Score C · 6.27/10Free to read$100K Hackathon · November 16, 2026 → December 16, 2026Nov 16, 2026

Cagette Route

Cagette Route turns a neighborhood grocer's WhatsApp, Instagram and phone orders into one aggregated pick list, customer labels and printable delivery route sheet.

One product, one founder, thirty days. Everything below is built to be executed inside the contest window — the official rules still decide what counts.

Decide in 20 seconds

The facts that decide whether this is worth thirty days of yours.

Speculative6.27/10
MVP build (solo)
5 days
Launch price
$149 · monthly
Sales to $1,000 revenue
7 customers
First paying customer
Direct French-language Instagram DM and phone calls to independent greengrocers and butchers in one city whose profiles advertise phone or WhatsApp local delivery.
Who pays

Owner-operator of an independent greengrocer, butcher or bakery in a French or Belgian city with 5-40 same-day local delivery orders per day and 1-2 drivers.

Biggest execution risk

Operational simplicity 3.8/10

Remove regulated data, human review steps and custom infrastructure; keep the product self-serve and stateless on managed services.

Why this could work in 30 days: one founder can ship the MVP in 5 days; the plan targets a first sale on day 9; 7 sales at $149 clears $1,000 of collected revenue; strongest signal is speed to first revenue at 8.3/10.

Score and projections come from the generated report and are not guarantees. The $100K Hackathon is run by Ship or Die — read the official rules before you build.

Your first 24 hours

12 concrete actions. Tick them off as you go — nothing is saved or sent.

0/12
Entry price
$149
Solo build
5 days
30d base revenue
$1,290
30d hackathon profit
$1,232
Section 1 of 8

Product opportunity and positioning

Target customer

Owner-operator of an independent greengrocer, butcher or bakery in a French or Belgian city with 5-40 same-day local delivery orders per day and 1-2 drivers.

Why now

The shop already has demand through WhatsApp and needs no marketplace cold start; a printable PDF can be demoed on the owner's own orders in 15 minutes.

Problem

Orders arrive as WhatsApp messages, screenshots and voicemails; the owner retypes them, walks the aisles repeatedly without an aggregated pick list and sequences addresses from memory.

Solution

Paste or forward raw order text or screenshots, review extracted customer/address/items/window, then export an aisle-sorted pick list and an A4 route sheet with labels and paid/cash status.

Differentiation · No commission, no consumer storefront, no fleet: the shop keeps its own customers and replaces the notebook-spreadsheet-Google Maps workaround with one printable artifact.

1. Product opportunity and positioning

The buyer already has demand. A greengrocer in Lyon or a butcher in Brussels taking 5-40 same-day orders through WhatsApp, Instagram DMs and the phone does not need a marketplace — they need their evening back. Today the owner retypes orders into a notebook or spreadsheet, walks the aisles three or four times because there is no aggregated pick list, and sequences addresses from memory before the driver leaves. That workaround costs an hour or more a day and produces missed items, late stops and repeated trips.

Cagette Route is not a storefront. It is the back-office artifact: one aisle-sorted pick list, per-customer labels, and a printable A4 route sheet carrying the delivery sequence, windows and paid/cash status. The shop keeps its own customers, pays no commission and joins no consumer network. That is the deliberate break from the two-sided marketplace the original EPICERY pursued: no network cold start, no fleet, no commission collection, no consumer app.

The wedge is narrow on purpose. One shop account, one city, one persona — the owner-operator whose phone ordering has crossed the threshold from manageable to messy. The painful output is concrete and demoable: the founder asks the owner to forward today's raw messages and returns their own route sheet before the call ends. No integration, no data migration, no procurement cycle, no regulatory approval.

JEV scored this concept 62.7/100 (band C) with speed-to-first-revenue highest at 8.3/10 and market demand lowest at 4.3/10. The plan compensates for the demand weakness in two ways: it only contacts shops that visibly advertise delivery or WhatsApp ordering, and it treats the free sample route sheet — not the pitch — as the sales mechanism. Operational simplicity scored 3.84/10, so the product stays intentionally thin: a mandated human review step is kept because it protects accuracy, not because it is elegant.

