Score C · 6.03/10Free to read$100K Hackathon · November 16, 2026 → December 16, 2026Nov 16, 2026

MenuProof Guide

MenuProof Guide turns a restaurant's existing menu into a branded, owner-verified healthy dining guide PDF and QR page in 24 hours, helping independent restaurants win health-conscious lunch customers.

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.

Speculative6.03/10
MVP build (solo)
4 days
Launch price
$99
Sales to $1,000 revenue
11 customers
First paying customer
Off-peak in-person visits to independent restaurants in one dense lunch district, opening with a pre-generated sample of their own menu.
Who pays

Owner or manager of an independent restaurant or cafe with 10-50 menu items and 1-3 locations in a dense urban lunch market (US/Canada).

Biggest execution risk

Market demand 4.0/10

Narrow the buyer to a segment with an urgent, already-budgeted problem and reference the specific trigger event that creates willingness to pay.

Why this could work in 30 days: one founder can ship the MVP in 4 days; the plan targets a first sale on day 10; 11 sales at $99 clears $1,000 of collected revenue; strongest signal is speed to first revenue at 8.7/10.

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
$99
Solo build
4 days
30d base revenue
$2,475
30d hackathon profit
$2,279
Section 1 of 8

Product opportunity and positioning

Target customer

Owner or manager of an independent restaurant or cafe with 10-50 menu items and 1-3 locations in a dense urban lunch market (US/Canada).

Why now

The restaurant already owns the menu, the deliverable is printable and linkable today, $99 is an operating expense, and the owner is reachable in person at 2pm on a Tuesday.

Problem

Health-conscious lunch customers ask about vegan, gluten-free, dairy-free and high-protein options; staff answer inconsistently and the restaurant has no artifact that shows what is actually available.

Solution

Owner uploads a menu PDF or photo, OCR and an LLM propose dietary tags with confidence flags, the owner confirms them in a 5-minute checklist, and the system emails a branded PDF guide plus a hosted QR page within 24 hours.

Differentiation · Not a menu rewrite, not a POS integration, not a delivery marketplace: a single verified artifact the owner signs off on, priced as a one-time op-ex line.

1. Product opportunity and positioning

Independent restaurants in dense lunch districts compete for the same office-worker foot traffic, and a growing share of that traffic screens menus for vegan, gluten-free, dairy-free, high-protein and low-carb options before walking in. The restaurant already has the answer buried in its own menu and in the head of one line cook. There is no artifact that says it publicly. Staff answer dietary questions inconsistently at the counter, and the owner does not have three hours to tag 40 dishes by hand.

MenuProof Guide sells a finished artifact, not software access. The owner uploads the menu they already have. OCR and an LLM extract dishes and ingredients and propose dietary tags with confidence flags. The owner confirms or edits those tags in a five-minute checklist and accepts an explicit attestation and disclaimer. Within 24 hours they receive a branded one-page Healthy Dining Guide PDF plus a hosted QR page with filters, so the guide can be printed for the host stand, taped to the counter, linked from an Instagram bio, or shown on a phone.

Who buys. The owner or manager of an independent restaurant or cafe with 10-50 menu items and 1-3 locations. The buying trigger is concrete: they are losing health-conscious lunch traffic to a competitor that markets a vegan bowl, or their staff keeps getting the same four dietary questions and answering them badly.

Why this wedge and not a platform. The product is deliberately narrow because the seller is one person with 30 days. There is no POS integration, no delivery logistics and no marketplace cold start. The output is visible before the owner pays: a sample guide built from their own menu. That is the entire sales mechanism. The 2-4pm window between lunch and dinner service is when an owner is physically present, not managing a rush, and can look at a printed page for ninety seconds and decide.

Where the value shows up. The owner gets a tag-checked dietary matrix they did not have to build, a printable artifact and a QR destination page, with a visible disclaimer that the restaurant attests to the tags. At $99 one-time against a single lunch cover table turning over faster, the arithmetic is trivial for the owner and does not require a budget cycle, a procurement process or a signature.

