Score B · 6.50/10Free to read$100K Hackathon · November 16, 2026 → December 16, 2026Nov 15, 2026

SpecAnswer Blocks

Generates paste-in pre-purchase answer blocks — spec-grounded FAQs with FAQPage schema — for Shopify stores whose buyers keep asking 'will this fit my car / printer / lens?', so the answer exists on the product page before checkout.

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.

Promising6.50/10
MVP build (solo)
4 days
Launch price
$49
Sales to $1,000 revenue
21 customers
First paying customer
Cold email carrying the finished, watermarked artifact built from the prospect's own public product page and reviews.
Who pays

Owner-operator of a compatibility-driven Shopify store (aftermarket auto parts, replacement filters, ink and toner, camera/audio accessories, appliance parts), 1-5 people, roughly $10k-$200k monthly revenue, selling in North America or the UK.

Biggest execution risk

Monetization potential 4.7/10

Tie the price to a measurable money or time outcome and add a recurring or repeat-purchase path instead of a single low-value charge.

Why this could work in 30 days: one founder can ship the MVP in 4 days; the plan targets a first sale on day 12; 21 sales at $49 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
$49
Solo build
4 days
30d base revenue
$2,282
30d hackathon profit
$2,019
Section 1 of 8

Product opportunity and positioning

Target customer

Owner-operator of a compatibility-driven Shopify store (aftermarket auto parts, replacement filters, ink and toner, camera/audio accessories, appliance parts), 1-5 people, roughly $10k-$200k monthly revenue, selling in North America or the UK.

Why now

November-December is peak season, so compatibility tickets spike and every wrong-fit order becomes a January return; the full deliverable can be produced from public data in under 15 minutes, so cold outreach carries a finished demo instead of a pitch.

Problem

Pre-purchase compatibility questions get answered by one person, one email at a time, dozens of times a day, while the product page shows a spec table the buyer cannot interpret and the questions the store genuinely cannot answer stay invisible to the owner.

Solution

Paste a public product URL; the pipeline pulls the product page, linked spec PDFs and visible review text, clusters the recurring pre-purchase questions, answers each one with a citation to the merchant's own source line or marks it UNRESOLVED, and returns a themed HTML accordion plus FAQPage JSON-LD and a Spec Gap Report.

Differentiation · No chatbot, no storefront script, no OAuth, no store access. The merchant receives a static portable file plus a ranked list of the questions their own pages cannot answer — something a general-purpose chatbot and a support widget do not produce.

Opportunity

The buyer is the owner-operator of a compatibility-driven Shopify store: aftermarket auto parts, replacement filters, ink and toner, camera and audio accessories, appliance parts. One to five people, roughly $10k to $200k monthly revenue, North America or the UK. Their most expensive recurring task is not marketing. It is answering "will this fit my X?" dozens of times a day, personally, in email and chat.

The page they already have does not solve it. It shows a spec table the buyer cannot interpret, and the real answer usually lives in a PDF manual nobody opens. Two costs follow: abandoned carts at the low end, wrong-fit orders and January returns at the high end. A third cost is invisible — the questions the store's own sources cannot answer never surface, so the owner keeps paying for the same tickets.

Positioning

SpecAnswer Blocks sells a finished artifact, not a subscription and not a widget. The output is a themed HTML accordion plus FAQPage JSON-LD, ready to paste into a theme, with every answer cited to a line from the merchant's own page, review text or spec PDF, and every unsupported question explicitly marked UNRESOLVED. Alongside it ships a Spec Gap Report ranking the top unanswered questions by how often they appear in the store's own reviews and page text.

The commercial logic is delivery format. Because the whole deliverable is produced from public, non-authenticated data, the founder can build the complete artifact for a prospect before that prospect has replied to an email. Outreach becomes a demo: the buyer opens an email and sees their own product page, answered, with a watermark on it. There is no app-store review, no OAuth consent, no procurement, no integration call, and the price is card-approvable in one click.

This is deliberately narrower than an agentic CRM. No cart monitoring, no SMS or email negotiation with shoppers, no discount automation, no live conversation, no script running on the storefront. It is static, portable, reviewable before publishing.