The commercial argument is arithmetic the owner can check. If the tool saves 45 minutes a day across owner and driver at a loaded cost of roughly $25/hour, the monthly value is over $500 against a $149 subscription. That is a hypothesis to be tested in demos, not a verified claim.

Section 2 of 8

Smallest sellable MVP

A one-shop web app that accepts pasted or uploaded orders, lets the owner correct them, and exports a printable pick list plus A4 route sheet — sold at $149/month.

Must-have features

  • Magic-link login for a single shop account
  • Product catalog CRUD with unit and category (aisle) fields
  • Paste box plus screenshot upload for raw WhatsApp/Instagram orders
  • LLM extraction to customer, address, items, quantities, delivery window, paid/cash
  • Human review queue with editable fields and low-confidence flags
  • Fuzzy product matching and quantity normalization against the catalog
  • Aggregated aisle-sorted pick list with per-product totals
  • A4 route sheet PDF with customer labels, sequence, windows and payment status
  • Drag-to-reorder stops and manual edit before export
  • Stripe Checkout subscription with a 7-day trial and account status page

Explicitly out of scope

  • Consumer-facing storefront or marketplace
  • Card processing for the shop's own customers
  • Driver mobile app or live GPS tracking
  • Inventory, accounting or POS integration
  • Automatic route optimization
  • Multi-shop franchise management
  • WhatsApp Business API integration (paste and forward only)

Demo flow (what the buyer sees)

  1. Owner signs in with a magic link on the founder's screen share
  2. Owner pastes three real WhatsApp orders or drops screenshots
  3. Extraction returns three drafts; owner fixes two wrong quantities in the review queue
  4. Pick list aggregates 14 items across the three orders, sorted by aisle
  5. Route sheet PDF renders three stops with labels, windows and cash status
  6. Owner downloads the PDF and prints it during the call
Build plan

Technical implementation plan

Stack

Next.js on VercelSupabase Auth, Postgres and Storage (EU region)OpenAI API with vision model for order extractionReact-PDF for route sheets and labelsStripe Checkout and BillingResend for transactional emailVercel Cron for the morning digest

Architecture

A single Next.js app on Vercel talks to Supabase for auth, Postgres and screenshot storage. Order intake posts raw text or an image URL to a server route that calls OpenAI with a strict JSON schema and stores the result as a draft order. The owner reviews drafts in the browser, corrected line items are fuzzy-matched to catalog rows, and a server-side aggregation query builds the pick list. React-PDF renders the route sheet on demand from saved orders, and Stripe webhooks flip the shop between trial and active. A cron job emails the morning pick-list link.

Build steps

  1. Scaffold the Next.js app and connect Supabase with RLS scoped per shop
  2. Create shops, products, orders, order_items and route_stops tables
  3. Ship magic-link auth and an empty dashboard
  4. Build the paste box and the screenshot upload path
  5. Write and freeze the extraction prompt with a strict JSON schema
  6. Build the review queue with editable fields and confidence flags
  7. Add fuzzy product matching and unit normalization
  8. Build the aggregation query and pick list view
  9. Render the route sheet and label PDFs with React-PDF
  10. Add drag-to-reorder with persisted sequence
  11. Wire Stripe Checkout, trial and webhook status page
  12. Publish the French landing page and a 40-second screen recording

AI components

  • OpenAI vision/text extraction of customer, address, items, quantities, delivery window and payment status from raw messages and screenshots
  • Fuzzy string matching of extracted item names to catalog SKUs with unit normalization
Hardest technical risk: Extraction accuracy on photographed handwritten orders; if corrections per order exceed roughly four, the time-saving promise collapses and image intake must be dropped in favour of paste-only.
Section 3 of 8

Pricing and monetization

Tournée Solo

Entry offer
$149 / month

For One independent shop with up to 300 delivery orders per month

  • WhatsApp and photo order parsing with review queue
  • Aggregated aisle-friendly pick list
  • Printable A4 route sheet and customer labels
  • Manual route reorder and PDF export

Tournée Pro

$249 / month

For Shop with two vans or higher order volume

  • Everything in Solo
  • Multi-van route split
  • Priority parsing
  • CSV export for driver sheets

Model: Monthly subscription per shop, no commission, with a first-month discount for the first ten shops.