Honest limits. JEV scored this concept in band C (60.3/100). Strength is speed to first revenue (8.68 normalized) and solo build feasibility (7.69). Weakness is market demand (3.96) and monetization potential (4.16): the buyer's pain is real but not urgent, and the price is a single low-value charge. The 30-day plan compensates by making the sale in person with a pre-built sample, and by opening a $149 repeat-purchase path (catering page plus three QR codes) so a single buyer can transact twice. Operational simplicity (4.97) is constrained by the human verification step and the 24-hour delivery promise; the plan keeps one revision and batch fulfilment so a solo operator can hold the SLA.

Section 2 of 8

Smallest sellable MVP

A paid Stripe Checkout that takes a menu PDF or photo, produces an LLM-tagged dish list for owner confirmation, and delivers a branded PDF guide plus hosted QR page by email within 24 hours.

Must-have features

  • Menu upload and OCR text extraction into a structured dish list
  • LLM dietary tag suggestions (vegan, vegetarian, gluten-free, dairy-free, high-protein, low-carb) with confidence flags
  • Owner verification form to confirm or edit tags and accept the attestation and disclaimer
  • Branded one-page PDF Healthy Dining Guide generator with restaurant logo and colors
  • Hosted QR landing page with dietary filters and a permanent visible disclaimer
  • Stripe Checkout for $99 Menu Guide and $149 Menu Guide + Catering Page
  • Gated download unlocked only after successful payment
  • Admin queue in Supabase plus Resend email delivery with tracking
  • One-revision workflow with a visible 24-hour delivery clock

Explicitly out of scope

  • Online ordering, delivery or POS integration
  • Calorie, macro or nutrition calculation claims
  • Any allergen-compliance or legal guarantee
  • Full restaurant website builder
  • Consumer-facing discovery marketplace
  • Multi-location enterprise workflow
  • Subscription billing or seat-based plans
  • Native mobile app

Demo flow (what the buyer sees)

  1. Owner uploads a menu PDF or photo on the landing page
  2. OCR extracts dish names, descriptions and prices into a structured list
  3. LLM proposes dietary tags per dish with confidence flags
  4. Owner opens the verification checklist, corrects tags and ticks the attestation
  5. Owner pays $99 through Stripe Checkout
  6. Admin runs guide generation and triggers delivery
  7. Owner receives a branded PDF and a hosted QR page link by email within 24 hours
  8. Owner prints the guide and posts the QR page link
Build plan

Technical implementation plan

Stack

Next.js 14 on VercelSupabase Postgres with row-level securityStripe Checkout and webhooksOpenAI GPT-4o mini for dish extraction and tag suggestionPDF.co for OCR with Tesseract.js fallbackreact-pdf / @react-pdf/renderer for the branded guideResend for transactional emailCloudflare R2 or Supabase Storage for menu and PDF filesQR code generation via a local library, no third-party redirect

Architecture

A single Next.js app on Vercel hosts the marketing page, the upload form, the owner verification checklist, the hosted QR guide page and an internal admin queue. Every order is a row in Supabase with a state machine: uploaded, parsed, tagged, awaiting_owner, paid, generated, delivered, revised. OCR and LLM calls happen in serverless route handlers immediately after upload; the resulting dish JSON is stored so regeneration never re-runs inference. Stripe webhooks flip an order to paid and unlock the PDF generation job. Generation writes a PDF to object storage and a Resend email sends the PDF plus the QR page URL. The hosted QR page is a dynamic route keyed by a public order slug, rendering the verified tags and filtering client-side. Admin queue is a password-protected route listing orders by state and SLA age.

