+import { useState, useCallback, useRef, useEffect, Suspense, Fragment } from "react";
 import { useSearchParams } from "next/navigation";
 import { useTransactions, useBanks, useUpdateTransaction, useBulkAction, useTags, useStatement, useCreateRule, useParticipants, useRecordPayment, useCurrentUser, useTrips, useAssignTransactionsToTrip, useRules } from "@/lib/hooks";
 import { CATEGORIES, formatCategory } from "@/lib/categories";
@@ -9,7 +9,8 @@ import { TagPicker } from "@/components/tag-picker";
 import { AddTransactionModal } from "@/components/add-transaction-modal";
 import { EditTransactionModal } from "@/components/edit-transaction-modal";
 import { CsvImportModal } from "@/components/csv-import-modal";
-import type { TransactionRow } from "@/lib/queries";
+import type { TransactionRow, RoutePointRow } from "@/lib/queries";
+import { OrderDetails } from "@/components/order-details";
 import type { RuleRow } from "@/lib/hooks";
 
 function formatDate(d: string) {
@@ -475,6 +476,25 @@ function MultiSelect({
   );
 }
 
+/**
+ * "Melbourne Airport (MEL) → Wyndham Vale VIC 3024" from the two stops on an
+ * Uber *trip* receipt. Deliveries are excluded by the caller: their merchant
+ * already identifies them, so the restaurant's street address would be clutter
+ * on every food order.
+ *
+ * Keeps the first two comma-segments of each address — a truncation, not a
+ * guess about geography. The venue or street comes first in Uber's format and
+ * is the identifying part; the full text stays in the title attribute.
+ */
+function routeSummary(route: RoutePointRow[] | null | undefined): string | null {
+  if (!route || route.length < 2) return null;
+  const short = (a: string) => a.split(",").slice(0, 2).join(",").trim();
+  const from = short(route[0].address);
+  const to = short(route[route.length - 1].address);
+  if (!from || !to) return null;
+  return `${from} → ${to}`;
+}
+
 export default function TransactionsPage() {
   return (
     <Suspense fallback={<p className="text-zinc-500 text-sm">Loading...</p>}>
@@ -537,6 +557,9 @@ function TransactionsContent() {
   const [splitModal, setSplitModal] = useState<{ transactionId?: number; transactionIds?: number[]; amount?: number; description: string; merchant?: string } | null>(null);
   const [addModal, setAddModal] = useState<{ prefill?: Parameters<typeof AddTransactionModal>[0]["prefill"]; title?: string } | null>(null);
   const [editModal, setEditModal] = useState<TransactionRow | null>(null);
+  // Rows expanded to show the ingested receipt. A set, not a single id: the
+  // point is comparing several orders without losing your place.
+  const [expanded, setExpanded] = useState<Set<number>>(new Set());
   const [showImportModal, setShowImportModal] = useState(false);
   const [paymentModal, setPaymentModal] = useState<TransactionRow | null>(null);
   const [rulePrompt, setRulePrompt] = useState<{
@@ -911,8 +934,8 @@ function TransactionsContent() {
               <tr><td colSpan={11} className="p-8 text-center text-zinc-500">No transactions found</td></tr>
             ) : (
               data.data.map((t) => (
+                <Fragment key={t.id}>
                 <tr
-                  key={t.id}
                   className={`border-b border-zinc-800/50 hover:bg-zinc-900/30 ${
                     selected.has(t.id) ? "bg-zinc-800/40" : ""
                   }`}
@@ -928,9 +951,39 @@ function TransactionsContent() {
                   <td className={`p-2 whitespace-nowrap sticky left-8 z-10 border-r border-zinc-800/80 ${selected.has(t.id) ? "bg-zinc-800" : "bg-zinc-950"}`}>{formatDate(t.transaction_date)}</td>
                   <td className="p-2 whitespace-nowrap text-zinc-500 text-xs">{formatDate(t.created_at)}</td>
                   <td className="p-2 max-w-xs">
+                    <div className="flex items-start gap-1.5">
+                      {t.order_platform && (
+                        // Only where there IS a receipt behind the row. A
+                        // disclosure arrow on every transaction would promise
+                        // detail that mostly does not exist.
+                        <button
+                          onClick={() => setExpanded((prev) => {
+                            const next = new Set(prev);
+                            if (next.has(t.id)) next.delete(t.id); else next.add(t.id);
+                            return next;
+                          })}
+                          className="text-zinc-600 hover:text-zinc-300 leading-none mt-0.5 shrink-0"
+                          title={expanded.has(t.id) ? "Hide receipt" : "Show the receipt this came from"}
+                          aria-expanded={expanded.has(t.id)}
+                        >
+                          {expanded.has(t.id) ? "▾" : "▸"}
+                        </button>
+                      )}
                       <p className="truncate" title={t.description}>{t.description}</p>
-                    {t.notes && (
+                    </div>
+                    {t.notes ? (
                       <p className="truncate text-xs text-zinc-500 italic mt-0.5" title={t.notes}>{t.notes}</p>
+                    ) : t.order_platform === "uber" && routeSummary(t.order_route) && (
+                      // Five rows all reading "Order - Uber Trip" are
+                      // indistinguishable. Where the trip went is what tells
+                      // them apart, and it was already stored. A note the user
+                      // wrote always wins — this only fills an empty line.
+                      <p
+                        className="truncate text-xs text-zinc-500 italic mt-0.5"
+                        title={t.order_route!.map((r) => `${r.label}${r.time ? ` ${r.time}` : ""}: ${r.address}`).join("\n")}
+                      >
+                        {routeSummary(t.order_route)}
+                      </p>
                     )}
                   </td>
                   <td className="p-2 max-w-[150px]">
@@ -1056,6 +1109,15 @@ function TransactionsContent() {
                     </button>
                   </td>
                 </tr>
+                {expanded.has(t.id) && (
+                  <tr className="border-b border-zinc-800/50 bg-zinc-900/40">
+                    <td />
+                    <td colSpan={10} className="px-4 py-3">
+                      <OrderDetails transactionId={t.id} currency={t.currency ?? null} bare />
+                    </td>
+                  </tr>
+                )}
+                </Fragment>
               ))
             )}
           </tbody>

src/app/trips/[id]/page.tsx

@@ -128,7 +128,10 @@ export default function TripDetailPage({ params }: { params: Promise<{ id: strin
 
       {/* Stat cards */}
       <div className="grid grid-cols-2 sm:grid-cols-4 gap-4">
-        <StatCard label="Total Spend" value={`$${Number(total_spend).toFixed(2)}`} sub="all transactions" color={t.color} />
+        {/* Deliberately every payer, not just this owner — a trip cost what the
+            group put into it. The split figures below are owner-scoped, so this
+            says whose money it counts to stop the two being read as one lens. */}
+        <StatCard label="Total Spend" value={`$${Number(total_spend).toFixed(2)}`} sub="all payers, net of refunds" color={t.color} />
         <StatCard label="Transactions" value={String(transaction_count)} sub="total" color={t.color} />
         <StatCard label="Daily Average" value={`$${Number(daily_average).toFixed(2)}`} sub="per day" color={t.color} />
         <StatCard label="Days" value={String(num_days)} sub={dateRange ?? "date range"} color={t.color} />
@@ -277,7 +280,7 @@ export default function TripDetailPage({ params }: { params: Promise<{ id: strin
               <table className="w-full text-sm">
                 <thead>
                   <tr className="border-b border-zinc-800">
-                    {["Person", "Share of this trip"].map((h) => (
+                    {["Person", "Outstanding on this trip"].map((h) => (
                       <th
                         key={h}
                         className={`px-5 py-2.5 text-xs text-zinc-500 font-medium ${h === "Person" ? "text-left" : "text-right"}`}
@@ -288,23 +291,46 @@ export default function TripDetailPage({ params }: { params: Promise<{ id: strin
                   </tr>
                 </thead>
                 <tbody>
-                  {participant_splits.map((p) => (
+                  {/* A negative outstanding means they have paid more towards this
+                      trip than their share of it — which reads as a typo unless the
+                      sign is spelled out. Shown as a magnitude plus a word, the same
+                      way Shared does it, so the two pages agree on what a direction
+                      means. */}
+                  {participant_splits.map((p) => {
+                    const owed = Number(p.owed);
+                    const square = Math.abs(owed) < 0.005;
+                    const theyOweMe = owed > 0;
+                    return (
                       <tr key={p.participant_id} className="border-b border-zinc-800/50 last:border-0">
                         <td className="px-5 py-3 font-medium">{p.name}</td>
-                      <td className="px-5 py-3 text-right tabular-nums font-mono">${Number(p.owed).toFixed(2)}</td>
+                        <td className="px-5 py-3 text-right tabular-nums font-mono">
+                          <span className={square ? "text-zinc-500" : theyOweMe ? "text-amber-400" : "text-blue-400"}>
+                            ${Math.abs(owed).toFixed(2)}
+                          </span>
+                          <span className="block text-[11px] text-zinc-500 mt-0.5 font-sans">
+                            {square ? "all square" : theyOweMe ? "owes you" : "ahead — you owe them"}
+                          </span>
+                          {p.unconverted_count > 0 && (
+                            <span className="block text-[11px] text-amber-500/80 mt-0.5 font-sans">
+                              approx · {p.unconverted_count} unconverted
+                            </span>
+                          )}
+                        </td>
                       </tr>
-                  ))}
+                    );
+                  })}
                 </tbody>
               </table>
-              {/* Settled/unsettled was reported from transaction_splits.settled,
+              {/* This note used to say a per-trip figure could not be computed,
-                  which nothing sets — so every trip showed 100% unsettled forever,
+                  because payments carried no trip attribution. Migration 0022 added
-                  including ones already paid in full. Settlement is tracked across
+                  split_payments.trip_id, so it can and now does — the figures above
-                  the whole relationship, not per trip: payments carry no trip
+                  are net of payments scoped to this trip. What the note has to say
-                  attribution, so a per-trip figure cannot be computed. */}
+                  instead is which payments are NOT in them. */}
               <p className="px-5 py-2.5 text-xs text-zinc-500 border-t border-zinc-800">
-                Settlement is tracked across all shared expenses, not per trip —
+                Net of payments recorded against this trip. Payments on the ongoing
+                household tab are not counted here —
                 see <Link href="/shared" className="text-zinc-400 hover:text-zinc-200 underline">Shared</Link> for
-                what is actually owed.
+                the overall balance.
               </p>
             </div>
           )}

src/components/edit-transaction-modal.tsx

@@ -10,6 +10,7 @@ import {
   useTrips,
 } from "@/lib/hooks";
 import { SplitModal } from "./split-modal";
+import { OrderDetails } from "./order-details";
 import { CATEGORIES, formatCategory } from "@/lib/categories";
 import type { TransactionRow, TagRow } from "@/lib/queries";
 