Honest read of the score

JEV scored this concept 65.0/100, band B. Speed to first revenue is the strongest criterion (8.71 normalized) and monetization the weakest (4.72), with the note that a single low-value charge needs a repeat or recurring path. The plan below answers that: one-time blocks as the wedge, a $399 full-catalog tier for stores where compatibility drives support volume, and repetition through referrals and multi-page expansion rather than an invented subscription.

Section 2 of 8

Smallest sellable MVP

For one merchant product page: a watermarked-then-unlocked HTML accordion plus FAQPage JSON-LD generated entirely from public data, delivered with a 2-page Spec Gap Report by email after a Stripe payment.

Must-have features

  • Public-URL ingester: product page HTML, linked spec PDFs and visible review text, no authentication
  • Question miner clustering review and page text into 10-25 ranked pre-purchase questions
  • Grounded answer writer producing a cited answer per question or an explicit UNRESOLVED flag
  • Export engine: themed HTML accordion plus FAQPage JSON-LD, copy-to-clipboard and .zip download
  • Spec Gap Report PDF listing the top 15 unresolved questions with the triggering evidence line
  • Stripe Checkout on a TrustMRR-connected account with instant Resend delivery of the unlocked files
  • Pre-build mode producing a watermarked preview for cold outreach

Explicitly out of scope

  • No chat widget or any script running on the storefront
  • No Shopify OAuth, Admin API, order, cart or customer data
  • No email or SMS sent to the merchant's shoppers
  • No auto-publishing to the merchant's theme
  • No multi-language output, analytics dashboard or A/B testing
  • No subscription tier or monthly refresh during the 30 days
  • No team accounts, roles or white-label portal

Demo flow (what the buyer sees)

  1. Founder runs pre-build mode on a prospect's public product URL
  2. Ingester extracts page text, spec PDF text and visible review text into source lines
  3. Question miner returns 10-25 ranked pre-purchase questions
  4. Answer writer attaches a citation to each answer, or marks it UNRESOLVED
  5. Watermarked accordion and Spec Gap Report PDF are generated
  6. Founder emails both to the store owner with the top unresolved question as the subject hook
  7. Buyer pays $149 by card; the same pipeline output is re-issued unwatermarked within minutes
  8. Export zip plus gap report lands in the buyer's inbox, with a 3-step paste guide
Build plan

Technical implementation plan

Stack

Next.js on VercelSupabase Postgres for jobs and source lines, Supabase Storage for blocks and PDFsFetch plus Readability for page extraction; a PDF text extraction library for manualsOpenAI or Anthropic API for question clustering and citation-grounded answersStripe Checkout on a TrustMRR-connected accountResend for delivery emailReact-PDF or headless Chrome for the Spec Gap ReportA queue table plus a cron worker so long builds never block a request

Architecture

A single Next.js app serves a private build form and a public order page. A build job is written to Supabase, then a cron-invoked worker fetches the product page, follows linked PDFs, extracts and chunks the text into numbered source lines, and stores them. The miner sends chunked review and page text to the model, asking only for clustered question strings. The writer is called once per question with the retrieved source lines and must return either an answer plus a source-line id or the literal token UNRESOLVED. A renderer turns verified answers into a scoped-CSS accordion and a JSON-LD FAQPage block, applies a watermark when the job belongs to a prospect, and writes the zip and PDF to storage. Stripe webhook marks the job paid, the worker re-renders without the watermark, and Resend sends the download links.

Build steps

  1. Scaffold Next.js on Vercel, create the Supabase project, set environment secrets
  2. Build the ingester: fetch URL, extract readable text, discover and parse linked spec PDFs, persist numbered source lines
  3. Build the question miner over review and page text, returning clustered questions with frequency counts
  4. Build the grounded answer writer with line-id citation and an explicit UNRESOLVED path
  5. Build the renderer: scoped-CSS accordion, FAQPage JSON-LD, watermark toggle, zip packaging
  6. Build the Spec Gap Report PDF generator from the unresolved set
  7. Wire Stripe Checkout and the TrustMRR-connected account, plus a webhook that triggers unwatermarked re-render
  8. Wire Resend delivery emails with signed download links
  9. Add an internal pre-build page that takes a URL and returns preview plus PDF in under 15 minutes
  10. Test the pasted block inside a Shopify development theme and fix any style bleed