Build steps

  1. Scaffold Next.js on Vercel, create Supabase project and orders/dish tables, wire env vars
  2. Build upload route that stores the file and runs OCR, returning raw text
  3. Add LLM extraction prompt returning strict JSON dish list; validate with a schema and store it
  4. Add LLM tagging prompt returning tag plus confidence per dish; render a confidence flag in the UI
  5. Build the owner verification checklist with edit, attestation checkbox and disclaimer
  6. Build the react-pdf guide template with logo, brand colors, tag legend and disclaimer footer
  7. Build the hosted QR page route with dietary filters and a permanent disclaimer banner
  8. Add Stripe Checkout with two price IDs and a webhook that sets paid=true
  9. Add Resend delivery with PDF attachment and QR link, plus the admin queue view
  10. Test end-to-end on 5 real public menus, fix parsing edge cases, then soft-launch

AI components

  • GPT-4o mini for structured dish extraction from OCR text
  • GPT-4o mini for dietary tag suggestion with per-tag confidence
  • Deterministic post-rules that force low-confidence tags to require manual owner confirmation
Hardest technical risk: OCR quality on photographed or heavily styled menus. Multi-column layouts, decorative fonts and handwritten specials boards will produce garbage text, and a bad tag suggestion on a real menu destroys trust in front of the owner.
Section 3 of 8

Pricing and monetization

Menu Guide

Entry offer
$99 one-time

For Independent restaurant owner

  • One menu up to 50 items
  • Owner verification checklist
  • Branded PDF guide
  • Hosted QR page
  • One revision
  • 24-hour email delivery

Menu Guide + Catering Page

$149 one-time

For Restaurant with office lunch or catering inquiries

  • Everything in Menu Guide
  • Separate catering menu guide
  • Three QR codes for tables and counters

Model: One-time fixed-fee productized deliverable, paid up front through Stripe Checkout, with a second higher-priced tier that adds a catering guide and extra QR codes.

First revenue offer: $99 for a published Healthy Dining Guide built from your current menu, delivered within 24 hours of a 5-minute tag check, with one revision included.

Expected ACV: $99

Upsell path: Every $99 buyer is offered the $149 catering guide within 7 days; seasonal menu changes create a natural $49 re-issue at month 2, which is the only path to repeat revenue in the first 60 days.

Target customers
30
Revenue model
$2,970
Costs
$130
Profit model
$2,840
Section 4 of 8

First-customer acquisition

Primary channel

Off-peak in-person visits to independent restaurants in one dense lunch district, opening with a pre-generated sample of their own menu.

Where to find them

  • A single walkable lunch district with 30+ independent cafes and restaurants
  • Public menus scraped from restaurant websites and Google Maps listings
  • Restaurant-owner Facebook groups for the metro area
  • Instagram bios of independent cafes that list a dietary claim
  • Neighbouring businesses referred by an existing buyer
  • Local chamber of commerce and business-improvement district newsletters

Exact outreach motions

  • Walk in at 2-4pm, ask for the owner or manager, hand them a printed guide built from their own menu
  • Open with the tag matrix, not the price: show the vegan/gluten-free rows already filled in
  • Offer to publish the PDF and QR page for $99 with delivery inside 24 hours of their tag check
  • If they hesitate, leave a QR card linking to their sample and email the next morning
  • Ask every buyer for one neighbouring restaurant introduction
  • Post a before/after menu sample in two local owner groups with the sample link

Message template

Hi — I'm the person behind MenuProof Guide. I already turned the menu on your website into a one-page Healthy Dining Guide (sample attached, no charge). It tags the vegan, gluten-free, dairy-free and high-protein dishes so your staff stops guessing at the counter. If you want it published with your logo plus a QR page, it's $99 and I deliver within 24 hours of your 5-minute tag check. Want me to send the checklist?
Week-one goal: Code deployed and payment live by day 5, 30 restaurants listed with menus captured, 10 free sample guides generated, 8 in-person visits completed by day 7.
Section 5 of 8

30-day execution roadmap

MVP deadline: day 5First sale target: day 1030 days planned
Week 1 · Build & validate · 7 days
Day 1Foundation

Repo and infrastructure

  • Scaffold Next.js on Vercel with Supabase connected.
  • Create orders, dishes and tags tables with RLS.