@@ -314,6 +315,8 @@ export function EditTransactionModal({
               )}
             </div>
 
+            <OrderDetails transactionId={transaction.id} currency={transaction.currency ?? null} />
+
           </div>
 
           {/* Footer */}

src/components/order-details.tsx

@@ -0,0 +1,95 @@
+"use client";
+
+import { useOrderReceipt, type OrderReceipt } from "@/lib/hooks";
+
+const PLATFORM_LABEL: Record<string, string> = {
+  doordash: "DoorDash",
+  ubereats: "Uber Eats",
+  uber: "Uber",
+};
+
+/**
+ * The receipt behind a delivery order: what was actually bought, and where it
+ * went. All of it was already stored at ingest and none of it was reachable —
+ * the row showed a merchant and a total and nothing else.
+ *
+ * Read-only on purpose. This is what a provider sent, not something to edit.
+ */
+export function OrderDetails({
+  transactionId,
+  currency,
+  bare = false,
+}: {
+  transactionId: number;
+  currency: string | null;
+  /** Drop the top border and heading spacing when embedded in a table row. */
+  bare?: boolean;
+}) {
+  const { data: receipt, isLoading } = useOrderReceipt(transactionId);
+  if (isLoading || !receipt) return null;
+
+  const cur = receipt.currency ?? currency ?? "AUD";
+  const fmt = (n: number) => (cur === "AUD" ? `$${n.toFixed(2)}` : `${cur} ${n.toFixed(2)}`);
+  const items: OrderReceipt["line_items"] = receipt.line_items ?? [];
+  const route: OrderReceipt["route"] = receipt.route ?? [];
+
+  return (
+    <div className={bare ? "" : "border-t border-zinc-800 pt-4"}>
+      <div className="flex items-baseline justify-between mb-2">
+        <p className="text-xs text-zinc-500">
+          Order details
+          {receipt.platform && (
+            <span className="ml-1.5 text-zinc-400">{PLATFORM_LABEL[receipt.platform] ?? receipt.platform}</span>
+          )}
+        </p>
+        {receipt.card_last4 && (
+          <span className="text-xs text-zinc-600">card ••••{receipt.card_last4}</span>
+        )}
+      </div>
+
+      {items.length > 0 ? (
+        <ul className="space-y-1.5 mb-3">
+          {items.map((it, i) => (
+            <li key={i} className="flex gap-2 text-xs">
+              <span className="text-zinc-600 tabular-nums shrink-0">{it.qty}×</span>
+              <span className="text-zinc-300 flex-1 min-w-0">
+                {it.description}
+                {it.options && it.options.length > 0 && (
+                  <span className="block text-zinc-600">{it.options.join(" · ")}</span>
+                )}
+              </span>
+              <span className="text-zinc-400 tabular-nums shrink-0">{fmt(Number(it.amount))}</span>
+            </li>
+          ))}
+        </ul>
+      ) : (
+        // Uber itemises groceries but not restaurant orders, and orders taken
+        // before this was parsed have none either. Say which, rather than
+        // showing an empty list that reads like a bug.
+        <p className="text-xs text-zinc-600 italic mb-3">
+          No itemised list on this receipt
+        </p>
+      )}
+
+      {route.length > 0 && (
+        <div className="space-y-1">
+          {route.map((pt, i) => (
+            <div key={i} className="flex gap-2 text-xs">
+              <span className="text-zinc-600 shrink-0 w-24">
+                {pt.label}
+                {pt.time && <span className="block text-zinc-700">{pt.time}</span>}
+              </span>
+              <span className="text-zinc-400 flex-1">{pt.address}</span>
+            </div>
+          ))}
+        </div>
+      )}
+
+      {receipt.order_reference && !receipt.order_reference.startsWith("msg:") && (
+        <p className="mt-3 text-[11px] text-zinc-700 font-mono break-all">
+          {receipt.order_reference}
+        </p>
+      )}
+    </div>
+  );
+}

src/lib/analytics-sql.ts

@@ -1,7 +1,7 @@
 // Shared SQL fragments for analytics queries, so spend/income semantics stay
 // identical across routes.
 //
-// Two rules every analytics query must follow:
+// Three rules every analytics query must follow:
 //
 // 1. Join `statements` with LEFT JOIN and scope on COALESCE(t.owner_id, s.owner_id).
 //    An INNER JOIN silently drops every manual/CSV transaction (statement_id IS
@@ -11,6 +11,23 @@
 //    bank account and again as the underlying purchases on the card statement.
 //    Categorising the money movement as `transfers` and excluding it here is what
 //    stops the double count. Investments are a balance-sheet move, not spend.
+// 3. Apply EXCLUDE_RECONCILED_SOURCE to every row-level query. The transaction
+//    queries have always done this (`queries.ts`); analytics never did, which is
+//    the other half of the same double count.
+
+/**
+ * `YYYY-MM-DD` for a Date, read in local time.
+ *
+ * `toISOString().slice(0, 10)` is the obvious thing and it is wrong here: these
+ * are calendar boundaries built with `new Date(y, m, 1)`, which is local
+ * midnight. In any timezone east of UTC that converts to the *previous* day, so
+ * every window silently started and ended a day early — visible once the fees
+ * endpoint began reporting the range it had used ("2026-04-30" for a window
+ * meant to open on 1 May).
+ */
+export function toDateStr(d: Date): string {
+  return `${d.getFullYear()}-${String(d.getMonth() + 1).padStart(2, "0")}-${String(d.getDate()).padStart(2, "0")}`;
+}
 
 /** Owner scoping that works for both statement-linked and manual transactions. */
 export const OWNER_SCOPE = `COALESCE(t.owner_id, s.owner_id)`;
@@ -18,6 +35,76 @@ export const OWNER_SCOPE = `COALESCE(t.owner_id, s.owner_id)`;
 /** Join clause to pair with OWNER_SCOPE. */
 export const STATEMENTS_JOIN = `LEFT JOIN statements s ON s.id = t.statement_id`;
 
+/**
+ * Drops rows that a different row has replaced. Two distinct cases:
+ *
+ * **1. The manual/CSV row a statement line superseded** (`reconciled_with_id`).
+ *
+ * Reconciliation keeps both rows: the manual one the user entered and the
+ * statement line it turned out to be. Only the statement line should count, or
+ * the same purchase is spent twice. `queries.ts` has always applied this; the
+ * analytics routes did not, so every reconciled row was double-counted in the
+ * category totals, movers, Pareto, and merchant rankings.
+ *
+ * Scoped to `statement_id IS NULL` deliberately: the *source* row is the manual
+ * one. A statement line pointing at something else is the survivor, not the
+ * duplicate.
+ *
+ * Order-receipt rows (`payment_method = 'credits'`) are unaffected — they are
+ * inserted with `reconciled_with_id` NULL and are held out of the reconcile
+ * queue by `needsCardMatch()`, so nothing ever sets it. If one is reconciled by
+ * hand against a card line, this is what stops it double-counting.
+ *
+ * **2. The statement row imported twice** (`superseded_by_id`, migration 0023).
+ *
+ * When two statements for one account bill overlapping periods, every
+ * transaction in the overlap arrives twice — 31 pairs on ANZ `4085-56264`, from
+ * statements 107/142/143. Case 1 cannot express this: its predicate is scoped
+ * to `statement_id IS NULL` on purpose, and here BOTH rows are statement lines.
+ *
+ * The superseded row is excluded rather than deleted because every child of
+ * `transactions` cascades on delete, and the curation is not reliably on the
+ * surviving side.
+ *
+ * Adding it to this fragment rather than making a new one is deliberate: every
+ * query that already asks "count each purchase once" now excludes both kinds
+ * without being edited. Anything summing transactions without this fragment
+ * still double-counts — that is the same gap that let reconciled rows into the
+ * analytics routes in the first place.
+ */
+export const EXCLUDE_RECONCILED_SOURCE = `NOT (t.statement_id IS NULL AND t.reconciled_with_id IS NOT NULL)
+  AND t.superseded_by_id IS NULL`;
+
+/**
+ * The currency `t.amount` is actually denominated in.
+ *
+ * Two different conventions meet here and the COALESCE order is what keeps them
+ * apart:
+ *
+ *  - A statement row is denominated in its statement's currency. If that row is
+ *    an overseas purchase on an AUD statement, `amount` is still AUD and
+ *    `foreign_currency_code` merely records what was originally charged — so
+ *    `s.currency` must win.
+ *  - An order-receipt row has no statement. There, `amount` IS the native
+ *    figure and `foreign_currency_code` names it (order-ingestion.ts leaves
+ *    `amount_aud` NULL rather than asserting an FX rate it does not have).
+ *
+ * Reading `s.currency` alone labels every foreign order row as AUD.
+ * `s` must be the statements alias in scope.
+ */
+export const NATIVE_CURRENCY = `COALESCE(s.currency, t.foreign_currency_code, 'AUD')`;
+
+/**
+ * True when a row's AUD value is unknown: it is denominated in something other
+ * than AUD and carries no converted figure.
+ *
+ * Every settlement total uses `COALESCE(t.amount_aud, t.amount)`, which for such
+ * a row silently nets a foreign figure against AUD ones. Rather than drop the
+ * row (which changes a balance with no trace) or convert it (with no rate),
+ * count these and let the UI say the balance is incomplete.
+ */
+export const AMOUNT_UNCONVERTED = `(t.amount_aud IS NULL AND ${NATIVE_CURRENCY} <> 'AUD')`;
+
 /** Transaction types that represent money going out. */
 export const SPEND_TYPES = `('debit', 'fee', 'interest')`;
 