First revenue offer: First month at $99 for the first ten shops, with the catalog built from their menu photos and their next 20 real orders run free before any charge.

Expected ACV: $149

Upsell path: Move two-van shops to Pro at $249, then sell paid catalog setup and a January onboarding sprint, keeping all revenue on the same monthly subscription.

Target customers
50
Revenue model
$7,450
Costs
$380
Profit model
$7,070
Section 4 of 8

First-customer acquisition

Primary channel

Direct French-language Instagram DM and phone calls to independent greengrocers and butchers in one city whose profiles advertise phone or WhatsApp local delivery.

Where to find them

  • Google Maps listings in one French metro filtered for 'livraison' and independent shops
  • Instagram profiles with 'commande par WhatsApp' in the bio
  • Local market hall tenant lists (Halles, marché couvert)
  • Facebook neighborhood groups where grocers post weekly baskets
  • Butchers and bakeries with a visible delivery van in post photos
  • Shops already replying publicly to DM orders in their comment threads

Exact outreach motions

  • Send 30 Instagram DMs daily with the free 20-order offer and no attachment
  • Call 15-20 shops between 10:00 and 12:00, when counters are quiet
  • On reply, ask for one screenshot of today's orders instead of a meeting
  • Run a 15-minute screen share and return their own route sheet PDF the same evening
  • Send the Stripe link during the call, not after, with the $99 first-month price
  • Follow up at 48 hours with the minutes-saved estimate and a catalog-setup offer

Message template

Bonjour, je suis développeur à [ville]. Je transforme les commandes WhatsApp et photos en liste de préparation et feuille de tournée imprimable, en 3 minutes. Je le fais gratuitement sur vos 20 prochaines commandes, sans engagement. Vous pouvez m'envoyer une capture d'une commande d'aujourd'hui ?
Week-one goal: 80 shops listed, 60 touches sent, 8 demos booked and 3 route sheets generated on real orders.
Section 5 of 8

30-day execution roadmap

MVP deadline: day 5First sale target: day 930 days planned
Week 1 · Build & validate · 7 days
Day 1Foundation

Auth and data model

  • Scaffold Next.js and Supabase on Vercel.
  • Create shops, products, orders and order_items tables with RLS.
  • Ship magic-link login and an empty dashboard.

Deliverable: Deployed app a shop can sign into.

Day 2Order intake

Raw messages to structured drafts

  • Build a paste box that accepts raw WhatsApp text.
  • Write an extraction prompt returning strict JSON for customer, address, items, window and paid status.
  • Add screenshot upload to Supabase Storage with an OpenAI vision call.

Deliverable: Any pasted or uploaded order returns draft JSON.

Day 3Review queue

Make corrections take seconds

  • Show the extracted order beside its raw source, fully editable.
  • Fuzzy-match items to the catalog with unit normalization.
  • Amber-flag low-confidence fields.

Deliverable: Owner saves a corrected order in under 60 seconds.

Day 4Output artifacts

Pick list and route sheet

  • Aggregate orders into an aisle-sorted pick list with totals.
  • Render A4 route sheet PDFs with labels, sequence, windows and paid status.
  • Allow drag-to-reorder stops before export.

Deliverable: Printable pick list and route sheet from real orders.

Day 5Launch surfaceMVP shipped

Landing page and checkout

  • Publish a French landing page with a 40-second screen recording.
  • Wire Stripe Checkout for Solo $149 and Pro $249 with a 7-day trial.
  • Add a catalog builder that parses pasted menu text or photos.