Deliverable: Deployed skeleton with a working database and health check.

Day 2Ingestion

Upload and OCR

  • Build menu upload to Supabase Storage.
  • Wire PDF.co OCR with Tesseract fallback and store raw text.

Deliverable: Upload returns extracted menu text for a PDF and a photo.

Day 3Intelligence

Extraction and tagging

  • Write GPT-4o mini extraction prompt returning strict dish JSON.
  • Add tag suggestion with confidence flags and a schema validator.

Deliverable: Any uploaded menu renders a ranked dish list with proposed tags.

Day 4Deliverable

PDF, QR page and payment

  • Build the react-pdf branded guide template.
  • Add hosted QR page route with filters and disclaimer.
  • Wire Stripe Checkout for $99 and $149 and the paid webhook.

Deliverable: Paid order produces a PDF and a live QR page URL.

Day 5LaunchMVP shipped

End-to-end and soft launch

  • Run 3 real public menus through the full pipeline.
  • Fix parsing failures and ship the public landing page.

Deliverable: Live product accepting payment, MVP complete.

Day 6Prospecting

Build the district list

  • List 30 independent restaurants in one walkable lunch district.
  • Capture each public menu PDF or photo into the system.

Deliverable: 30-row prospect sheet with menus attached.

Day 7Samples

Pre-build proof

  • Generate 10 free sample guides from prospect menus.
  • Print samples and 30 QR cards at a local print shop.

Deliverable: 10 printed samples and outreach kit ready.

Week 2 · Launch & first sale · 7 days
Day 8Sales readiness

Pitch and logistics

  • Write the 90-second counter script and a one-page price sheet.
  • Set up Resend sequences for follow-up day 1 and day 3.

Deliverable: Pitch script, price sheet and two follow-up templates.

Day 9Field sales

First visits

  • Visit 8 restaurants between 2pm and 4pm.
  • Log every conversation outcome in the prospect sheet.

Deliverable: 8 logged conversations, first warm leads.

Day 10Field salesFirst sale target

Close the first order

  • Visit 8 more restaurants and send day-1 follow-up emails.
  • Collect payment on-site via a Stripe payment link if the owner says yes.

Deliverable: First paid order.

Day 11Fulfilment

Hit the 24-hour SLA

  • Deliver the first order: checklist, PDF, QR page, email.
  • Time each fulfilment step and cut the slowest one.

Deliverable: Delivered order and a measured fulfilment time.

Day 12Field sales

Volume visits and referrals

  • Visit 10 restaurants with printed samples.
  • Ask every buyer for one neighbouring introduction.

Deliverable: 10 visits and 2 referral names.

Day 13Organic

Owner communities

  • Post a before/after sample in two local restaurant-owner groups.
  • Reply to every comment with the sample link, no hard pitch.

Deliverable: Two posts live and inbound replies logged.

Day 14Review

Week 2 metrics

  • Measure visits, samples shown, closes and fulfilment hours.
  • Fix the single biggest drop-off point in the funnel.

Deliverable: Written week-2 funnel numbers and one shipped fix.

Week 3 · Repeat & systematize · 7 days
Day 15Pipeline

Warm lead follow-up

  • Email every non-converter from days 9-12.
  • Build a second district list of 30 restaurants.

Deliverable: 30 warm follow-ups sent and a second district list.

Day 16Field sales

Second district

  • Visit 10 restaurants in the second district.
  • Show the first delivered guide as social proof.

Deliverable: 10 visits and 2-3 new paid orders.

Day 17Product

Revision workflow

  • Ship the one-revision request form and SLA timer.
  • Add an automated delivery confirmation email.

Deliverable: Self-serve revision flow and SLA status visible in admin.

Day 18Field sales

Testimonial capture

  • Visit 10 restaurants and hand out samples.
  • Ask each buyer for one sentence of real feedback on the guide.

Deliverable: 10 visits and at least 2 genuine buyer quotes.