AI components

  • Question clustering and normalisation across noisy review phrasings
  • Retrieval-augmented answer writing constrained to merchant source lines
  • A verification pass that rejects answers whose cited line does not support them
  • Optional product-page summarisation used only to build the source-line index
Hardest technical risk: Citation fidelity: the model will occasionally invent a plausible spec line or cite a line that does not contain the answer, which turns a delivered block into a liability. Mitigation is a mandatory line-id return, a server-side check that the cited text exists and shares key tokens with the question, and a hard UNRESOLVED fallback.
Section 3 of 8

Pricing and monetization

Gap Report

Entry offer
$49 one-time

For Stores that want the diagnosis before publishing anything

  • Top 15 unresolved pre-sale questions ranked
  • The exact review or page line evidencing each one
  • Recommended spec additions in priority order

Answer Block

$149 one-time

For Stores ready to publish answers on their product pages

  • Paste-in HTML accordion plus FAQPage JSON-LD for up to 25 product pages
  • Every answer cited to the merchant's own spec line
  • Full Spec Gap Report
  • One regeneration after the merchant edits or adds specs

Full Catalog

$399 one-time

For Stores with 100+ SKUs where compatibility drives support volume

  • Blocks for up to 150 product pages with SKU-to-block mapping
  • CSV index so the merchant can find each block by SKU
  • 90-day refresh of any block whose spec text changed

Model: One-time per-deliverable pricing in three tiers, paid by card at the moment the artifact is already built.

First revenue offer: $149 for the unwatermarked block on up to 25 product pages, already built and previewed in the email; payment removes the watermark and unlocks the export. A $99 code is available for 48 hours on the day-3 follow-up.

Expected ACV: $163

Upsell path: Move any buyer with more than 50 compatibility SKUs to the $399 Full Catalog with a SKU-mapped CSV index, then collect one named referral per buyer and run the pipeline on that store the same day. Repeat purchases come from new product lines and seasonal catalog refreshes, not from a subscription invented before demand exists.

Target customers
90
Revenue model
$13,410
Costs
$620
Profit model
$12,790
Section 4 of 8

First-customer acquisition

Primary channel

Cold email carrying the finished, watermarked artifact built from the prospect's own public product page and reviews.

Where to find them

  • Public product pages in aftermarket auto parts with fitment tables or 'fits' language
  • Replacement filter and consumable stores (HVAC, water, vacuum) with dense spec tables
  • Ink and toner stores where cartridge compatibility is the main support driver
  • Camera and audio accessory stores with lens-mount and impedance compatibility text
  • Appliance parts stores with model-number matching
  • Shopify stores surfaced by searching site:myshopify.com plus category terms
  • Stores whose public reviews contain 'does this fit', 'will it work with', 'compatible with'
  • Existing buyers' introductions to other stores in the same category

Exact outreach motions

  • Build the watermarked block and gap report before sending anything; the artifact is the pitch
  • Send under 40 emails a day from a warmed domain, plain text, with the preview link and PDF attached
  • Subject angle: 'Your page cannot answer this — your own reviews ask it'
  • Body is three lines: the single most-asked unresolved question, the evidence line from their own review text, and a link to the preview
  • Day 3 follow-up names a second unresolved question and includes a 48-hour $99 code
  • Day 7 follow-up unlocks the full gap report as a free PDF with no code
  • After a sale, ask for one introduction to a store in the same category and run the pipeline the same day

Message template

Subject: Your page cannot answer this — your own reviews ask it

Hi {first_name} — on {product_url} buyers keep asking "{top_question}". Your own review from {review_month} says: "{evidence_line}".

I already built the answer block and a 2-page gap report for that page: {preview_link}

$149 removes the watermark and gives you the paste-in block for up to 25 pages. No app install, no store access needed. Want it?
Week-one goal: 200-store prospect list logged with a specific worst-served product URL each, 20 watermarked pre-builds generated, 50 cold emails sent, and at least 2 replies that name a price or ask how to publish.
Section 5 of 8

30-day execution roadmap

MVP deadline: day 5First sale target: day 1230 days planned
Week 1 · Build & validate · 7 days
Day 1Foundation

Repo, hosting, payments account

  • Register the domain and create the Next.js plus Vercel project
  • Create the Supabase project and storage buckets
  • Confirm the Stripe account is connected to TrustMRR before anything ships