@@ -104,3 +191,62 @@ export const SPEND_SIGNED = `CASE
   WHEN t.transaction_type IN ('refund', 'credit') THEN -(${SPEND_BASE})
   ELSE (${SPEND_BASE})
 END`;
+
+/**
+ * A split that still counts towards what someone owes.
+ *
+ * This is the line between the two questions the same table answers, and
+ * conflating them is what made three different "owed" figures disagree:
+ *
+ *  - **Owed** — what is still outstanding between two people. Must apply this.
+ *  - **Spend** — what a purchase cost me. Must NOT apply this.
+ *
+ * A settled split is still a real expense: my half of a 2025 grocery shop is my
+ * spend whether or not the other half was ever repaid. Filtering settled rows
+ * out of `myShare`/`mySplitOf` would re-inflate exactly the figures that
+ * importing settled history exists to correct.
+ *
+ * `settled` marks obligations discharged OUTSIDE this app — imported
+ * SplitMyExpenses history, whose repayments happened on a platform we no longer
+ * run and which therefore has no `split_payments` row here.
+ *
+ * A live obligation is NOT settled by flipping this. It is settled by recording
+ * the payment, and the balance nets to zero on its own. Doing both would
+ * subtract the settlement twice — the splits leave the sum AND the payment is
+ * deducted — driving the balance negative by the amount repaid. So there is
+ * deliberately no "mark settled" action anywhere: settling up is recording a
+ * payment, and this column is only ever written by the historical import.
+ *
+ * **The date is the real guard, and the flag is only a refinement of it.**
+ * Nothing before SPLIT_CUTOVER can be owed, because carryover transaction 2348
+ * already carries the entire pre-cutover balance as a single figure. A split on
+ * an older transaction is therefore free to describe *how an expense was shared*
+ * — which is what stops it inflating spend — without ever asserting a debt.
+ *
+ * That separation is what makes splitting history safe. Before it, the only
+ * thing keeping $37,233.28 of paid debt out of the balances was a boolean that
+ * any delete-and-recreate write path silently reset to false. Now losing the
+ * flag on a pre-cutover row costs nothing: the date still excludes it. The flag
+ * matters only for rows on or after the cutover, where it marks the handful
+ * settled outside this app.
+ *
+ * The boundary is inclusive because transaction 2348 is itself dated
+ * 2026-01-09 — an exclusive bound would drop the carryover and with it the
+ * entire pre-cutover balance.
+ *
+ * Assumes the `transaction_splits` alias is `ts` and `transactions` is `t`.
+ */
+export const SPLIT_CUTOVER = "2026-01-09";
+
+export const ACTIVE_OBLIGATION = `ts.settled = false
+  AND t.transaction_date >= '${SPLIT_CUTOVER}'`;
+
+/**
+ * The tab a split belongs to: its transaction's trip, else the household.
+ *
+ * Membership already lives on `transaction_overrides.trip_id`, so this is a
+ * read of existing data rather than a new grouping key. Assumes an
+ * `transaction_overrides` alias `o` is joined (LEFT — a transaction with no
+ * override row has no trip, which is the common case and means household).
+ */
+export const SPLIT_SCOPE = `o.trip_id`;

src/lib/hooks.ts

@@ -1,7 +1,7 @@
 "use client";
 
 import { useQuery, useMutation, useQueryClient } from "@tanstack/react-query";
-import type { TransactionRow, StatementRow, TagRow, TripRow, TripAnalytics } from "./queries";
+import type { TransactionRow, StatementRow, TagRow, TripRow, TripAnalytics, ParticipantBalance } from "./queries";
 export type { TripRow, TripAnalytics };
 import type { CurrentUser } from "./auth";
 
@@ -216,7 +216,7 @@ export function useParticipants() {
 }
 
 export function useParticipantBalances(tagIds?: string[]) {
-  return useQuery<{ id: number; name: string; total_owed: number; unsettled_count: number }[]>({
+  return useQuery<ParticipantBalance[]>({
     queryKey: ["participant-balances", tagIds],
     queryFn: async () => {
       const params = tagIds?.length ? `?tag_ids=${tagIds.join(",")}` : "";
@@ -250,6 +250,33 @@ export function useTransactionSplits(transactionId: number) {
   });
 }
 
+export interface OrderReceipt {
+  platform: "doordash" | "ubereats" | "uber" | null;
+  order_reference: string | null;
+  line_items: { qty: number; description: string; amount: number; options?: string[] }[];
+  route: { label: string; time: string | null; address: string }[];
+  subtotal: string | null;
+  amount: string | null;
+  currency: string | null;
+  card_last4: string | null;
+  flags: string[];
+  source_email_subject: string | null;
+  transaction_date: string | null;
+}
+
+/** The receipt behind a transaction, or null when it did not come from one. */
+export function useOrderReceipt(transactionId: number) {
+  return useQuery<OrderReceipt | null>({
+    queryKey: ["order-receipt", transactionId],
+    queryFn: async () => {
+      const res = await fetch(`/api/transactions/${transactionId}/order`);
+      if (!res.ok) return null;
+      return res.json();
+    },
+    staleTime: Infinity, // a receipt never changes
+  });
+}
+
 export function useSetSplits() {
   const qc = useQueryClient();
   return useMutation({
@@ -748,6 +775,27 @@ export function useMonthlyAnalytics(months?: number) {
   });
 }
 
+/**
+ * Sparse by day-of-month; a missing day means zero.
+ *   daily      → { "2026-07": { 3: 42.10 } }
+ *   byCategory → { "2026-07": { dining: { 3: 42.10 } } }
+ */
+export interface DailySpend {
+  daily: Record<string, Record<number, number>>;
+  byCategory: Record<string, Record<string, Record<number, number>>>;
+}
+
+export function useDailySpend(months?: number) {
+  const m = months || 12;
+  return useQuery<DailySpend>({
+    queryKey: ["analytics", "daily", m],
+    queryFn: async () => {
+      const res = await fetch(`/api/analytics/daily?months=${m}`);
+      return res.json();
+    },
+  });
+}
+
 export interface SubscriptionRow {
   merchant: string;
   category: string;
@@ -788,16 +836,25 @@ export interface FeeTxnRow {
   bank_name: string;
 }
 
-export function useFees() {
+export interface FeePeriod {
+  months: number;
+  from: string | null;
+  to: string | null;
+  all_time: boolean;
+}
+
+/** `months = 0` means all time. */
+export function useFees(months = 12) {
   return useQuery<{
     by_bank: FeeBankRow[];
     transactions: FeeTxnRow[];
     total_fees: number;
     total_interest: number;
+    period: FeePeriod;
   }>({
-    queryKey: ["analytics", "fees"],
+    queryKey: ["analytics", "fees", months],
     queryFn: async () => {
-      const res = await fetch("/api/analytics/fees");
+      const res = await fetch(`/api/analytics/fees?months=${months}`);
       return res.json();
     },
   });

src/lib/order-ingestion.ts

@@ -5,6 +5,36 @@ export * from "./order-parse";
 
 export const CUTOVER_DATE = "2026-01-09";
 
+/**
+ * Owner for ingested orders. Analytics scope on COALESCE(t.owner_id,
+ * s.owner_id); an ingested order has no statement, so leaving owner_id NULL
+ * hides it from every view in the app while it sits in the table.
+ */
+export const DEFAULT_OWNER_ID = 1;
+
+/** Human labels for the platform a receipt came from. */
+export const PLATFORM_LABEL: Record<ParsedOrder["platform"], string> = {
+  doordash: "DoorDash",
+  ubereats: "Uber Eats",
+  uber: "Uber",
+};
+
+/**
+ * Transaction description.
+ *
+ * The merchant alone ("Order - Burger Corner") does not say where to go and
+ * look for the detail, and there are restaurants on both platforms. The
+ * platform is the one thing the parser always knows and used to discard.
+ */
+export function orderDescription(platform: ParsedOrder["platform"], merchant: string): string {
+  const label = PLATFORM_LABEL[platform];
+  // A trip's merchant is literally "Uber Trip", so the suffix would restate it
+  // — "Order - Uber Trip (Uber)". The addresses that actually identify a trip
+  // are in the Order details panel, not squeezed into the description.
+  if (merchant.toLowerCase().includes(label.toLowerCase())) return `Order - ${merchant}`;
+  return `Order - ${merchant} (${label})`;
+}
+
 export interface IngestResult {
   transactionId: number | null;
   metadataId: number | null;
@@ -86,7 +116,13 @@ async function ensureTag(name: string): Promise<number> {
  */
 export async function processOrderIngestion(
   order: ParsedOrder,
-  options: { messageId?: string; backfillMode?: boolean } = {}
+  options: {
+    messageId?: string;
+    backfillMode?: boolean;
+    ownerId?: number;
+    subject?: string;
+    sender?: string;
+  } = {}
 ): Promise<IngestResult> {
   const flags = [...order.flags];
   const day = order.order_datetime.slice(0, 10);
@@ -108,7 +144,6 @@ export async function processOrderIngestion(
 