Day 19Monetization

Catering upsell

  • Pitch the $149 catering guide to every existing $99 buyer.
  • Build the catering guide template variant.

Deliverable: Catering template live and upsell sent to all buyers.

Day 20Organic

Owner communities round two

  • Post one cuisine-specific sample guide in three owner groups.
  • Email the 30 prospect list from day 6 a second time.

Deliverable: Three posts and 30 second-touch emails sent.

Day 21Review

Mid-point verification

  • Reconcile Stripe payouts against TrustMRR records.
  • Recalculate verified profit and adjust the visit quota.

Deliverable: Mid-point profit reconciliation and adjusted plan.

Week 4 · Scale what works · 7 days
Day 22Field sales

Scale visits

  • Visit 12 restaurants with printed samples.
  • Target cafe and bakery owners who are on-site in the afternoon.

Deliverable: 12 visits and 3+ new orders.

Day 23Closing

Convert warm leads

  • Call or visit every lead that said maybe.
  • Send Stripe payment links to all pending conversations.

Deliverable: All pending leads given a payment link.

Day 24Product

Accuracy and speed

  • Add cuisine-specific tag rules for the top three cuisines seen.
  • Reduce manual cleanup time per menu and log the new average.

Deliverable: Lower average fulfilment time per order.

Day 25Field sales

Volume and upsell

  • Visit 12 restaurants and hand out samples.
  • Offer the $149 tier first to restaurants with catering signage.

Deliverable: 12 visits and at least 2 catering-tier orders.

Day 26Partnership test

Distribution experiment

  • Pitch three local business associations on a free sample guide.
  • Offer members a $79 rate in exchange for a newsletter mention.

Deliverable: 3 partnership emails sent, 1 reply minimum.

Day 27Follow-up

Third touch

  • Send a third and final follow-up to all non-responders.
  • Ask every buyer for a referral name and a neighbouring introduction.

Deliverable: All non-responders closed out with referrals captured.

Day 28Fulfilment

Batch delivery

  • Batch-generate and deliver every outstanding order.
  • Request one revision per buyer to lock in satisfaction.

Deliverable: Zero outstanding orders, SLA satisfied.

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

Close and collect

  • Visit every warm lead one final time before the deadline.
  • Collect all outstanding payments through Stripe before December 16 13:00 UTC.

Deliverable: All invoices paid and cleared.

Day 30Verification

Reconcile and retro

  • Reconcile Stripe, refunds and TrustMRR records for the final profit number.
  • Write a one-page retro on conversion rate and fulfilment cost.

Deliverable: Final verified profit figure and written retro.

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: 10 orders from roughly 60 in-person visits plus follow-up emails, all at the $99 tier.
  • Base: 25 orders, of which 3 upgrade to $149, from roughly 150 visits and 40 emails.
  • Upside: 45 orders with 9 at the $149 tier, from roughly 250 visits, referrals and owner-group inbound.
  • Payment processing assumed at Stripe standard rates, roughly 2.9% plus $0.30 per transaction, about 3.9-4.5% of collected revenue.
  • No paid marketing in Conservative or Base; Upside assumes $200 of boosted local posts and printed sample costs treated as paid acquisition.
  • Refunds modelled at 0 in Conservative because delivery is manual and pre-agreed; 1 order in Base and Upside if a menu is unparseable.
  • Operating costs cover Vercel Pro, Supabase, PDF.co OCR, GPT-4o mini inference, Resend and domain, roughly $1-2 per delivered guide plus fixed hosting.
  • All figures are projections built from stated assumptions, not verified revenue.