Deliverable: Deployed skeleton with a private build form and a working Supabase connection

Day 2Ingestion

Public data extraction

  • Build the URL fetcher with Readability extraction
  • Add linked-PDF discovery and text extraction
  • Persist numbered source lines per build job

Deliverable: Ingester returns 40+ numbered source lines from three real test product pages

Day 3Intelligence

Question miner and grounded writer

  • Cluster review and page text into ranked questions
  • Write answers constrained to source lines with line-id citations
  • Enforce the UNRESOLVED fallback and a server-side citation check

Deliverable: End-to-end run producing 15 cited questions and 3 UNRESOLVED on a real page

Day 4Export

Paste-ready artifact

  • Render the scoped-CSS accordion and FAQPage JSON-LD
  • Add watermark toggle, zip packaging and copy-to-clipboard
  • Test the pasted block in a Shopify development theme

Deliverable: One paste-ready block verified inside a live theme

Day 5CommerceMVP shipped

Checkout and delivery

  • Wire Stripe Checkout with the three tiers and a TrustMRR-connected account
  • Add the webhook that re-renders without watermark on payment
  • Send the delivery email via Resend with signed download links

Deliverable: Test purchase delivers the unwatermarked zip and PDF in under 60 seconds

Day 6List building

Prospect sourcing

  • Find 60 compatibility-driven Shopify stores via category and site: searches
  • Log store, worst-served product URL and a contact email per row
  • Prioritise stores whose reviews contain fitment questions

Deliverable: 60-row prospect sheet with a specific product URL per store

Day 7Pre-builds

Artifact production at volume

  • Run pre-build mode on 20 stores across four categories
  • Generate the Spec Gap Report PDF template
  • Record pipeline time and source-line yield per store

Deliverable: 20 watermarked previews plus 20 gap reports ready to attach

Week 2 · Launch & first sale · 7 days
Day 8First send

Outreach begins

  • Send 20 cold emails with the preview link and PDF
  • Log opens, bounces and replies in the sheet
  • Check SPF, DKIM and DMARC on the sending domain

Deliverable: 20 sends logged with deliverability confirmed

Day 9Volume

List to 120 and send

  • Extend the prospect sheet to 120 stores
  • Send 30 more emails using the same angle
  • Cap daily sends at 40 to protect domain reputation

Deliverable: 50 total sends, bounce rate under 5%

Day 10Reply handling

Convert replies

  • Answer every reply within 2 hours with the payment link
  • Offer to run one more page free for anyone hesitating
  • Fix any block that rendered badly in a buyer's theme

Deliverable: Five warm replies and the first payment link clicked

Day 11Follow-up one

Day-3 touch

  • Send the day-3 follow-up with a second unresolved question
  • Attach the 48-hour $99 code to the first 20 prospects
  • Send 30 fresh first-touch emails

Deliverable: 80 cumulative sends, follow-up sequence running

Day 12First saleFirst sale target

Close and deliver

  • Close the warmest prospect with a live preview walkthrough
  • Collect payment through the TrustMRR-connected Stripe account
  • Deliver the unwatermarked zip within one hour of payment

Deliverable: First collected payment inside the contest window

Day 13Fulfilment

Delivery loop

  • Ship the paid order with a 3-step paste guide
  • Ask the buyer to confirm the block renders on their theme
  • Request one introduction to a store in the same category

Deliverable: One delivered order, one confirmed render, one referral ask sent

Day 14Copy iteration

Reply-rate repair

  • Review subject-line and first-line performance against replies
  • Write variant B around the revenue-recovery framing
  • Skip any store with fewer than 15 usable source lines

Deliverable: Variant B email copy and a filtered prospect segment

Week 3 · Repeat & systematize · 7 days
Day 15Variant test

Send variant B

  • Send 40 emails using variant B
  • Tag replies by variant in the sheet
  • Kill whichever variant underperforms on reply rate

Deliverable: 120 cumulative sends with a winning variant identified

Day 16Upsell build

Full Catalog tier

  • Build the SKU-to-block CSV mapping and 150-page batch runner
  • Publish the $399 Full Catalog checkout link
  • Write the one-paragraph pitch for stores with 50+ compatibility SKUs