   // ---- resolve the credits portion ----------------------------------------
   let creditsAmount: number | null = null;
-  let cardAmount: number | null = order.payment.card_amount;
 
   if (order.payment.ambiguous) {
     const { cardAmount: reconciled } = await reconcileCardLeg(order);
@@ -122,9 +157,7 @@ export async function processOrderIngestion(
       // resolves it once the statement lands. Backfill hits the same path and
       // resolves immediately, because those statements are already imported.
       flags.push("awaiting_card_statement");
-      cardAmount = null;
     } else {
-      cardAmount = reconciled;
       const remainder = Number((order.totals.total_charged - reconciled).toFixed(2));
       if (remainder > 0.02) {
         creditsAmount = remainder;
@@ -151,11 +184,11 @@ export async function processOrderIngestion(
          transaction_date, description, amount, amount_aud, category, payment_method,
          merchant_name, merchant_normalized, transaction_type,
          foreign_currency_amount, foreign_currency_code, owner_id
-       ) VALUES ($1,$2,$3,$4,$5,'credits',$6,$6,'debit',$7,$8,NULL)
+       ) VALUES ($1,$2,$3,$4,$5,'credits',$6,$6,'debit',$7,$8,$9)
        RETURNING id`,
       [
         day,
-        `Order - ${order.merchant_name}`,
+        orderDescription(order.platform, order.merchant_name),
         creditsAmount,
         // No FX rate is available at ingest, so amount_aud is left NULL for
         // foreign orders rather than asserting a conversion we cannot make.
@@ -164,6 +197,11 @@ export async function processOrderIngestion(
         order.merchant_name,
         isAud ? null : creditsAmount,
         isAud ? null : order.currency,
+        // Owner scoping is COALESCE(t.owner_id, s.owner_id). These rows carry
+        // no statement, so a NULL owner_id makes them invisible in every view
+        // in the app — present in the table, absent from the UI. The backfill
+        // inserted 85 rows nobody could see.
+        options.ownerId ?? DEFAULT_OWNER_ID,
       ]
     );
     transactionId = txn!.id;
@@ -185,8 +223,9 @@ export async function processOrderIngestion(
     `INSERT INTO expense_metadata (
        transaction_id, source, source_message_id, order_reference, line_items,
        subtotal, amount, merchant_normalized, transaction_date,
-       card_last4, currency, flags, reconciled_at
+       card_last4, currency, flags, reconciled_at,
-     ) VALUES ($1,'email',$2,$3,$4::jsonb,$5,$6,$7,$8,$9,$10,$11::jsonb,$12)
+       platform, source_email_subject, source_email_from, route
+     ) VALUES ($1,'email',$2,$3,$4::jsonb,$5,$6,$7,$8,$9,$10,$11::jsonb,$12,$13,$14,$15,$16::jsonb)
      RETURNING id`,
     [
       transactionId,
@@ -201,6 +240,10 @@ export async function processOrderIngestion(
       order.currency,
       JSON.stringify(flags),
       pending ? null : new Date().toISOString(),
+      order.platform,
+      options.subject ?? null,
+      options.sender ?? null,
+      JSON.stringify(order.route ?? []),
     ]
   );
 
@@ -230,9 +273,10 @@ export async function reconcilePendingOrders(): Promise<{
     merchant_normalized: string;
     card_last4: string | null;
     currency: string | null;
+    platform: ParsedOrder["platform"] | null;
   }>(
     `SELECT id, order_reference, amount::text, transaction_date::text,
-            merchant_normalized, card_last4, currency
+            merchant_normalized, card_last4, currency, platform
        FROM expense_metadata
       WHERE transaction_id IS NULL
         AND reconciled_at IS NULL
@@ -246,7 +290,7 @@ export async function reconcilePendingOrders(): Promise<{
     const total = Number(row.amount);
     const probe: ParsedOrder = {
       order_reference: row.order_reference,
-      platform: "doordash",
+      platform: row.platform ?? "doordash",
       merchant_name: row.merchant_normalized,
       order_datetime: `${row.transaction_date}T00:00:00Z`,
       currency: row.currency || "AUD",
@@ -256,6 +300,7 @@ export async function reconcilePendingOrders(): Promise<{
         service_fee: null, tip: null, discounts: null, total_charged: total,
       },
       line_items: [],
+      route: [],
       is_family: false,
       flags: [],
     };
@@ -280,9 +325,9 @@ export async function reconcilePendingOrders(): Promise<{
         `INSERT INTO transactions (
            transaction_date, description, amount, amount_aud, category,
            payment_method, merchant_name, merchant_normalized, transaction_type, owner_id
-         ) VALUES ($1,$2,$3,$3,$5,'credits',$4,$4,'debit',NULL)
+         ) VALUES ($1,$2,$3,$3,$5,'credits',$4,$4,'debit',$6)
          RETURNING id`,
-        [row.transaction_date, `Order - ${row.merchant_normalized}`, remainder, row.merchant_normalized, category]
+        [row.transaction_date, orderDescription(row.platform ?? "doordash", row.merchant_normalized), remainder, row.merchant_normalized, category, DEFAULT_OWNER_ID]
       );
       txnId = txn!.id;
       created++;

src/lib/order-parse.ts

@@ -20,6 +20,18 @@ export interface LineItem {
   options?: string[];
 }
 
+/**
+ * A stop on the receipt's map: pick-up, delivery, or (for a trip) the ride's
+ * start and end. Uber prints these for every order under `Order details`.
+ */
+export interface RoutePoint {
+  /** "Pick-up" / "Delivery" — whatever the receipt itself calls it. */
+  label: string;
+  /** Local time as printed, e.g. "1:20 pm". No date; the receipt gives none. */
+  time: string | null;
+  address: string;
+}
+
 export interface PaymentBreakdown {
   credits_amount: number | null;
   card_amount: number | null;
@@ -47,6 +59,8 @@ export interface ParsedOrder {
   payment: PaymentBreakdown;
   totals: OrderTotals;
   line_items: LineItem[];
+  /** Uber only. Empty for DoorDash, whose receipts carry no addresses. */
+  route: RoutePoint[];
   is_family: boolean;
   flags: string[];
 }
@@ -98,15 +112,6 @@ const decodeEntities = (s: string) =>
 
 const collapse = (s: string) => s.replace(/\s+/g, " ").trim();
 
-/** URL-decodes without throwing on malformed percent-escapes. */
-function safeDecode(s: string): string {
-  try {
-    return decodeURIComponent(s.replace(/%(?![0-9a-f]{2})/gi, "%25"));
-  } catch {
-    return s;
-  }
-}
-
 const money = (raw: string): number => Math.abs(parseFloat(raw.replace(/[$,]/g, "")));
 
 /**
@@ -125,6 +130,103 @@ function tdPairValue(html: string, label: string): number | null {
   return m ? money(m[1]) : null;
 }
 
+/**
+ * Uber itemises only *grocery* orders — a restaurant receipt states a total and
+ * nothing else, which is why 67 of the 101 backfilled orders have no items.
+ * When it does itemise, the markup is far better than DoorDash's: every cell
+ * carries a `data-testid` naming its role and the item's own uuid, so quantity,
+ * title and amount can be bound to each other by id rather than by position.
+ */
+function parseUberLineItems(html: string): LineItem[] {
+  const items: LineItem[] = [];
+  const titleRe =
+    /data-testid="shoppingCart_item_title_([0-9a-f-]+)"[^>]*>([\s\S]*?)<\/td>/gi;
+
+  for (const m of html.matchAll(titleRe)) {
+    const [, id, rawTitle] = m;
+    const description = collapse(decodeEntities(stripTags(rawTitle)));
+    if (!description) continue;
+
+    const qtyM = html.match(
+      new RegExp(`data-testid="shoppingCart_item_quantity_${id}"[^>]*>\\s*(\\d+)\\s*<`, "i")
+    );
+    const amtM = html.match(
+      new RegExp(
+        `data-testid="shoppingCart_item_amount_${id}"[^>]*>([\\s\\S]*?)<\\/td>`,
+        "i"
+      )
+    );
+    const amtText = amtM ? collapse(decodeEntities(stripTags(amtM[1]))) : "";
+    const amtNum = amtText.match(/(-?[\d,]+\.\d{2})/);
+
+    items.push({
+      qty: qtyM ? parseInt(qtyM[1], 10) : 1,
+      description,
+      // A sold-out item prints 0.00 and is genuinely part of the order — it
+      // explains a total that does not match what was asked for. Keep it.
+      amount: amtNum ? money(amtNum[1]) : 0,
+    });
+  }
+  return items;
+}
+
+/**
+ * Uber's `Order details` block, anchored on `data-testid="address_point_N_*"`.
+ *
+ * The template repeats the whole block twice (once hidden for narrow screens),
+ * so the same stop appears more than once and has to be de-duplicated. This is
+ * the same markup a *trip* receipt uses for its start and destination — rides
+ * are not ingested today, but the reader will not need changing when they are.
+ */
+function parseUberRoute(html: string): RoutePoint[] {
+  const seen = new Set<string>();
+  const points: RoutePoint[] = [];
+
+  const labelRe = /data-testid="address_point_(\d+)_time"[^>]*>([\s\S]*?)<\/td>/gi;
+  for (const m of html.matchAll(labelRe)) {
+    const [, idx, rawLabel] = m;
+    const addrM = html.match(
+      new RegExp(`data-testid="address_point_${idx}_address"[^>]*>([\\s\\S]*?)<\\/td>`, "i")
+    );
+    if (!addrM) continue;
+
+    const address = collapse(decodeEntities(stripTags(addrM[1])));
+    // Delivery receipts share one cell between time and label — "1:20 pm -
+    // Pick-up". Trip receipts print the time alone, with no label at all, so
+    // the naive split put the time in `label` and left `time` null. Position
+    // carries the meaning there: first stop is where the ride began.
+    const combined = collapse(decodeEntities(stripTags(rawLabel)));
+    const split = combined.match(/^(.*?)\s+-\s+(.*)$/);
+    let time: string | null;
+    let label: string;
+    if (split) {
+      time = split[1];
+      label = split[2];
+    } else if (/^\d{1,2}:\d{2}\s*(am|pm)?$/i.test(combined)) {
+      time = combined;
+      label = ""; // filled in positionally below — the receipt gives none
+    } else {
+      time = null;
+      label = combined;
+    }
+
+    const key = `${label}|${time}|${address}`;
+    if (!address || seen.has(key)) continue;
+    seen.add(key);
+    points.push({ label, time, address });
+  }
+
+  // A trip receipt labels neither end. Position is the only thing that says
+  // which is which, and for a two-stop trip it says it unambiguously. Only
+  // filled where the receipt itself was silent, so a future template that does
+  // label its stops keeps its own wording.
+  if (points.length === 2 && points.every((p) => !p.label)) {
+    points[0].label = "Pick-up";
+    points[1].label = "Drop-off";
+  }
+  return points;
+}
+
 function parseDoorDashLineItems(html: string): LineItem[] {
   // <td width="10%">1x</td><td width="75%"><b>Name</b> (Cat)<br><font>• Opt</font>…</td><td width="15%">$22.10</td>
   const re =
@@ -181,7 +283,6 @@ function parseMerchant(platform: string, meta: MessageMeta, text: string): strin
 }
 