Deliverable: Public URL with working checkout — MVP deadline.

Day 6Outbound

Build the 80-shop list

  • List 80 shops in one metro from Google Maps and Instagram where delivery is visible.
  • Send 30 French Instagram DMs with the free 20-order offer.
  • Call 15 shops between 10:00 and 12:00.

Deliverable: 80-shop list and 45 touches sent.

Day 7First demos

Demo on the owners' own orders

  • Run four fifteen-minute screen shares.
  • Ask each owner to forward today's raw orders and build their route sheet live.
  • Email each owner their own PDF the same evening.

Deliverable: 4 sample route sheets on real orders.

Week 2 · Launch & first sale · 7 days
Day 8Fix loop

Repair what demos exposed

  • Fix the two worst parsing failures from demo calls.
  • Add shop-specific synonyms so local product names map to SKUs.
  • Record a 90-second Loom of the demo for follow-ups.

Deliverable: Higher extraction accuracy on the demo set.

Day 9First saleFirst sale target

Ask for the card

  • Send the Stripe link to the two warmest owners with the $99 first-month offer.
  • Offer free catalog setup from menu photos.
  • Ask for a yes or no today instead of an open maybe.

Deliverable: First paid subscription collected in-window.

Day 10Onboarding

One shop runs a real morning

  • Hand-build catalogs for every shop that said yes.
  • Walk their first delivery run end to end on the call.
  • Record minutes saved before and after.

Deliverable: 1-2 shops printing a real morning route sheet.

Day 11Outreach volume

Push past 100 touches

  • Send 40 DMs quoting the measured minutes saved.
  • Call 20 shops, prioritising butchers for smaller catalogs.
  • Log every no with its stated reason.

Deliverable: 85 cumulative touches and 8 booked demos.

Day 12Demo block

Five calls back to back

  • Run five demos with the Stripe link ready at the end of each.
  • Note every request that widens the wedge.
  • Send same-day PDFs to all five.

Deliverable: 5 demos and 1-2 new trials.

Day 13Referrals

Borrow trust from paying shops

  • Ask each paying owner for one other shop owner they rate.
  • Send a two-line intro they can forward on WhatsApp.
  • Turn an anonymized route sheet into a one-page case note.

Deliverable: 3 warm intros and 1 case note.

Day 14Week 2 review

Read the funnel honestly

  • Tabulate contacts, replies, demos, trials and paid.
  • Kill the message variant with the worst reply rate.
  • Patch the noisiest support complaint.

Deliverable: Funnel table and one repeatable message.

Week 3 · Repeat & systematize · 7 days
Day 15Pro tier

Give two-van shops a reason to pay more

  • Add multi-van route split behind a plan flag.
  • Add CSV export for driver sheets.
  • Offer Pro at $249 to any shop with two drivers.

Deliverable: Pro tier live and priced.

Day 16Outreach volume

A second list of 40 shops

  • Extract 40 new shops from Instagram hashtags and market hall lists.
  • Send 40 DMs and make 20 calls.
  • Re-test the winning message in a new arrondissement.

Deliverable: 125 cumulative touches.

Day 17Demo block

Five more demos

  • Run five demos with live PDF generation.
  • Send the Stripe link on the call, not after.
  • Book a dated follow-up for anyone hesitating.

Deliverable: 5 demos and follow-up dates.

Day 18Onboarding

Cut setup to 20 minutes

  • Turn catalog setup into a reused checklist.
  • Pre-fill addresses and windows from past screenshots.
  • Confirm the 07:30 printing ritual with each owner.

Deliverable: Reusable 20-minute onboarding.

Day 19Retention

Make the morning automatic

  • Email a 07:00 digest when the day's pick list is ready.
  • Add a one-click today's-route-sheet link.
  • Track who opens it two mornings in a row.

Deliverable: Daily digest email live via Resend and cron.

Day 20Money hygiene

Keep the ledger clean

  • Confirm Stripe is the only revenue path and connected to TrustMRR.
  • Reconcile every charge and refund against the entry window.
  • Remove all test-mode payment artifacts.