Deliverable: $399 tier live and deliverable end to end

Day 17Upsell pitch

Raise order value

  • Pitch Full Catalog to any buyer or warm lead with a large SKU count
  • Offer to map the first 25 SKUs free as proof
  • Log objections verbatim

Deliverable: At least one $399 conversation in progress

Day 18Follow-up two

Day-7 touch

  • Send the day-7 follow-up unlocking the full gap report free
  • Send 40 fresh first-touch emails
  • Re-run pre-builds for any store whose catalog changed

Deliverable: 160 cumulative sends, gap reports opened by non-buyers

Day 19Public proof

Inbound surface

  • Publish one anonymised store teardown showing an unanswered fitment question
  • Add the preview link and price to the teardown page
  • Share the teardown from the founder account, no paid promotion

Deliverable: Live teardown page with a shareable URL

Day 20Throughput

Send and triage

  • Send 40 emails to the ink, toner and appliance parts segment
  • Triage replies into buy, later and no
  • Send payment links to every 'later'

Deliverable: 200 cumulative sends across four categories

Day 21Midpoint review

Conversion math

  • Compute reply rate, reply-to-paid rate and revenue per 100 sends
  • Identify the single most common objection and rewrite the offer around it
  • Decide whether to push $149 volume or $399 catalog deals for the remainder

Deliverable: One-page midpoint scorecard and a revised offer sheet

Week 4 · Scale what works · 7 days
Day 22New segment

Fresh prospects

  • Build 60 new prospect rows in untested categories
  • Send 40 first-touch emails to the new segment
  • Pre-build 10 previews for the strongest rows

Deliverable: 60 fresh rows and 40 sends with previews attached

Day 23Referrals

Warm introductions

  • Ask every buyer for one named introduction
  • Run the pipeline on each introduced store the same day
  • Send the introduced store their own preview at no charge

Deliverable: At least three warm introductions with previews delivered

Day 24Reactivation

Non-openers

  • Re-send to non-openers with a new subject line
  • Offer the $99 code with a hard 48-hour expiry
  • Send 40 fresh first-touch emails

Deliverable: Reactivation batch sent and expiry tracked

Day 25Refund defence

Delivery quality

  • Add a 'verified in theme' checklist email to every order
  • Ship a Liquid snippet variant for themes that break pasted HTML
  • Audit 10 delivered blocks for citation accuracy

Deliverable: Checklist email live and zero citation errors in the audit

Day 26Volume

Produce and send

  • Batch-build 10 previews for the newest segment
  • Send 40 emails with previews attached
  • Log every build's pipeline time

Deliverable: 10 previews built under 15 minutes each and 40 sends logged

Day 27Close push

Warm pipeline

  • Email every warm lead with a same-day delivery promise
  • Offer the $399 catalog price to any lead with 50+ SKUs
  • Confirm delivery can complete before December 16

Deliverable: All warm leads contacted with a dated delivery commitment

Day 28Final send

Last outreach wave

  • Send the final 40 first-touch emails
  • Send a last-call note to every non-buyer who opened
  • Stop all new builds that cannot be delivered before the deadline

Deliverable: Outreach closed with a full send and reply ledger

Days 29-30 · Consolidate · 2 days
Day 29Fulfilment

Clear the backlog

  • Deliver every outstanding paid order
  • Confirm each buyer received and pasted their block
  • Refund anything that cannot be delivered before the deadline

Deliverable: Zero undelivered orders

Day 30Accounting

Final tally

  • Reconcile gross revenue, refunds, chargebacks and processing fees in Stripe
  • Record verified profit and submit the contest entry
  • Write down the next 30-day plan from the real conversion numbers

Deliverable: Final verified profit figure and submitted entry

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: 3 sales in 30 days at a blended $149, mostly the Answer Block tier, with a cold-email reply rate under 3%.
  • Base: 14 sales at a blended $163, mixing $49 Gap Reports, $149 Answer Blocks and two $399 Full Catalog orders.
  • Upside: 32 sales at a blended $178 including about five Full Catalog orders, which requires a 6%+ reply rate and roughly 25% reply-to-paid conversion.
  • Payment processing assumed at Stripe 2.9% + $0.30 per transaction, modelled here as about 5% of gross revenue.
  • No paid marketing in any scenario: ad_spend_usd is 0 because acquisition is organic cold email carrying a pre-built artifact.
  • Operating costs cover Vercel, Supabase, Resend and model API calls, at roughly $1.50-$2.00 of inference per store build.
  • Refunds modelled at 0-5% of gross because every answer is cited and reviewed by the merchant before pasting.
  • Chargebacks modelled at 0: prices are $49-$399, delivered instantly by email, with a stated refund policy.
  • Hackathon profit excludes operating costs; business profit subtracts them. Both are derived downstream from these inputs.
  • All figures are projections built from stated assumptions, not verified revenue, existing customers or observed traction.