 function parsePayment(platform: string, html: string, text: string): PaymentBreakdown {
-  // eslint-disable-next-line no-param-reassign
   const out: PaymentBreakdown = {
     credits_amount: null,
     card_amount: null,
@@ -226,8 +327,14 @@ function parsePayment(platform: string, html: string, text: string): PaymentBrea
 
   const cash = text.match(/Uber Cash\s*(?:[A-Z]{3})?\s*\$?([\d,]+\.\d{2})/i);
   if (cash) out.credits_amount = money(cash[1]);
+  // Anchor on the masking, not on a list of card brands. Uber labels the card
+  // leg with whatever the issuer is called — "Westpac ••••8032 $15.33",
+  // "Mastercard ••••3893 (CBA Ultimate) CHF 51.23" — so a brand allowlist
+  // silently drops the card half of a mixed payment. Found in the backfill
+  // dry-run: Uber Cash $1.17 + Westpac ••••8032 $15.33 against a $16.50 total,
+  // which validateOrderTotals correctly refused rather than under-recording.
   const card = text.match(
-    /(?:Visa|MasterCard|American Express|Amex)[^\d]*(\d{4})[^\d]*(?:[A-Z]{3})?\s*\$?([\d,]+\.\d{2})/i
+    /(?:••••|\*{4}|\u2022{4})\s*(\d{4})[^\d]{0,40}?\$?\s*([\d,]+\.\d{2})/
   );
   if (card) {
     out.card_last4 = card[1];
@@ -272,6 +379,20 @@ export function parseOrderHTML(html: string, meta: MessageMeta): ParsedOrder {
     );
   }
 
+  // Uber sends TWO mails per trip with the same subject and the same total: a
+  // "charge summary" when the trip ends, then the real receipt once payment
+  // settles. The summary says so itself — "This is not a payment receipt ...
+  // You will receive a trip receipt when the payment is processed with payment
+  // information" — and it carries no tripReference, so order_reference would
+  // fall back to `msg:<message-id>` and I7 could not dedupe it against the
+  // receipt that follows. Every trip would be recorded twice.
+  if (/This is not a payment receipt|This is your charge summary/i.test(text)) {
+    throw new NotAReceiptError(
+      "charge summary, not a payment receipt — the real receipt follows",
+      meta.messageId
+    );
+  }
+
   const platform = detectPlatform(meta, text);
   const merchant_name = parseMerchant(platform, meta, text);
 
@@ -379,14 +500,15 @@ export function parseOrderHTML(html: string, meta: MessageMeta): ParsedOrder {
   const payment = parsePayment(platform, clean, text);
   if (payment.ambiguous && is_family) {
     // [Family] receipts name the payer, not an instrument ("Payments Siddharth
-    // LKR 3,783.20"), so no split is recoverable and there is no card leg to
+    // LKR 3,783.20"). An earlier version read that as credits-funded. It is
-    // reconcile against — parking them would mean never importing them, which
+    // not: the card statement carries all four of them (CBA ...3893, exact
-    // fails the actual requirement (import, tag, exclude from budgets).
+    // foreign_currency_amount matches), so creating a transaction duplicated
-    // Treated as credits so the order is recorded and tagged. Safe because the
+    // spend that was already recorded — precisely the double-count I5 exists to
-    // family tag removes it from every budget regardless of instrument.
+    // prevent.
-    payment.ambiguous = false;
+    //
-    payment.credits_amount = totals.total_charged;
+    // Treated as card-settled: provenance only, no transaction. The statement
-    flags.push("family_payment_assumed_credits");
+    // line IS the transaction, and it is what should carry the `family` tag.
+    flags.push("family_card_settled_no_transaction");
   } else if (payment.ambiguous) {
     // Split not stated and resolvable from the card statement — left for the
     // ingestion runner to reconcile, not guessed here.
@@ -400,11 +522,15 @@ export function parseOrderHTML(html: string, meta: MessageMeta): ParsedOrder {
   // ---- line items ----------------------------------------------------------
   // Uber Eats receipts carry no itemisation (verified across 29 real mails).
   const line_items =
-    platform === "doordash" ? parseDoorDashLineItems(clean) : [];
+    platform === "doordash" ? parseDoorDashLineItems(clean) : parseUberLineItems(clean);
   if (platform === "doordash" && line_items.length === 0) {
     flags.push("no_line_items_parsed");
   }
 
+  // Uber prints addresses on every receipt; DoorDash prints none at all, so an
+  // empty route there is expected rather than a parse failure.
+  const route = platform === "doordash" ? [] : parseUberRoute(clean);
+
   const currency =
     explicitCurrency ||
     (/\b(NZD|USD|LKR|CHF|EUR|GBP|SGD|INR)\b/.test(text)
@@ -420,6 +546,7 @@ export function parseOrderHTML(html: string, meta: MessageMeta): ParsedOrder {
     payment,
     totals,
     line_items,
+    route,
     is_family,
     flags,
   };

src/lib/queries.ts

@@ -1,4 +1,11 @@
 import { queryRaw } from "./db";
+import { EXCLUDE_RECONCILED_SOURCE, NATIVE_CURRENCY, AMOUNT_UNCONVERTED, ACTIVE_OBLIGATION, STATEMENTS_JOIN, OWNER_SCOPE, NET_SPEND_ROWS, SPEND_SIGNED } from "./analytics-sql";
+
+export interface RoutePointRow {
+  label: string;
+  time: string | null;
+  address: string;
+}
 
 export interface TagRow {
   id: number;
@@ -26,8 +33,11 @@ export interface TransactionRow {
   principal_amount: number | null;
   interest_amount: number | null;
   // How it was paid (migration 0016). NULL = unknown, treated as reconcilable.
-  // 'cash' is excluded from reconciliation — see notCash().
+  // 'cash' and 'credits' are excluded from reconciliation — see needsCardMatch().
   payment_method: string | null;
+  /** Uber pick-up/drop-off, when this row came from an order receipt. */
+  order_route: RoutePointRow[] | null;
+  order_platform: "doordash" | "ubereats" | "uber" | null;
   // override fields
   category_override: string | null;
   merchant_override: string | null;
@@ -40,9 +50,13 @@ export interface TransactionRow {
   my_amount: number;
   // statement context (null for manual transactions)
   bank_name: string;
-  // Native currency of the statement this row came from ('AUD' for manual rows).
+  // The currency `amount` is denominated in; `amount_aud` is the converted
-  // `amount` is in this currency; `amount_aud` is the converted figure.
+  // figure where one exists. Usually the statement's currency, but an
+  // order-receipt row has no statement and carries its own — see
+  // NATIVE_CURRENCY. Not simply 'AUD' for every statement-less row.
   currency: string;
+  /** True when `amount` is non-AUD and no converted figure exists. */
+  amount_unconverted: boolean;
   owner_id: number;
   owner_name: string;
   // tags
@@ -91,6 +105,12 @@ export interface StatementRow {
   // opening/closing balance to check against.
   expected_closing: number | null;
   balance_diff: number | null;
+  /**
+   * Other statements for this account billing the same days — see
+   * STATEMENT_OVERLAPS. Non-empty means some of these transactions are almost
+   * certainly imported twice. Empty array, never null.
+   */
+  overlaps: { id: number; days: number }[];
 }
 