Deliverable: Reconciled revenue ledger.

Day 21Week 3 review

Check who actually reuses it

  • Count shops that printed a route sheet on three or more days.
  • Call anyone who paid but went quiet.
  • Decide the single retention feature worth building.

Deliverable: Activation table and one retention decision.

Week 4 · Scale what works · 7 days
Day 22Holiday wedge

Sell into Christmas week chaos

  • Offer to pre-load Christmas-week orders from existing screenshots.
  • Position the tool as the fix for the busiest eight days of the year.
  • Add bulk paste mode for ten orders at once.

Deliverable: Bulk intake and a holiday offer message.

Day 23Outreach volume

Fourth wave

  • DM 40 shops on the not-now list with the holiday angle.
  • Call 20 shops that never replied.
  • Ask two paying shops to share the case note.

Deliverable: 160 cumulative touches.

Day 24Demo block

Five demos with holiday framing

  • Run five demos on a twelve-order Christmas morning.
  • Send the Stripe link within ten minutes of hanging up.
  • Offer to build their first Christmas list myself.

Deliverable: 5 demos and 1-2 closes.

Day 25Friction removal

One feature, earned

  • Add a per-shop inbound email address so exports can be forwarded.
  • Keep the review queue mandatory so accuracy holds.
  • Refuse anything requiring POS or a driver app.

Deliverable: Forward-by-email intake live.

Day 26Proof sheet

Make the numbers concrete

  • Compile real before and after minutes from onboarded shops.
  • Cut a 30-second clip of a route sheet printing.
  • Rewrite the DM opener around the strongest number.

Deliverable: One-page proof sheet and new opener.

Day 27Pipeline for January

Do not let the window end the pipeline

  • Book January demos with every after-Christmas reply.
  • Send a calendar link with two fixed slots per day.
  • Tag every non-buyer by objection.

Deliverable: 6+ January demos booked.

Day 28Week 4 review

Close the loop

  • Compare paid count against the three scenarios.
  • Fix the biggest refund or churn cause.
  • Freeze features and stop building.

Deliverable: Scenario-versus-actual table.

Days 29-30 · Consolidate · 2 days
Day 29Final close

Collect everything inside the window

  • Call every trial that has not converted.
  • Send a final invoice link with a one-day expiry.
  • Confirm each payment lands before December 16 13:00 UTC.

Deliverable: All collectable revenue collected.

Day 30Wrap

Leave with a verified number and a plan

  • Export the TrustMRR-verified profit figure.
  • Write the next 30 days on retention, referrals and January onboarding.
  • Publish a short build log with no invented metrics.

Deliverable: Verified profit number and next-30-day plan.

Section 6 of 8

Revenue and profit scenarios

How the numbers are defined

Hackathon profit — Collected revenue − refunds − chargebacks − payment processing fees − paid marketing spend.

Business profit — Hackathon profit − operating costs (hosting, domains, tooling, AI/API delivery costs).

The contest ranks entrants on hackathon profit only. Hosting, tooling and AI/API delivery costs are not deducted by the organisers, but each sale still has to cover them — so we show both. Definitions verified on 2026-11-15 against the official rules.

Stated assumptions

  • Conservative: 160 outbound touches convert to 7 demos and 3 paid shops, all on the $99 first-month price.
  • Base: 320 touches, 20 demos, 8 paid from demos plus 2 referrals, blended $129 because most pay the intro price in month one.
  • Upside: 450 touches, 38 demos, 16 paid plus 6 referrals, blended $122 with one refunded month.
  • Payment processing assumed at 4.5% of collected revenue (Stripe Checkout, EU cards).
  • No paid marketing in conservative or base; upside includes $150 of sponsored Instagram posts.
  • Operating costs are hosting, Supabase, domains, Resend and OpenAI extraction only; they are not deducted from hackathon profit.
  • The concept's 50-shop / $7,450 figure is treated as a ceiling requiring roughly 400 completed contacts and 90 demos, not as the plan.
  • All figures are projections from stated assumptions, not verified revenue or existing customers.