Conservative

3 customers
$447
collected revenue · avg payment $149
Refunds
$0
Chargebacks
$0
Payment fees
$22
Paid marketing
$0
Hackathon profit
$425
Operating costs
$35
Business profit
$390

Base case

14 customers
$2,282
collected revenue · avg payment $163
Refunds
$149
Chargebacks
$0
Payment fees
$114
Paid marketing
$0
Hackathon profit
$2,019
Operating costs
$90
Business profit
$1,929

Optimistic

32 customers
$5,696
collected revenue · avg payment $178
Refunds
$298
Chargebacks
$0
Payment fees
$285
Paid marketing
$0
Hackathon profit
$5,113
Operating costs
$220
Business profit
$4,893

Model your own numbers

Runs in your browser · nothing is saved
Collected revenue
$686
Payment fees (3.9%)
$27
Paid marketing
$0
Hackathon profit
$659
Business profit
$569

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.

Scenario models from stated assumptions, not verified revenue. All three cases are projections only; no customers, payments or traction exist at the start.

Section 7 of 8

Main risks and validation experiments

Risk 1The buyer concludes a free general chatbot can do this and refuses to pay $149.

Mitigation

Lead with the citation discipline and the ranked Spec Gap Report, not with the answers, and show the watermark proof that the artifact already exists.

Validation experiment

Split the first 80 sends between a 'cited answers' subject and a 'your own reviews ask this' subject and compare replies that mention price.

Kill criteria

If 150 artifact-carrying sends produce zero replies mentioning price or publishing, stop and re-wedge before spending the remaining days.

Risk 2Cold email deliverability collapses and reply rates fall to near zero within days.

Mitigation

Warm the sending domain, configure SPF, DKIM and DMARC, cap sends at 40 per day, keep messages plain text and link to previews rather than attaching large files.

Validation experiment

Track bounce, open and reply per 40-send batch from day 8 to day 11.

Kill criteria

If bounce exceeds 5% or reply rate stays under 1% after 100 sends, change domain and channel tactics before continuing.

Risk 3Reviews render client-side or spec manuals sit behind a login, so the pipeline cannot mine questions for the best-fit stores.

Mitigation

Fall back to page text and the spec table, score each store on source-line yield, and skip low-evidence stores rather than shipping a thin block.

Validation experiment

Run the pipeline on the first 20 stores and record usable source lines and resolved-question count per store.

Kill criteria

If more than 60% of top-fit stores yield under 15 usable source lines, narrow to categories that publish specs in page HTML.

Risk 4A wrong compatibility answer causes a return and a refund, damaging the product's core claim.

Mitigation

Cite a source line on every answer, keep an explicit UNRESOLVED state, verify the cited line server-side, and require merchant review before pasting.

Validation experiment

Audit 10 delivered blocks and count answers whose citation does not support them.

Kill criteria

If two refunds in the first ten sales cite a wrong answer, add a mandatory human review step before delivery.

Risk 5Some Shopify themes break pasted HTML blocks, producing refund requests in week one.

Mitigation

Ship scoped-CSS with a Liquid snippet variant, test in a development theme, and send a 'verified in theme' checklist with every order.

Validation experiment

Ask the first five buyers to confirm the block renders and to name their theme.

Kill criteria

If more than 30% of buyers report a broken render, switch the default deliverable to the Liquid snippet variant.

Risk 6Peak season means merchants are too busy to implement, even though their ticket volume is highest.

Mitigation

Promise a paste-ready file that takes under 10 minutes, and offer to pre-fill the first page for them at no extra charge.

Validation experiment

Track how many buyers paste within 48 hours of delivery versus deferring to January.

Kill criteria

If fewer than half of buyers implement within 48 hours, shift the offer toward the $49 Gap Report as the primary purchase.