Conservative

10 customers
$990
collected revenue · avg payment $99
Refunds
$0
Chargebacks
$0
Payment fees
$39
Paid marketing
$0
Hackathon profit
$951
Operating costs
$55
Business profit
$896

Base case

25 customers
$2,475
collected revenue · avg payment $99
Refunds
$99
Chargebacks
$0
Payment fees
$97
Paid marketing
$0
Hackathon profit
$2,279
Operating costs
$110
Business profit
$2,169

Optimistic

45 customers
$4,770
collected revenue · avg payment $106
Refunds
$106
Chargebacks
$0
Payment fees
$215
Paid marketing
$200
Hackathon profit
$4,249
Operating costs
$180
Business profit
$4,069

Model your own numbers

Runs in your browser · nothing is saved
Collected revenue
$2,475
Payment fees (3.9%)
$97
Paid marketing
$0
Hackathon profit
$2,378
Business profit
$2,268

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 scenarios are projections derived from the stated assumptions, not verified revenue, customers or traction. Hackathon profit excludes hosting and delivery costs; business profit subtracts operating_costs_usd.

Section 7 of 8

Main risks and validation experiments

Risk 1Restaurant owners are on-site but not interruptible, so in-person visits convert far below the needed rate.

Mitigation

Visit strictly at 2-4pm, lead with a printed sample of their own menu, and keep the ask to a 90-second look.

Validation experiment

Visit 8 restaurants on day 9 and record how many owners or managers actually appear and look at the sample.

Kill criteria

If fewer than 5 of 8 visits reach an owner or manager, switch to pre-booked 10-minute slots by email.

Risk 2OCR and LLM extraction fails on real menus, so manual cleanup eats the margin and the 24-hour SLA.

Mitigation

Start from menu PDFs and website HTML rather than phone photos, and require owner confirmation on every low-confidence tag.

Validation experiment

Run 10 real prospect menus through the pipeline on days 5-7 and time the manual cleanup per menu.

Kill criteria

If average cleanup exceeds 45 minutes per menu, raise the price or restrict to PDF menus only.

Risk 3Dietary tag inaccuracy creates liability or refund pressure.

Mitigation

Owner attestation checkbox, per-tag confidence flags, one visible disclaimer on PDF and QR page, and no allergen guarantee language anywhere.

Validation experiment

Show 5 owners the checklist and ask directly whether the disclaimer is enough for them to put their name on it.

Kill criteria

If 3 or more owners refuse to attest, re-scope the output to a customer-facing preference filter without any health claim.

Risk 4$99 is too high without proof, so owners take the free sample and never pay.

Mitigation

Offer a $79 first-district rate in exchange for one referral, and always show the finished sample before naming a price.

Validation experiment

Split the first 16 visits evenly between $99 and $79 and compare close rates.

Kill criteria

If close rate stays below 10% across both prices after 30 visits, change the buyer to multi-location cafes or repackage as a $29 monthly QR page.

Risk 5The single low-value charge caps revenue and there is no repeat path.

Mitigation

Pitch the $149 catering tier to every buyer within 7 days and offer a $49 seasonal re-issue at the next menu change.

Validation experiment

Offer $149 to the first 5 buyers and count how many accept within 7 days.

Kill criteria

If zero of the first 5 buyers upgrade, stop building the catering template and reinvest the time in more visits.

Section 8 of 8

Hour-by-hour: the first 24 hours

Hours 1-2

  • Create the Next.js repo, deploy an empty app to Vercel, create the Supabase project.
  • Define the orders, dishes and tags tables with RLS enabled.

Output: Deployed app URL and a working database schema.

Hours 3-4

  • Build the menu upload route and file storage.
  • Connect PDF.co OCR and store extracted text on the order row.

Output: An upload that returns raw menu text.

Hours 5-6

  • Write the GPT-4o mini extraction prompt with strict JSON output.
  • Add schema validation and render the dish list in a review screen.

Output: First real menu converted into a structured dish list.

Hours 7-8

  • Write the dietary tagging prompt with confidence flags.
  • Add the owner verification checklist with edit, attestation and disclaimer.

Output: Working checklist screen on a real menu.

Hours 9-10

  • Build the react-pdf branded guide template with logo, tag legend and disclaimer.
  • Render one guide PDF from the day's test menu.

Output: First generated Healthy Dining Guide PDF.