Conservative

3 customers
$297
collected revenue · avg payment $99
Refunds
$0
Chargebacks
$0
Payment fees
$13
Paid marketing
$0
Hackathon profit
$284
Operating costs
$35
Business profit
$249

Base case

10 customers
$1,290
collected revenue · avg payment $129
Refunds
$0
Chargebacks
$0
Payment fees
$58
Paid marketing
$0
Hackathon profit
$1,232
Operating costs
$85
Business profit
$1,147

Optimistic

22 customers
$2,684
collected revenue · avg payment $122
Refunds
$99
Chargebacks
$0
Payment fees
$121
Paid marketing
$150
Hackathon profit
$2,314
Operating costs
$160
Business profit
$2,154

Model your own numbers

Runs in your browser · nothing is saved
Collected revenue
$1,490
Payment fees (3.9%)
$58
Paid marketing
$0
Hackathon profit
$1,432
Business profit
$1,347

Hypothetical arithmetic, not a forecast. The fee rate is a flat 3.9% assumption, refunds and chargebacks are left at zero here, and none of this is tracked data.

All scenario figures are projections built from the stated assumptions. They are not verified revenue, existing customers or real traction, and any outcome depends on outreach response in the founder's own city.

Section 7 of 8

Main risks and validation experiments

Risk 1Cold outreach response is too low to book enough demos before December 16.

Mitigation

Run DM and phone in parallel; shop owners answer the phone between 10:00 and 12:00 far more reliably than DMs.

Validation experiment

On day 6 send 30 DMs and make 15 calls, then count replies and demo bookings within 48 hours.

Kill criteria

Fewer than 3 demo bookings from 60 touches by day 8: stop feature work and switch to walk-in visits at market halls before spending another day building.

Risk 2Photo and handwritten order extraction is too inaccurate to save the owner time.

Mitigation

Paste-first intake with vision as secondary; mandatory review queue and amber confidence flags keep errors visible.

Validation experiment

Run 20 real order screenshots through the day-3 build and count corrections per order.

Kill criteria

Average above four corrections per order: drop image intake entirely and ship paste and CSV only.

Risk 3Owners demand a consumer order page or POS integration, making the wedge feel too narrow.

Mitigation

Scripted answer that consumer ordering is a different product; log every request instead of building it.

Validation experiment

Track how many of the first ten demos ask for a customer-facing ordering page.

Kill criteria

More than half of demoed shops say the printable sheet is insufficient: re-scope month two to butchers only or exit the wedge.

Risk 4$149 per month exceeds impulse approval for very small shops.

Mitigation

First ten shops pay $99 for month one; demos always quantify minutes saved against the owner's own hourly cost.

Validation experiment

Offer $149 to five demoed shops and $99 to five others, then compare close rates.

Kill criteria

Zero of 15 demoed shops accept $99 or more: drop to $79 and re-test with ten fresh demos before continuing outreach.

Risk 5Order data includes customer names and addresses, creating GDPR obligations a solo founder handles badly.

Mitigation

Supabase EU region, minimal fields, self-serve deletion, published privacy note, no data resale or model training.

Validation experiment

Write a one-page privacy note and ask three owners whether it is enough for them to sign.

Kill criteria

Shops refuse to store order data in a third-party tool without a signed contract: pause and reconsider the data model.

Section 8 of 8

Hour-by-hour: the first 24 hours

Hours 1-2

  • Pick one French metro area and commit to it for 30 days.
  • Write the one-sentence French offer and the DM script verbatim.

Output: Chosen city and a frozen French message.

Hours 3-5

  • Build the 80-shop list from Google Maps and Instagram.
  • Record handle, phone, visible delivery signal and shop type for each.

Output: 80-row CSV ready for outbound.

Hours 6-8

  • Scaffold Next.js and Supabase, deploy to Vercel.
  • Create the core tables and magic-link login.