Section 8 of 8

Hour-by-hour: the first 24 hours

Hours 1-2

  • Register the domain and create the Next.js plus Vercel project
  • Create the Supabase project and confirm Stripe is connected to TrustMRR

Output: Deployed skeleton, repository and a live payments account ready for the contest window

Hours 3-5

  • Build the fetcher with Readability extraction
  • Add linked spec PDF text extraction and numbered source lines
  • Test on three real product pages across different categories

Output: Numbered source-line JSON for three real product pages

Hours 6-8

  • Implement question clustering over review and page text
  • Implement citation-grounded answers with a line-id return and UNRESOLVED fallback

Output: 15 ranked questions with citations and 3 explicit UNRESOLVED flags on a real page

Hours 9-11

  • Render the scoped-CSS accordion and FAQPage JSON-LD
  • Add watermark toggle and zip packaging
  • Paste the output into a Shopify development theme

Output: One paste-ready block rendering correctly inside a live theme

Hours 12-14

  • Seed a 20-row prospect sheet with one worst-served product URL each
  • Draft the three-line outreach email and the $149 offer

Output: 20-row prospect sheet plus send-ready email copy

Complete report

Full 30-day launch report

Read the full generated report

SpecAnswer Blocks — 30-Day Launch Report

Concept: SpecAnswer Blocks, version 1. JEV score: 6.5 weighted, 65.0/100, band B, confidence 0.71. Window: Monday 16 November 2026 13:00 UTC to Wednesday 16 December 2026 13:00 UTC. Builder: one person, no team, no funding, no brand.

1. Product opportunity and positioning

The buyer is the owner-operator of a compatibility-driven Shopify store — aftermarket auto parts, replacement filters, ink and toner, camera and audio accessories, appliance parts — one to five people, roughly $10k to $200k monthly revenue. Their most expensive recurring task is answering "will this fit my X?" personally, dozens of times a day. Two costs follow: abandoned carts at the low end, wrong-fit orders and January returns at the high end. A third cost is invisible — the questions the store's own sources cannot answer never surface, so the same tickets repeat.

The wedge is delivery format. The merchant pastes a public product URL. The pipeline pulls the page, linked spec PDFs and visible review text, clusters the recurring pre-purchase questions, and answers each one with a citation to the merchant's own source line — or marks it UNRESOLVED. The output is a themed HTML accordion plus FAQPage JSON-LD and a Spec Gap Report ranking what the store cannot answer.

Why this is sellable now: November–December is the trigger event, and the entire deliverable can be produced from public data in under 15 minutes, before the merchant has replied to an email. Cold outreach becomes a demo. Differentiation against a general chatbot is citation discipline and a ranked gap list; differentiation against a live chat app is that nothing is installed and no script runs on the storefront.

2. Smallest sellable MVP

One product page, one paste, one cited answer set. The minimum that can take money is: a public-URL ingester, a question miner that returns 10–25 ranked questions, a grounded writer that cites or refuses, an export that produces accordion plus JSON-LD, a gap-report PDF, and a Stripe checkout on a TrustMRR-connected account that unlocks the unwatermarked files by email within a minute.

Everything else is out of scope for 30 days: no chat widget, no OAuth, no Admin API, no messages to the merchant's shoppers, no auto-publish, no dashboard, no subscription, no multi-language. Those are not deferred features; they are the reasons the product is buyable without a procurement conversation.

3. Technical implementation plan

Next.js on Vercel, Supabase for jobs, source lines and file storage, Fetch plus Readability for pages, a PDF text extractor for manuals, an LLM for clustering and grounded answers, Stripe Checkout, Resend for delivery, and a cron worker so long builds never block a request. A server-side check verifies that every cited line id exists and shares key tokens with the question; failure forces UNRESOLVED. That check is the product's integrity guarantee and the hardest thing to get right — the model will otherwise invent a plausible spec line, and one invented line turns a delivered block into a liability.

4. Pricing and monetization

Three one-time tiers: $49 Gap Report, $149 Answer Block for up to 25 pages, $399 Full Catalog for up to 150 pages with a SKU-mapped CSV. The offer sent this week is $149, on an artifact already built and previewed in the email, with a 48-hour $99 code on the day-3 follow-up. Expected blended revenue per customer is about $163. JEV's weakest criterion for this concept is monetization, and the answer is not to invent a subscription before demand exists: it is to move stores with 50+ compatibility SKUs onto the $399 tier and to earn repeat orders from new product lines and referrals.