Hours 11-12

  • Wire Stripe Checkout with the $99 price and the paid webhook.
  • Draft the 30-restaurant prospect list for one lunch district.

Output: Live payment link and a 30-row prospect sheet.

Complete report

Full 30-day launch report

Read the full generated report

MenuProof Guide — 30-Day Launch Report

1. Product opportunity and positioning

Independent restaurants in dense lunch districts compete for office-worker foot traffic, and part of that traffic screens menus for vegan, gluten-free, dairy-free, high-protein and low-carb options before walking in. The restaurant already holds the answer in its own menu and in the head of one line cook. What it does not have is an artifact that states it publicly. Staff answer dietary questions inconsistently at the counter, and the owner has no spare afternoon to tag forty dishes by hand.

MenuProof Guide sells a finished artifact rather than software access. The owner uploads the menu they already have. OCR and an LLM extract dishes and ingredients and propose dietary tags with confidence flags. The owner confirms or edits those tags in a five-minute checklist, accepts an attestation and disclaimer, and receives a branded one-page Healthy Dining Guide PDF plus a hosted QR page with filters within 24 hours. The guide gets printed for the host stand, taped to the counter, linked from an Instagram bio, or shown on a phone.

The buyer is the owner or manager of an independent restaurant or cafe with 10-50 menu items and 1-3 locations. The trigger is concrete: they are losing health-conscious lunch traffic, or their staff keeps fielding the same four dietary questions and answering them badly.

The wedge stays narrow on purpose. There is no POS integration, no delivery logistics and no marketplace cold start. The output is visible before the owner pays, because the founder walks in with a sample built from that restaurant's own menu. The 2-4pm window is when an owner is physically present and not managing a rush.

JEV scored this concept in band C at 60.3/100 with 0.76 confidence. Speed to first revenue (8.68 normalized) and solo build feasibility (7.69) are the strengths. Market demand (3.96), monetization potential (4.16) and operational simplicity (4.97) are the weaknesses, and the plan is built to answer those three specifically: sell in person against a free sample, open a $149 repeat-purchase path, and keep one revision with batch fulfilment.

2. Smallest sellable MVP

The smallest sellable thing is: upload a menu, get a tagged dish list, confirm tags, pay $99, receive a PDF plus QR page link by email inside 24 hours. Everything else is optional.

Must-have: upload and OCR; LLM tag suggestions with confidence flags; owner verification form with attestation; branded PDF generator; hosted QR page with filters and a permanent disclaimer; Stripe Checkout at $99 and $149; gated download; admin queue; Resend delivery; one-revision workflow.

Explicitly not built: ordering, delivery, POS integration, calorie or macro claims, allergen guarantees, a website builder, a consumer discovery app, multi-location enterprise workflow, subscriptions, or a native app. Each of those would add a week and none of them affects whether the first owner hands over $99.

3. Technical implementation plan

Next.js 14 on Vercel, Supabase Postgres with RLS, Stripe Checkout and webhooks, GPT-4o mini for extraction and tagging, PDF.co OCR with a Tesseract fallback, react-pdf for the guide, Resend for email, and object storage for menus and PDFs.

Every order is a Supabase row with a state machine: uploaded, parsed, tagged, awaiting_owner, paid, generated, delivered, revised. OCR and inference run once at upload and the dish JSON is stored, so regeneration never re-bills inference. The Stripe webhook flips an order to paid and unlocks generation. The QR page is a dynamic route keyed by a public order slug, rendering verified tags and filtering client-side. The admin queue is a password-protected list sorted by SLA age.

The hardest technical risk is OCR quality on photographed or decorative menus. Multi-column layouts and hand-lettered specials boards produce garbage text, and one bad tag in front of an owner destroys the pitch. Mitigation: start from website menus and PDFs, and force every low-confidence tag through manual owner confirmation.

4. Pricing and monetization

Menu Guide at $99 one-time: one menu up to 50 items, owner verification, branded PDF, hosted QR page, one revision, 24-hour delivery. Menu Guide + Catering Page at $149: everything above plus a separate catering guide and three QR codes for tables and counters.