Output: Live skeleton URL with a working sign-in.

Hours 9-11

  • Write the extraction prompt and strict JSON schema.
  • Test it on ten raw WhatsApp order messages you write yourself.

Output: Prompt file plus fixture orders passing the schema.

Hours 12-14

  • Send 20 Instagram DMs from the list.
  • Make 5 phone calls between the calls window and log replies.

Output: First real replies and two demo candidates.

Hours 15-16

  • Draft the French landing page copy and the 40-second demo outline.
  • Create Stripe products for Solo $149 and Pro $249 in test mode.

Output: Page copy, demo storyboard and price IDs.

Complete report

Full 30-day launch report

Read the full generated report

Cagette Route — 30-Day Launch Report

One founder. Window: November 16, 2026 13:00 UTC to December 16, 2026 13:00 UTC. All financial figures below are projections.

1. Product opportunity and positioning

The buyer already has demand. A greengrocer in Lyon or a butcher in Brussels taking 5-40 same-day orders through WhatsApp, Instagram and the phone does not need a marketplace — they need their evening back. Today the owner retypes orders into a notebook or spreadsheet, walks the aisles three or four times because there is no aggregated pick list, and sequences addresses from memory before the driver leaves. That workaround costs an hour a day and produces missed items, late stops and repeated trips.

Cagette Route is not a storefront. It is the back-office artifact: one aisle-sorted pick list, per-customer labels, and a printable A4 route sheet carrying delivery sequence, windows and paid/cash status. The shop keeps its own customers, pays no commission and joins no consumer network. That is the deliberate break from the two-sided marketplace the original EPICERY pursued — no network cold start, no fleet, no commission collection.

The wedge is narrow on purpose: one shop account, one city, one persona — the owner-operator whose phone ordering has just crossed from manageable to messy. The painful output is concrete and demoable. The founder asks the owner to forward today's raw messages and returns their own route sheet before the call ends. No integration, no migration, no procurement, no approval body.

JEV scored the concept 62.7/100 (band C), highest on speed-to-first-revenue (8.3/10) and lowest on market demand (4.3/10). The plan answers that weakness by only contacting shops that visibly advertise delivery or WhatsApp ordering, and by treating the free sample route sheet — not the pitch — as the sales mechanism. Operational simplicity scored 3.84/10, so the product stays thin; the mandatory human review step is kept because it protects accuracy, not because it is elegant.

The commercial argument is arithmetic the owner can check: 45 minutes a day saved across owner and driver is worth over $500 a month against a $149 subscription. That is a hypothesis to be tested in demos, not a verified claim.

2. Smallest sellable MVP

The MVP is one web app, one shop, three outputs: an editable review queue, an aisle-sorted pick list, and an A4 route sheet PDF with labels. Intake is a paste box plus screenshot upload. Catalog is a simple CRUD table with an aisle field. Billing is Stripe Checkout with a seven-day trial. That is enough to run a real morning in a real shop.

Deliberately absent: consumer storefront, card processing for the shop's customers, driver app, GPS, POS integration, automatic route optimization. Each of these converts a one-week build into a one-quarter build and none of them is required for the first payment.

3. Technical implementation plan

Next.js on Vercel, Supabase for auth, Postgres and screenshot storage in an EU region, OpenAI vision and text for extraction, React-PDF for the sheet, Stripe for billing, Resend for the morning digest. A server route posts raw text or an image URL to OpenAI with a strict JSON schema and saves a draft order. Corrections are fuzzy-matched to catalog rows. An aggregation query builds the pick list; React-PDF renders the route sheet on demand; Stripe webhooks flip trial to active.

The single largest technical risk is extraction accuracy on photographs of handwriting. Mitigation is ordering: paste works on day 2, vision is secondary, and the review queue is permanent. If the average order needs more than four corrections, image intake is removed rather than propped up.