5. First-customer acquisition

One channel: cold email that carries the finished, watermarked artifact. Build a 200-store list from public search results — fitment tables, cartridge compatibility, lens-mount text, appliance model matching — and log the single worst-served product URL per store. Run pre-build mode on the strongest 20. Send under 40 emails a day from a warmed domain, plain text, with a preview link. The body is three lines: the most-asked unresolved question, the evidence line from their own review text, and the link. Follow up on day 3 with a second question and the $99 code, and on day 7 with the unlocked gap report. Week-one target: 200 rows, 20 pre-builds, 50 sends, two replies that mention price.

6. 30-day execution roadmap

Days 1–5 (build): scaffold and payments account on day 1, ingester on day 2, question miner and grounded writer on day 3, export and theme test on day 4, Stripe plus Resend delivery on day 5. MVP is live at the end of day 5.

Days 6–14 (first revenue): 60-store list on day 6, 20 pre-builds on day 7, first 20 sends on day 8, list to 120 and 30 more sends on day 9, reply handling and the first payment link on day 10, day-3 follow-ups with the $99 code on day 11, first sale target on day 12, delivery and referral ask on day 13, copy iteration to variant B on day 14.

Days 15–21 (scale): variant B at 40 sends per day, the $399 Full Catalog batch runner and CSV mapping built on day 16 and pitched on day 17, day-7 follow-ups on day 18, one public anonymised teardown on day 19, 200 cumulative sends by day 20, midpoint scorecard on day 21 that decides whether to push volume or catalog deals.

Days 22–30 (harvest): new segments and fresh prospects, a referral push that runs the pipeline the same day an introduction arrives, non-opener reactivation with a hard-expiring $99 code, a refund-defence checklist and Liquid snippet variant, and a final send wave. Day 29 clears every outstanding order; day 30 reconciles revenue, refunds, chargebacks and fees, and submits.

7. Revenue and profit scenarios

All three cases are projections from stated assumptions. No customers, payments or traction exist at the start.

CaseCustomersBlended ARPUGrossRefundsFeesAd spendOperating
Conservative3$149$447$0$22$0$35
Base14$163$2,282$149$114$0$90
Upside32$178$5,696$298$285$0$220

Hackathon profit is gross minus refunds, chargebacks, processing fees and paid marketing; hosting, domains, tooling and inference are not deducted. Business profit subtracts operating costs as well. Processing is modelled at roughly 5% of gross on Stripe. There is no paid marketing in any case, because acquisition is organic cold email. Even the upside case is an order of magnitude below the $100,000 target — this is a one-time-artifact business, and the honest 30-day goal is a validated paid channel and a handful of verified dollars, not a six-figure profit.

8. Main risks and validation experiments

Free-alternative risk. A buyer may decide a general chatbot is good enough. Test: two subject angles across 80 sends; kill if 150 sends produce no price conversation.

Deliverability risk. Reply rates can collapse in days. Test: bounce and reply per 40-send batch; kill if bounce exceeds 5% or reply stays under 1% after 100 sends.

Data-access risk. Some stores hide reviews or manuals. Test: source-line yield across the first 20 stores; kill if more than 60% of top-fit stores fall under 15 usable lines.

Accuracy risk. One wrong compatibility answer creates a return. Test: audit 10 delivered blocks for citation support; add mandatory review if two of the first ten sales refund on accuracy.

Theme friction risk. Pasted HTML can break. Test: ask the first five buyers to confirm rendering; switch the default to a Liquid snippet if more than 30% report breakage.

9. First 24-hour action plan

Hours 1–2: domain, repo, Vercel, Supabase, and confirmation that Stripe is connected to TrustMRR. Hours 3–5: ingester working on three real product pages. Hours 6–8: question miner plus grounded writer producing 15 cited questions and 3 UNRESOLVED. Hours 9–11: accordion, JSON-LD and zip, pasted successfully into a Shopify development theme. Hours 12–14: a 20-row prospect sheet and the three-line outreach email ready to send.

By the end of day one the founder should be able to hand a stranger their own product page, answered and watermarked. Everything after that is repetition.

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 →