The first-revenue offer is $99 for a published guide built from the current menu, delivered within 24 hours of a five-minute tag check, one revision included. Payment is collected up front through Stripe Checkout, so fulfilment is never speculative.

The upsell path is the only route to repeat revenue in the first 60 days: offer $149 to every $99 buyer within a week, and offer a $49 seasonal re-issue when the menu changes. Assume roughly one in six buyers upgrades.

5. First-customer acquisition

One channel: off-peak in-person visits in a single walkable lunch district, opening with a sample of the restaurant's own menu. Build a 30-restaurant list, capture public menus, pre-generate ten free samples, print them, and walk in between 2pm and 4pm.

The script is the artifact, not the pitch: show the tag matrix with vegan and gluten-free rows already filled in, then name the price. If the owner hesitates, leave a QR card linking to their sample and follow up by email the next morning. Ask every buyer for one neighbouring introduction, and post before/after samples in two local owner groups. Week-one goal: product live and payment working by day 5, 30 prospects listed, ten samples printed, eight visits completed.

6. 30-day execution roadmap

Days 1-5 are build: scaffold and database, upload and OCR, extraction and tagging, PDF and QR page and Stripe, then end-to-end testing and public launch. Day 5 is the MVP deadline.

Days 6-8 are preparation: 30-prospect list with menus captured, ten printed samples, 30 QR cards, pitch script, price sheet and two follow-up templates.

Days 9-14 are the first sales push. Eight visits, then eight more, first order targeted on day 10, delivery on day 11 inside the SLA, ten more visits on day 12, owner-group posts on day 13, funnel review on day 14.

Days 15-21 extend to a second district, ship the revision workflow and SLA timer, pitch $149 to existing buyers, run a second round of owner-group posts, and reconcile Stripe against TrustMRR at the midpoint.

Days 22-28 are volume plus closing: 12 visits, payment links to every hesitant lead, cuisine-specific tag rules to cut cleanup time, 12 more visits with $149 offered first to restaurants with catering signage, a three-association partnership test, a final follow-up round, and a batch fulfilment day that clears every outstanding order.

Days 29-30 close out: final visits, all payments collected before December 16 13:00 UTC, refunds reconciled, verified profit recorded, one-page retro written.

7. Revenue and profit scenarios

Projections only. Conservative: 10 orders at $99, $990 gross, $39 fees, $55 operating, zero ad spend. Base: 25 orders, $2,475 gross, one refund, $97 fees, $110 operating. Upside: 45 orders at a $106 blended price, $4,770 gross, one refund, $215 fees, $200 paid promotion, $180 operating.

Hackathon profit excludes hosting and delivery costs; business profit subtracts them. Even the conservative case returns positive hackathon profit because ad spend is zero and the deliverable is manual.

8. Main risks and validation experiments

First, owner availability: if fewer than five of eight visits reach an owner, switch to pre-booked ten-minute slots. Second, extraction quality: if manual cleanup exceeds 45 minutes per menu, restrict to PDF menus or raise the price. Third, liability fear: if three owners refuse to attest, re-scope to a preference filter with no health claim. Fourth, price resistance: split 16 visits between $99 and $79 and compare close rates. Fifth, revenue ceiling: if zero of the first five buyers take the $149 upgrade, stop building the catering template.

9. First 24-hour action plan

Hours 1-2: repo, Vercel deploy, Supabase project, tables with RLS. Hours 3-4: upload route and OCR returning raw text. Hours 5-6: extraction prompt with strict JSON and a review screen. Hours 7-8: tagging prompt with confidence flags and the owner checklist. Hours 9-10: react-pdf guide template and one rendered PDF. Hours 11-12: Stripe Checkout, paid webhook, and a 30-restaurant prospect list.

All revenue figures in this report are projections derived from stated assumptions, not verified revenue, customers or traction.

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 →