 interface TransactionFilters {
@@ -115,7 +135,7 @@ interface TransactionFilters {
 export async function getTransactions(ownerId: number, filters: TransactionFilters) {
   const conditions: string[] = [
     `(COALESCE(t.owner_id, s.owner_id) = $1 OR EXISTS (SELECT 1 FROM transaction_splits ts_me WHERE ts_me.transaction_id = t.id AND ts_me.participant_id = $1))`,
-    `NOT (t.statement_id IS NULL AND t.reconciled_with_id IS NOT NULL)`,
+    EXCLUDE_RECONCILED_SOURCE,
   ];
   const params: unknown[] = [ownerId];
   let paramIdx = 2;
@@ -133,17 +153,24 @@ export async function getTransactions(ownerId: number, filters: TransactionFilte
     params.push(filters.categories);
   }
   if (filters.bank_names?.length) {
+    // "Manual" and "Gift Card" are not banks — they are the two shapes a
+    // statement-less row can take, and bankLabel() decides which. The filter
+    // has to split on the same condition or the chip selects nothing.
     const hasManual = filters.bank_names.includes("Manual");
-    const bankList = filters.bank_names.filter((b) => b !== "Manual");
+    const hasGiftCard = filters.bank_names.includes("Gift Card");
-    if (hasManual && bankList.length > 0) {
+    const bankList = filters.bank_names.filter((b) => b !== "Manual" && b !== "Gift Card");
-      conditions.push(`(t.statement_id IS NULL OR s.bank_name = ANY($${paramIdx++}::text[]))`);
+    const alternatives: string[] = [];
-      params.push(bankList);
+    if (hasManual) {
-    } else if (hasManual) {
+      alternatives.push(`(t.statement_id IS NULL AND t.payment_method IS DISTINCT FROM 'credits')`);
-      conditions.push(`t.statement_id IS NULL`);
+    }
-    } else {
+    if (hasGiftCard) {
-      conditions.push(`s.bank_name = ANY($${paramIdx++}::text[])`);
+      alternatives.push(`(t.statement_id IS NULL AND t.payment_method = 'credits')`);
+    }
+    if (bankList.length > 0) {
+      alternatives.push(`s.bank_name = ANY($${paramIdx++}::text[])`);
       params.push(bankList);
     }
+    conditions.push(`(${alternatives.join(" OR ")})`);
   }
   if (filters.tag_ids?.length) {
     const noTags = filters.tag_ids.includes("untagged");
@@ -210,8 +237,9 @@ export async function getTransactions(ownerId: number, filters: TransactionFilte
       o.category_override, o.merchant_normalized as merchant_override, o.notes, o.my_share_percent,
       COALESCE(o.category_override, t.category) as effective_category,
       COALESCE(o.merchant_normalized, t.merchant_normalized, t.merchant_name) as effective_merchant,
-      COALESCE(s.bank_name, 'Manual') as bank_name,
+      ${bankLabel()} as bank_name,
-      COALESCE(s.currency, 'AUD') as currency,
+      ${NATIVE_CURRENCY} as currency,
+      ${AMOUNT_UNCONVERTED} as amount_unconverted,
       -- My share, resolved the same way analytics does it (see myShare in
       -- analytics-sql.ts): explicit split row, then override, then whatever is
       -- left after everyone else. Computed here so the UI cannot drift from
@@ -237,10 +265,23 @@ export async function getTransactions(ownerId: number, filters: TransactionFilte
       tr.name as trip_name,
       tr.color as trip_color,
       txn_tags.tags,
-      txn_splits.splits
+      txn_splits.splits,
+      order_ctx.route as order_route,
+      order_ctx.platform as order_platform
     FROM transactions t
     LEFT JOIN transaction_overrides o ON o.transaction_id = t.id
     LEFT JOIN statements s ON s.id = t.statement_id
+    -- Order provenance, for the sub-line under the description. Five rows all
+    -- reading "Order - Uber Trip" are indistinguishable; where they went is the
+    -- only thing that tells them apart, and it was already stored.
+    -- Both directions, because a card-settled order has no transaction of its
+    -- own and points at the statement line instead (I5).
+    LEFT JOIN LATERAL (
+      SELECT em.route, em.platform
+        FROM expense_metadata em
+       WHERE em.transaction_id = t.id OR em.matched_transaction_id = t.id
+       LIMIT 1
+    ) order_ctx ON true
     LEFT JOIN participants p ON p.id = COALESCE(t.owner_id, s.owner_id)
     LEFT JOIN transactions src ON src.reconciled_with_id = t.id AND src.statement_id IS NULL
     LEFT JOIN trips tr ON tr.id = o.trip_id
@@ -294,7 +335,7 @@ export async function getTransactionById(id: number) {
       o.category_override, o.merchant_normalized as merchant_override, o.notes, o.my_share_percent,
       COALESCE(o.category_override, t.category) as effective_category,
       COALESCE(o.merchant_normalized, t.merchant_normalized, t.merchant_name) as effective_merchant,
-      COALESCE(s.bank_name, 'Manual') as bank_name,
+      ${bankLabel()} as bank_name,
       COALESCE(t.owner_id, s.owner_id) as owner_id,
       p.name as owner_name
     FROM transactions t
@@ -326,15 +367,56 @@ export const BALANCE_DELTA = `SUM(CASE
     ELSE CASE WHEN t.transaction_type IN ('debit', 'fee', 'interest') THEN -t.amount ELSE t.amount END
 END)`;
 
+/**
+ * Other statements for the same account whose billing period overlaps this one.
+ *
+ * An account cannot be billed twice for the same day, so an overlap means the
+ * same transactions were imported twice. This is not hypothetical: ANZ
+ * statements 107 and 143 overlap by 118 days and put ~$42,000 of duplicate rows
+ * in the ledger, which silently inflated spend and dragged the CSV split match
+ * rate down to 46.5%.
+ *
+ * Two details are what make it actually catch that case:
+ *
+ *  - The account number is compared with non-digits stripped. The duplicate got
+ *    in precisely because the existing duplicate key compared raw text, and ANZ
+ *    wrote the same account as `408556264` on one statement and `4085-56264` on
+ *    the other.
+ *  - The range is half-open `[)`. These statements are issued back-to-back with
+ *    one period's end date equal to the next one's start, so inclusive bounds
+ *    flag every consecutive pair — 5 hits of which 3 were false. Half-open
+ *    leaves exactly the 2 real ones.
+ *
+ * NULL bounds are excluded rather than passed to `daterange`, where NULL means
+ * unbounded and would make an undated statement overlap the entire history.
+ */
+const STATEMENT_OVERLAPS = `
+      SELECT COALESCE(json_agg(json_build_object(
+               'id', o.id,
+               'days', (LEAST(s.billing_end_date, o.billing_end_date)
+                        - GREATEST(s.billing_start_date, o.billing_start_date))
+             ) ORDER BY o.id), '[]'::json) AS overlaps
+      FROM statements o
+      WHERE o.id <> s.id
+        AND o.owner_id = s.owner_id
+        AND regexp_replace(o.account_number, '\\D', '', 'g')
+          = regexp_replace(s.account_number, '\\D', '', 'g')
+        AND o.billing_start_date IS NOT NULL AND o.billing_end_date IS NOT NULL
+        AND s.billing_start_date IS NOT NULL AND s.billing_end_date IS NOT NULL
+        AND daterange(s.billing_start_date, s.billing_end_date, '[)')
+         && daterange(o.billing_start_date, o.billing_end_date, '[)')`;
+
 export async function getStatements(ownerId: number) {
   const sql = `
     SELECT s.*,
       (SELECT COUNT(*)::int FROM transactions t WHERE t.statement_id = s.id) as transaction_count,
       p.name as owner_name,
       recon.expected_closing,
-      recon.balance_diff
+      recon.balance_diff,
+      ov.overlaps
     FROM statements s
     LEFT JOIN participants p ON p.id = s.owner_id
+    LEFT JOIN LATERAL (${STATEMENT_OVERLAPS}) ov ON true
     LEFT JOIN LATERAL (
       SELECT
         (s.opening_balance + ${BALANCE_DELTA})::numeric(12,2) as expected_closing,
@@ -376,12 +458,20 @@ export async function getMerchantSuggestions(search: string) {
 }
 
 export async function getBankNames() {
-  const [bankRows, manualCount] = await Promise.all([
+  const [bankRows, statementless] = await Promise.all([
     queryRaw<{ bank_name: string }>(`SELECT DISTINCT bank_name FROM statements ORDER BY bank_name`),
-    queryRaw<{ count: number }>(`SELECT COUNT(*)::int as count FROM transactions WHERE statement_id IS NULL`),
+    queryRaw<{ label: string }>(
+      `SELECT DISTINCT ${bankLabel("t", "s")} as label
+         FROM transactions t LEFT JOIN statements s ON s.id = t.statement_id
+        WHERE t.statement_id IS NULL`
+    ),
   ]);
   const banks = bankRows.map((r) => r.bank_name);
-  if (manualCount[0]?.count > 0) banks.push("Manual");
+  // Order matters for the filter chips: real banks first, then the
+  // statement-less kinds, in a stable order rather than whatever the DB returns.
+  for (const label of ["Manual", "Gift Card"]) {
+    if (statementless.some((r) => r.label === label)) banks.push(label);
+  }
   return banks;
 }
 
@@ -390,6 +480,8 @@ export interface ParticipantBalance {
   name: string;
   total_owed: number;
   unsettled_count: number;
+  /** Splits counted at a non-AUD figure because no converted amount exists. */
+  unconverted_count: number;
 }
 
 export async function getParticipantBalances(ownerId: number, tagIds?: number[]) {
@@ -403,6 +495,13 @@ export async function getParticipantBalances(ownerId: number, tagIds?: number[])
   // Payments settle the total relationship between two people, not a specific tag.
   // Only subtract payments when viewing the unfiltered total; with a tag filter
   // active, show the raw split amount for that tag context only.
+  //
+  // That asymmetry is a symptom, not a design: a tag is a view and has no
+  // payments, so a tag-filtered balance had nothing honest to subtract. A trip
+  // does have payments (`split_payments.trip_id`, migration 0022), which is why
+  // the per-trip figure in getTripAnalytics can be netted and this one cannot.
+  // The fix for the tag case is to stop showing a balance there, not to invent
+  // one — see docs/shared-expenses-design.md.
   const paymentsJoin = tagIds?.length ? "" : `
     LEFT JOIN (
       SELECT
@@ -419,7 +518,11 @@ export async function getParticipantBalances(ownerId: number, tagIds?: number[])
     SELECT p.id, p.name,
       COALESCE(SUM(splits.signed_amount), 0)::numeric(12,2)
         ${paymentsSelect} AS total_owed,
-      COALESCE(SUM(splits.split_count), 0)::int AS unsettled_count
+      COALESCE(SUM(splits.split_count), 0)::int AS unsettled_count,
+      -- Splits whose AUD value is unknown. They are still summed above (as
+      -- their native figure), so a non-zero count means this balance is
+      -- approximate and the UI has to say so.
+      COALESCE(SUM(splits.unconverted_count), 0)::int AS unconverted_count
     FROM participants p
 
     LEFT JOIN (
@@ -428,12 +531,14 @@ export async function getParticipantBalances(ownerId: number, tagIds?: number[])
       -- currency, so splitting on it nets a USD figure against AUD ones.
       SELECT ts.participant_id AS pid,
         (CASE WHEN t.transaction_type IN ('debit', 'fee', 'interest') THEN COALESCE(t.amount_aud, t.amount) ELSE -COALESCE(t.amount_aud, t.amount) END) * ts.share_percent / 100 AS signed_amount,
-        1 AS split_count
+        1 AS split_count,
+        (CASE WHEN ${AMOUNT_UNCONVERTED} THEN 1 ELSE 0 END) AS unconverted_count
       FROM transaction_splits ts
       JOIN transactions t ON t.id = ts.transaction_id
       LEFT JOIN statements s ON s.id = t.statement_id
       WHERE COALESCE(t.owner_id, s.owner_id) = $1 AND ts.participant_id != $1
-        AND NOT (t.statement_id IS NULL AND t.reconciled_with_id IS NOT NULL)
+        AND ${EXCLUDE_RECONCILED_SOURCE}
+        AND ${ACTIVE_OBLIGATION}
       ${tagFilter}
 
