Raw purchase claim
Amount, invented fixture currency, market, date, local unit and contributor statement stay together. A submission is not yet confirmed.
A submitted price is not yet a comparable price.
A fictional/composite LaunchReady sample for founders testing informal-market price intelligence: preserve the paid amount and local unit, review the photo and context separately, normalize only with evidence, and keep confidence visible.
Fictional sample: “Harbour City”, all markets, contributors, records, dates, photo references and amounts are invented. FXC is an invented fixture currency. No real city price, merchant, stall, contributor, image, saving or customer result is represented. The Bango public signal motivated the schema problem; LaunchReady is not affiliated with Bango.
1. Three evidence states before a price card
The product should show what happened to each submission—not compress every row into “verified”.
Photo and context check
Reviewer records whether a usable image, date, location and quantity cue exist. The fictional CSV contains placeholders, not published photos.
Comparable quantity
Only add weight/volume and price-per-standard-unit after evidence supports the conversion. Every normalized field below is intentionally blank.
2. The deliberately incomplete fixture
Eight fictional rows expose seven local-unit labels. Amounts cannot be responsibly ranked while size, grade, ripeness, measured quantity and photo review remain unresolved.
| Row | Market | Commodity | Raw local unit | Paid amount | Photo state | Normalized quantity | Verification |
|---|---|---|---|---|---|---|---|
| FX-P01 | North Market | Tomato | 1 basket | 24 FXC | Fictional placeholder | Blank | Not verified |
| FX-P02 | North Market | Tomato | 1 bucket | 18 FXC | Fictional placeholder | Blank | Not verified |
| FX-P03 | Riverside Market | Tomato | 3 small piles | 11 FXC | Fictional placeholder | Blank | Not verified |
| FX-P04 | Central Market | Onion | 0.5 bag | 31 FXC | Fictional placeholder | Blank | Not verified |
| FX-P05 | Central Market | Onion | 2 bowls | 9 FXC | Fictional placeholder | Blank | Not verified |
| FX-P06 | East Market | Rice | 4 tins | 26 FXC | Fictional placeholder | Blank | Not verified |
| FX-P07 | East Market | Pepper | 5 small piles | 8 FXC | Fictional placeholder | Blank | Not verified |
| FX-P08 | South Market | Plantain | 1 bunch | 15 FXC | Fictional placeholder | Blank | Not verified |
3. Founder-readable five-bullet review
1
Preserve the raw local unit
Do not overwrite “basket”, “bucket” or “small pile” with guessed kilograms. Keep raw amount, quantity label, market and date immutable; add normalization in separate fields.
Immediate change: use
local_unit, local_unit_quantity, normalized_quantity, normalized_unit and normalization_method.2
Make photo review a state, not decoration
A photo can support recency, unit shape and visible quality context, but does not prove weight, identity, payment or universal market price.
Immediate change: separate
photo_received, photo_reviewed, quantity_cue_visible and privacy_redaction_complete.3
Do not publish a cheapest-market ranking too early
Grade, variety, ripeness, bundle size and contributor selection can create false spreads. Show incomparability before a leaderboard.
Immediate change: require comparable-unit, quality-band and freshness gates before generating any market spread.
4
Keep contributor confidence auditable
Repeat evidence, disagreement, reviewer notes and recency can inform a confidence label. Do not infer trust from identity, neighbourhood or one accepted upload.
Immediate change: record repeat-submission count, conflict count, last review and the exact rule behind each confidence state.
5
Sell a five-commodity brief before a marketplace
The first paid outcome is a bounded weekly evidence brief for one city/category—not inventory, delivery, cold-chain operations or collective buying.
Immediate change: earn 100 reviewable rows, 10 repeat contributors and three buyer requests before expanding beyond a static brief.
4. The bounded paid scope
A$149–299 Public-source Price Ledger Readiness Review
One public/sample page plus 8–20 supplied public or redacted rows, returned in 48–72 hours with source/consent fields, unit-normalization policy, photo-evidence states, confidence/missingness rules, one buyer-brief template, safe-claims edits and an approval-gated distribution queue.
One public/sample page plus 8–20 supplied public or redacted rows, returned in 48–72 hours with source/consent fields, unit-normalization policy, photo-evidence states, confidence/missingness rules, one buyer-brief template, safe-claims edits and an approval-gated distribution queue.
Promote gate
One target city reaches 10 repeat contributor candidates, 100 reviewable rows and three buyer requests—or one paid city/category brief.
Kill gate
Most records cannot be normalized/reviewed, privacy-safe evidence is infeasible, contributors require continuing high subsidies, or value depends on fulfilment.
Boundary: no price-accuracy, complete-coverage, food-quality, supplier-quality, savings or procurement claim. No real contributors or photos in this sample. No inventory, ordering, payments, delivery, cold chain or marketplace operation.