4. Pricing and monetization

Tournée Solo at $149/month covers up to 300 delivery orders a month, one shop. Tournée Pro at $249/month adds multi-van split, priority parsing and CSV export for driver sheets. The first ten shops pay $99 for month one, with the catalog built from their menu photos at no charge. No commission is charged on order value — that is the pricing story that separates this from every marketplace the owner has already refused.

The upsell path is volume-driven, not feature-driven: two-van shops move to Pro, and paid catalog setup absorbs the founder's idle hours without becoming consulting revenue.

5. First-customer acquisition

One channel only: French-language Instagram DM plus phone calls in a single metro area. The list comes from Google Maps and Instagram profiles that mention delivery or WhatsApp ordering. The message is short and offers work before money: the founder processes the next 20 real orders free.

The funnel to plan against, as a projection: 320 touches produce roughly 38 replies, 20 demos and 8 paid shops, plus 2 referrals. The conservative case assumes half that reply rate; the upside assumes 450 touches and a working referral loop. The demo is the product: a 15-minute screen share, the owner forwards real orders, the PDF lands in their inbox that evening, and the Stripe link is sent during the call rather than after it.

6. 30-day execution roadmap

Days 1-5 build the sellable core and end with a public page and working checkout. Day 4 must produce a real route sheet from real orders; if it does not, day 5 slips and the first-sale target moves with it.

Days 6-14 are outreach, demos and the first payment, with day 9 as the first-sale target and day 8 reserved for fixing whatever the demo calls exposed. Days 15-21 add the Pro tier, a morning digest email, and a ledger reconciliation so every charge and refund is attributable inside the contest window.

Days 22-28 use the holiday peak as the wedge — offering to pre-load Christmas-week orders — and keep building January pipeline with anyone who says after Christmas. Days 29-30 are collection and wrap: no features, only calls, invoices and a verified profit number.

7. Revenue and profit scenarios

Three projections, all built from stated assumptions. Conservative: 3 shops, $99 blended, $297 gross, $13 processing, $35 operating. Base: 10 shops, $129 blended, $1,290 gross, $58 processing, $85 operating. Upside: 22 shops, $122 blended, $2,684 gross, $99 refunded, $121 processing, $150 in sponsored posts, $160 operating.

Two profit numbers matter and are not interchangeable. Hackathon profit deducts refunds, chargebacks, processing and paid marketing only — hosting and AI delivery costs are excluded from the ranking. Business profit deducts operating costs as well. Every scenario here clears its delivery cost per sale, so no shop is sold below cost. The concept's own $7,450 target implies roughly 50 shops and 400 completed contacts; the plan treats that as a ceiling, not a forecast.

8. Main risks and validation experiments

Outreach response is the first risk, tested on day 6 with 30 DMs and 15 calls; fewer than three demo bookings from 60 touches means switching to market-hall walk-ins before writing more code. Extraction accuracy is the second, tested by running 20 real screenshots through the day-3 build and counting corrections. Scope pressure is the third: if more than half of demoed shops demand a consumer ordering page, the wedge is wrong and month two re-scopes to butchers only. Price resistance is the fourth, tested with a split between $99 and $149 offers across ten demos. GDPR handling is the fifth: EU hosting, minimal fields, self-serve deletion, and a privacy note shown in demos.

9. First 24-hour action plan

Pick one metro and freeze the French offer (hours 1-2). Build the 80-shop list (3-5). Scaffold and deploy the app skeleton (6-8). Write and test the extraction prompt on ten self-written messages (9-11). Send 20 DMs and make five calls (12-14). Draft the landing page and create the Stripe products (15-16). By the end of day one the founder should have a live URL, a tested prompt, a usable list and at least two real replies.

All revenue and profit statements in this report are projections from stated assumptions, not verified results. No customers, partnerships or sources are claimed.

This brief was generated from public data about established startups, then narrowed and scored for a solo 30-day launch. Revenue and profit figures are scenario models, not verified results, guarantees or evidence of existing customers. Validate demand yourself before you spend a month on it — and check the contest rules for what counts toward your score.

Related ideas

All ideas →