       UNION ALL
@@ -441,12 +546,14 @@ export async function getParticipantBalances(ownerId: number, tagIds?: number[])
       -- I owe them: my splits on transactions they own
       SELECT COALESCE(t.owner_id, s.owner_id) AS pid,
         -((CASE WHEN t.transaction_type IN ('debit', 'fee', 'interest') THEN COALESCE(t.amount_aud, t.amount) ELSE -COALESCE(t.amount_aud, t.amount) END) * ts.share_percent / 100) AS signed_amount,
-        0 AS split_count
+        0 AS split_count,
+        (CASE WHEN ${AMOUNT_UNCONVERTED} THEN 1 ELSE 0 END) AS unconverted_count
       FROM transaction_splits ts
       JOIN transactions t ON t.id = ts.transaction_id
       LEFT JOIN statements s ON s.id = t.statement_id
       WHERE ts.participant_id = $1 AND COALESCE(t.owner_id, s.owner_id) != $1
-        AND NOT (t.statement_id IS NULL AND t.reconciled_with_id IS NOT NULL)
+        AND ${EXCLUDE_RECONCILED_SOURCE}
+        AND ${ACTIVE_OBLIGATION}
       ${tagFilter}
     ) splits ON splits.pid = p.id
     ${paymentsJoin}
@@ -533,8 +640,25 @@ export async function batchInsertCSVTransactions(
  * transaction accounts are imported, and NULL means unknown — both stay
  * candidates, which preserves the behaviour of every pre-existing row.
  */
-export const notCash = (alias = "t") =>
+/**
-  `(${alias}.payment_method IS NULL OR ${alias}.payment_method <> 'cash')`;
+ * Payment methods that can still be matched against a card statement line.
+ *
+ * Cash never appears on one. Neither does a credits-funded delivery order: the
+ * gift card already paid it, so there is no card leg coming, ever. Leaving
+ * those in the queue meant 81 orders sat in "pending reconciliation" waiting
+ * for a match that could not exist (user, 2026-07-27).
+ */
+export const needsCardMatch = (alias = "t") =>
+  `(${alias}.payment_method IS NULL OR ${alias}.payment_method NOT IN ('cash', 'credits'))`;
+
+/**
+ * Bank label for a transaction. A row with no statement was not imported from
+ * one, and the label has to say *why*: "Manual" reads as "hand-entered, still
+ * awaiting a card line", which is wrong for a gift-card order — nothing is
+ * awaited. `s` must be the statements alias in scope.
+ */
+export const bankLabel = (t = "t", s = "s") =>
+  `COALESCE(${s}.bank_name, CASE WHEN ${t}.payment_method = 'credits' THEN 'Gift Card' ELSE 'Manual' END)`;
 
 export interface PotentialMatch {
   id: number;
@@ -578,7 +702,7 @@ export async function getPendingReconciliations(ownerId: number): Promise<Manual
        WHERE ts.transaction_id = t.id
      ) txn_splits ON true
      WHERE t.statement_id IS NULL AND t.owner_id = $1 AND t.reconciled_with_id IS NULL
-       AND ${notCash("t")}
+       AND ${needsCardMatch("t")}
      ORDER BY t.transaction_date DESC, t.row_index ASC`,
     [ownerId]
   );
@@ -620,7 +744,7 @@ export async function getPendingReconciliations(ownerId: number): Promise<Manual
        WHERE m.statement_id IS NULL
          AND m.owner_id = $1
          AND m.reconciled_with_id IS NULL
-         AND ${notCash("m")}
+         AND ${needsCardMatch("m")}
          AND COALESCE(t.owner_id, s.owner_id) = $1
          AND NOT EXISTS (
            SELECT 1 FROM transactions mt WHERE mt.reconciled_with_id = t.id
@@ -687,7 +811,11 @@ export async function getSharedTransactions(ownerId: number, tagIds?: number[],
       o.category_override, o.merchant_normalized as merchant_override, o.notes,
       COALESCE(o.category_override, t.category) as effective_category,
       COALESCE(o.merchant_normalized, t.merchant_normalized, t.merchant_name) as effective_merchant,
-      COALESCE(s.bank_name, 'Manual') as bank_name,
+      ${bankLabel()} as bank_name,
+      -- The table renders t.amount, which is not always AUD. Without these the
+      -- rows visibly disagreed with the participant balances, which do convert.
+      ${NATIVE_CURRENCY} as currency,
+      ${AMOUNT_UNCONVERTED} as amount_unconverted,
       COALESCE(t.owner_id, s.owner_id) as owner_id,
       p_owner.name as owner_name,
       COALESCE(src.created_at, t.created_at) as created_at,
@@ -714,10 +842,10 @@ export async function getSharedTransactions(ownerId: number, tagIds?: number[],
           AND EXISTS (SELECT 1 FROM transaction_splits ts_me WHERE ts_me.transaction_id = t.id AND ts_me.participant_id = $1)
         )
       )
-      AND NOT (t.statement_id IS NULL AND t.reconciled_with_id IS NOT NULL)
+      AND ${EXCLUDE_RECONCILED_SOURCE}
     ${tagClause}
     ${participantClause}
-    GROUP BY t.id, o.category_override, o.merchant_normalized, o.notes, s.bank_name, s.owner_id, p_owner.name, src.created_at
+    GROUP BY t.id, o.category_override, o.merchant_normalized, o.notes, s.bank_name, s.currency, s.owner_id, p_owner.name, src.created_at
     ORDER BY t.transaction_date DESC
   `, params);
 
@@ -753,43 +881,53 @@ export interface TripAnalytics {
   daily_spend: { date: string; amount: number }[];
   top_merchants: { merchant: string; amount: number; count: number }[];
   tag_breakdown: { tag_id: number; name: string; color: string; amount: number; count: number }[];
-  participant_splits: { participant_id: number; name: string; owed: number }[];
+  participant_splits: {
+    participant_id: number;
+    name: string;
+    /** Their share of this trip, net of payments scoped to it. */
+    owed: number;
+    /** Splits counted at a non-AUD figure because no converted amount exists. */
+    unconverted_count: number;
+  }[];
 }
 
+// `total_spend` is the headline figure on the trips list and the trip header,
+// and it nets refunds for the same reason getTripAnalytics does — see the note
+// there. Trips are aliased `tr` so that `t` can be `transactions`, which is the
+// alias the shared fragments assume.
+const TRIP_TOTAL_SPEND = `COALESCE(SUM(
+        CASE WHEN ${NET_SPEND_ROWS}
+             AND ${EXCLUDE_RECONCILED_SOURCE}
+             AND COALESCE(o.category_override, t.category, 'other') NOT IN ('transfers', 'investment')
+        THEN ${SPEND_SIGNED} ELSE 0 END
+      ), 0)::float AS total_spend`;
+
 export async function getTrips(ownerId: number): Promise<TripRow[]> {
   return queryRaw<TripRow>(`
     SELECT
-      t.*,
+      tr.*,
-      COALESCE(SUM(
+      ${TRIP_TOTAL_SPEND},
-        CASE WHEN tx.transaction_type IN ('debit','fee','interest')
-             AND COALESCE(o.category_override, tx.category, 'other') NOT IN ('transfers', 'investment')
-        THEN COALESCE(tx.amount_aud, tx.amount) ELSE 0 END
-      ), 0)::float AS total_spend,
       COUNT(o.transaction_id)::int AS transaction_count
-    FROM trips t
+    FROM trips tr
-    LEFT JOIN transaction_overrides o ON o.trip_id = t.id
+    LEFT JOIN transaction_overrides o ON o.trip_id = tr.id
-    LEFT JOIN transactions tx ON tx.id = o.transaction_id
+    LEFT JOIN transactions t ON t.id = o.transaction_id
-    WHERE t.owner_id = $1
+    WHERE tr.owner_id = $1
-    GROUP BY t.id
+    GROUP BY tr.id
-    ORDER BY t.created_at DESC
+    ORDER BY tr.created_at DESC
   `, [ownerId]);
 }
 
 export async function getTripById(id: number, ownerId: number): Promise<TripRow | null> {
   const rows = await queryRaw<TripRow>(`
     SELECT
-      t.*,
+      tr.*,
-      COALESCE(SUM(
+      ${TRIP_TOTAL_SPEND},
-        CASE WHEN tx.transaction_type IN ('debit','fee','interest')
-             AND COALESCE(o.category_override, tx.category, 'other') NOT IN ('transfers', 'investment')
-        THEN COALESCE(tx.amount_aud, tx.amount) ELSE 0 END
-      ), 0)::float AS total_spend,
       COUNT(o.transaction_id)::int AS transaction_count
-    FROM trips t
+    FROM trips tr
-    LEFT JOIN transaction_overrides o ON o.trip_id = t.id
+    LEFT JOIN transaction_overrides o ON o.trip_id = tr.id
-    LEFT JOIN transactions tx ON tx.id = o.transaction_id
+    LEFT JOIN transactions t ON t.id = o.transaction_id
-    WHERE t.id = $1 AND t.owner_id = $2
+    WHERE tr.id = $1 AND tr.owner_id = $2
-    GROUP BY t.id
+    GROUP BY tr.id
   `, [id, ownerId]);
   return rows[0] ?? null;
 }
@@ -798,44 +936,66 @@ export async function getTripAnalytics(tripId: number, ownerId: number): Promise
   const trip = await getTripById(tripId, ownerId);
   if (!trip) throw new Error("Trip not found");
 
+  // What the trip cost, with refunds subtracted.
+  //
+  // These four queries filtered on `transaction_type IN ('debit','fee','interest')`,
+  // which drops every refund and credit — so money that came back was still
+  // counted as trip spend. A partly-refunded booking read at its full price and
+  // a fully-refunded one read as pure cost.
+  //
+  // NET_SPEND_ROWS admits the refunds and SPEND_SIGNED carries their direction,
+  // the same pair the general analytics adopted after a refunded Expedia
+  // purchase read as $2,888.92 of spend. `transactions` is aliased `t` because
+  // the fragments assume that alias.
+  //
+  // A cancelled booking is a different case and is NOT handled here: both its
+  // legs are untagged from the trip by hand, because a trip the booking was
+  // cancelled from never incurred that cost at all. This nets the partial
+  // refunds — a price adjustment on a booking that did happen.
+  //
+  // COUNT(*) deliberately still counts refund rows: a refund is a transaction
+  // that occurred on the trip, even though it subtracts from the total.
   const [categoryRows, dailyRows, merchantRows, tagRows, splitRows] = await Promise.all([
     queryRaw<{ category: string; amount: number; count: number }>(`
       SELECT
-        COALESCE(o.category_override, tx.category, 'other') AS category,
+        COALESCE(o.category_override, t.category, 'other') AS category,
-        SUM(COALESCE(tx.amount_aud, tx.amount))::float AS amount,
+        SUM(${SPEND_SIGNED})::float AS amount,
         COUNT(*)::int AS count
       FROM transaction_overrides o
-      JOIN transactions tx ON tx.id = o.transaction_id
+      JOIN transactions t ON t.id = o.transaction_id
       WHERE o.trip_id = $1
-        AND tx.transaction_type IN ('debit','fee','interest')
+        AND ${NET_SPEND_ROWS}
-        AND COALESCE(o.category_override, tx.category, 'other') NOT IN ('transfers', 'investment')
+        AND ${EXCLUDE_RECONCILED_SOURCE}
+        AND COALESCE(o.category_override, t.category, 'other') NOT IN ('transfers', 'investment')
       GROUP BY 1
       ORDER BY 2 DESC
     `, [tripId]),
 
     queryRaw<{ date: string; amount: number }>(`
       SELECT
-        tx.transaction_date::text AS date,
+        t.transaction_date::text AS date,
-        SUM(COALESCE(tx.amount_aud, tx.amount))::float AS amount
+        SUM(${SPEND_SIGNED})::float AS amount
       FROM transaction_overrides o
-      JOIN transactions tx ON tx.id = o.transaction_id
+      JOIN transactions t ON t.id = o.transaction_id
       WHERE o.trip_id = $1
-        AND tx.transaction_type IN ('debit','fee','interest')
+        AND ${NET_SPEND_ROWS}
-        AND COALESCE(o.category_override, tx.category, 'other') NOT IN ('transfers', 'investment')
+        AND ${EXCLUDE_RECONCILED_SOURCE}
+        AND COALESCE(o.category_override, t.category, 'other') NOT IN ('transfers', 'investment')
       GROUP BY 1
       ORDER BY 1
     `, [tripId]),
 
     queryRaw<{ merchant: string; amount: number; count: number }>(`
       SELECT
-        COALESCE(o.merchant_normalized, tx.merchant_normalized, tx.merchant_name, tx.description) AS merchant,
+        COALESCE(o.merchant_normalized, t.merchant_normalized, t.merchant_name, t.description) AS merchant,
-        SUM(COALESCE(tx.amount_aud, tx.amount))::float AS amount,
+        SUM(${SPEND_SIGNED})::float AS amount,
         COUNT(*)::int AS count
       FROM transaction_overrides o
-      JOIN transactions tx ON tx.id = o.transaction_id
+      JOIN transactions t ON t.id = o.transaction_id
       WHERE o.trip_id = $1
-        AND tx.transaction_type IN ('debit','fee','interest')
+        AND ${NET_SPEND_ROWS}
-        AND COALESCE(o.category_override, tx.category, 'other') NOT IN ('transfers', 'investment')
+        AND ${EXCLUDE_RECONCILED_SOURCE}
+        AND COALESCE(o.category_override, t.category, 'other') NOT IN ('transfers', 'investment')
       GROUP BY 1
       ORDER BY 2 DESC
       LIMIT 10
@@ -844,42 +1004,91 @@ export async function getTripAnalytics(tripId: number, ownerId: number): Promise
     queryRaw<{ tag_id: number; name: string; color: string; amount: number; count: number }>(`
       SELECT
         tg.id AS tag_id, tg.name, tg.color,
-        SUM(COALESCE(tx.amount_aud, tx.amount))::float AS amount,
+        SUM(${SPEND_SIGNED})::float AS amount,
-        COUNT(DISTINCT tx.id)::int AS count
+        COUNT(DISTINCT t.id)::int AS count
       FROM transaction_overrides o
-      JOIN transactions tx ON tx.id = o.transaction_id
+      JOIN transactions t ON t.id = o.transaction_id
-      JOIN transaction_tags tt ON tt.transaction_id = tx.id
+      JOIN transaction_tags tt ON tt.transaction_id = t.id
       JOIN tags tg ON tg.id = tt.tag_id
       WHERE o.trip_id = $1
-        AND tx.transaction_type IN ('debit','fee','interest')
+        AND ${NET_SPEND_ROWS}
-        AND COALESCE(o.category_override, tx.category, 'other') NOT IN ('transfers', 'investment')
+        AND ${EXCLUDE_RECONCILED_SOURCE}
+        AND COALESCE(o.category_override, t.category, 'other') NOT IN ('transfers', 'investment')
       GROUP BY tg.id
       ORDER BY 4 DESC
     `, [tripId]),
 
-    // No settled/unsettled breakdown here. It was computed from
+    // Owed per participant for THIS trip, net of payments made against it.
-    // transaction_splits.settled, which only /api/splits/settle writes and
+    //
-    // nothing in the UI calls — so it is false on all 673 splits and every trip
+    // This was gross splits, with a comment explaining why it could not be
-    // reported 100% unsettled, including trips paid in full. A real per-trip
+    // anything better: `split_payments` carried no trip attribution, so a
-    // figure is not computable either: split_payments carries no trip
+    // payment could not be assigned to a trip and every trip reported 100%
-    // attribution, so a payment cannot be assigned to a trip. Settlement is a
+    // unsettled including trips paid in full. `split_payments.trip_id`
-    // property of the whole relationship until settlement contexts exist
+    // (migration 0022) closes that, so the figure is now real.
-    // (see docs/shared-expenses-design.md).
+    //
-    queryRaw<{ participant_id: number; name: string; owed: number }>(`
+    // Three exclusions, all load-bearing:
-      SELECT
+    //  - ACTIVE_OBLIGATION drops settled splits, so a closed trip reads zero
-        p.id AS participant_id,
+    //    rather than its original gross.
-        p.name,
+    //  - EXCLUDE_RECONCILED_SOURCE drops the manual row a statement line has
-        SUM(ts.share_percent / 100.0 * COALESCE(tx.amount_aud, tx.amount))::float AS owed
+    //    superseded. The trip queries never applied it, so a reconciled trip
+    //    expense was counted twice here.
+    //  - AMOUNT_UNCONVERTED counts rows whose AUD value is unknown, the same
+    //    way getParticipantBalances does. They are still summed (at their
+    //    native figure), so a non-zero count means this total is approximate
+    //    and the UI has to say so. A trip is where foreign rows actually live,
+    //    so netting a EUR figure against AUD ones silently is most likely to
+    //    bite exactly here.
+    //
+    // A fourth, and it is what "owed" actually means: only rows THIS owner paid
+    // for. Without ${OWNER_SCOPE} the figure sums every split on every trip
+    // transaction regardless of who paid, so it silently mixes debts owed to
+    // different people. On Europe 2026 that put $1,605.49 of Molina's share of
+    // Sonu-paid rows into a number labelled as owed to the owner — a debt that
+    // is real, but between the other two participants, and which they settled
+    // directly (split_payments id 5, Molina -> Sonu, exactly $1,605.49).
+    // A participant's own share of a row they paid for was in there too, which
+    // is nobody's debt at all.
+    //
+    // `transactions` is aliased `t` so the shared fragments apply directly —
+    // they assume that alias, and hand-inlining a copy is what let the
+    // reconciled-row exclusion drift out of the analytics routes to begin with.
+    queryRaw<{ participant_id: number; name: string; owed: number; unconverted_count: number }>(`
+      WITH owed AS (
+        SELECT ts.participant_id AS pid,
+          SUM(ts.share_percent / 100.0 * COALESCE(t.amount_aud, t.amount)) AS gross,
+          SUM(CASE WHEN ${AMOUNT_UNCONVERTED} THEN 1 ELSE 0 END) AS unconverted
         FROM transaction_overrides o
-      JOIN transactions tx ON tx.id = o.transaction_id
+        JOIN transactions t ON t.id = o.transaction_id
-      JOIN transaction_splits ts ON ts.transaction_id = tx.id
+        ${STATEMENTS_JOIN}
-      JOIN participants p ON p.id = ts.participant_id
+        JOIN transaction_splits ts ON ts.transaction_id = t.id
         WHERE o.trip_id = $1
-        AND tx.transaction_type IN ('debit','fee','interest')
+          AND ${OWNER_SCOPE} = $2
-        AND COALESCE(o.category_override, tx.category, 'other') NOT IN ('transfers', 'investment')
+          AND ts.participant_id <> $2
-      GROUP BY p.id
+          AND t.transaction_type IN ('debit','fee','interest')
+          AND COALESCE(o.category_override, t.category, 'other') NOT IN ('transfers', 'investment')
+          AND ${ACTIVE_OBLIGATION}
+          AND ${EXCLUDE_RECONCILED_SOURCE}
+        GROUP BY ts.participant_id
+      ),
+      -- Only payments made TO this owner. Payment 5 on Europe is Molina -> Sonu:
+      -- a real settlement, but of a debt between those two, so it must not
+      -- reduce what Molina owes here. Symmetrical with the owner scoping above.
+      paid AS (
+        SELECT sp.from_participant_id AS pid, SUM(sp.amount) AS amt
+        FROM split_payments sp
+        WHERE sp.trip_id = $1
+          AND sp.to_participant_id = $2
+        GROUP BY sp.from_participant_id
+      )
+      SELECT p.id AS participant_id, p.name,
+        (COALESCE(owed.gross, 0) - COALESCE(paid.amt, 0))::float AS owed,
+        COALESCE(owed.unconverted, 0)::int AS unconverted_count
+      FROM participants p
+      LEFT JOIN owed ON owed.pid = p.id
+      LEFT JOIN paid ON paid.pid = p.id
+      WHERE owed.pid IS NOT NULL OR paid.pid IS NOT NULL
       ORDER BY 3 DESC
-    `, [tripId]),
+    `, [tripId, ownerId]),
   ]);
 
   const num_days = (trip.start_date && trip.end_date)