Score B · 6.91/10Free to read$100K Hackathon · November 16, 2026 → December 16, 2026Nov 15, 2026

Proofwright

Proofwright runs the customer interview on autopilot and returns a publish-ready case study draft, pull quotes and social copy, white-labeled under the writer's own brand.

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.

Promising6.91/10
MVP build (solo)
5 days
Launch price
$79 · monthly
Sales to $1,000 revenue
13 customers
First paying customer
LinkedIn DMs to self-identified freelance case-study writers and small content-agency owners.
Who pays

Solo B2B case-study writers, freelance content strategists and 2-10 person content agencies who sell case studies as a fixed deliverable and are reachable on LinkedIn, Superpath Slack and X.

Biggest execution risk

Operational simplicity 5.3/10

Why this could work in 30 days: one founder can ship the MVP in 5 days; the plan targets a first sale on day 11; 13 sales at $79 clears $1,000 of collected revenue; strongest signal is speed to first revenue at 8.0/10.

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

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

0/15
Entry price
$79
Solo build
5 days
30d base revenue
$1,071
30d hackathon profit
$1,023
Section 1 of 8

Product opportunity and positioning

Target customer

Solo B2B case-study writers, freelance content strategists and 2-10 person content agencies who sell case studies as a fixed deliverable and are reachable on LinkedIn, Superpath Slack and X.

Why now

The buyer already sells this exact artifact at a fixed price with a deadline this week, approves $79 to $199 without procurement, and judges the tool on a single edit pass.

Problem

Every case study eats a scheduling round-trip with the client's customer, a 45-minute call, transcript cleanup and a blank-page first draft - the least billable hours in the engagement, and clients stall because their own customers are hard to book.

Solution

The writer sends a branded interview link; a text-first AI interviewer runs a fixed 12-question script, forces concrete answers, and returns a 700-word draft in the writer's template plus 5 pull quotes, 3 stat callouts, a headline and a LinkedIn post, with every quote traceable to the transcript.

Differentiation · Generic AI writers produce prose you must fact-check; Proofwright produces prose you can verify, because every quote carries a span back to the transcript and unsourced sentences are flagged.

1. Product opportunity and positioning

The buyer is one person under deadline: a solo B2B case-study writer, or the owner of a two-to-ten person content shop selling case studies at a fixed price. Their day is not blocked by writing. It is blocked by booking a customer interview through a client, then through that client's customer, sometimes twice. Every slipped booking is unbilled time and a slipping deadline, and the case study is the least billable artifact in the engagement.

Proofwright compresses that funnel into a link. The writer brands it, sends it, and the customer answers a twelve-question interview in a browser at 11pm if that is when they have five minutes. The AI moderator refuses vague answers and re-asks until it captures a number, an alternative the customer tried, or the objection someone had to overcome - the exact material a case study cannot be written without. The output is not a chat log. It is a 700-word draft in the writer's template, five pull quotes, three stat callouts, a headline and a LinkedIn post.

The defensible detail is verification, not generation. Generic AI writing tools produce plausible prose that a careful professional must fact-check line by line, which eats the time saved. Here every pull quote carries a span back to the transcript and the editor flags any sentence with no source, so the writer does one edit pass instead of a truth audit. That single mechanic is what turns an AI draft from a liability into something a professional will put their name on.

White-label is the second lever. The client's customer never sees Proofwright, only the writer's logo and colours on a subdomain. The writer can therefore resell the speed as their own process - I turn these around in 48 hours - which is a sales argument they can charge for, not just a convenience.

This concept scores 69.1/100 (band B) under JEV. Speed to first revenue is its strongest sub-score and operational simplicity its weakest, so this plan front-loads a sellable artifact and strips every surface that is not the interview-to-draft loop. The riskiest assumption is not demand; it is whether the writer's end customer will finish a link-based interview at all.

Section 2 of 8

Smallest sellable MVP

One branded interview link that produces one complete case-study package - 700-word draft in the writer's template, 5 pull quotes, 3 stat callouts, headline and LinkedIn post - delivered within 24 hours of the customer finishing.

Must-have features

  • Branded interview page on a subdomain with logo and colours
  • Fixed 12-question text interview with AI follow-ups that demand concrete detail
  • Optional browser voice note transcribed server-side
  • Draft generator emitting draft, quotes, stats, headline and LinkedIn post as structured data
  • Editor with inline quote highlighting and a verify-quote toggle flagging untraceable sentences
  • One-click export to Markdown, PDF and Google Doc
  • Stripe Checkout with credit metering per plan
  • Dashboard statuses: sent, in progress, completed, draft ready

Explicitly out of scope

  • Outbound AI phone calls
  • HubSpot or Salesforce integration
  • Team seats, roles or permissions
  • Non-English support
  • Template marketplace or custom template builder
  • Analytics or benchmarking dashboards
  • Billing by seat or usage overage

Demo flow (what the buyer sees)

  1. Writer pastes their client's customer email and picks a template
  2. System emails a branded interview link valid for 14 days
  3. Customer answers 12 questions; the AI re-asks vague answers until one concrete detail lands
  4. On completion the writer gets an email and a draft-ready row in the dashboard
  5. Draft package renders with highlighted quotes and flagged unsourced sentences
  6. Writer edits, exports to Google Doc or PDF, and sends to their client
Build plan

Technical implementation plan

Stack

Next.js on VercelSupabase (Postgres, auth, storage)OpenAI API for follow-ups and draftingWhisper API for voice notesStripe Checkout and webhooksResend for transactional email

Architecture

A public interview route keyed by an opaque slug streams answers to the OpenAI follow-up loop and writes each turn to Postgres. On completion a single structured generation call produces the draft package; quote text is stored with character offsets into the stored transcript so highlighting and verification are deterministic rather than a second model guess. The writer's dashboard reads drafts and triggers exports; Stripe webhooks flip subscription status and credits gate new interviews.

Build steps

  1. Deploy Next.js skeleton and migrate Supabase schema (writers, interviews, turns, drafts, credits)
  2. Add magic-link auth and an empty dashboard shell
  3. Encode the 12-question script as JSON and build the streaming interview route
  4. Add follow-up logic with a 20-turn cap and finish-early escape
  5. Persist every turn and the final transcript
  6. Generate the draft package in one JSON call and store quote offsets
  7. Build the editor, highlight layer and verify-quote toggle
  8. Add Markdown, PDF and .docx export plus subdomain branding settings
  9. Wire Stripe Checkout, credits and the Resend email sequence
  10. Load-test with two seeded transcripts and fix prompt regressions

AI components

  • OpenAI chat model for follow-up questioning and answer specificity checks
  • OpenAI chat model in structured JSON mode for the draft, quotes, stats, headline and social copy
  • Whisper for optional voice-note transcription
  • No fine-tuning; all behaviour lives in prompts and the JSON script
Hardest technical risk: A runaway or truncated interview: if the model loops or the customer abandons at question 9, the draft quality collapses. Mitigate with a 20-turn cap, two re-asks maximum per question, and partial-transcript drafting.
Section 3 of 8

Pricing and monetization

Solo

Entry offer
$79 / month

For Individual freelance writers running 1-3 case studies a month

  • 3 case-study drafts per month
  • 5 pull quotes and 3 stat callouts per draft
  • Quote-verification view
  • Markdown and PDF export

Studio

$199 / month

For Content agencies and busy freelancers running 4-12 case studies a month

  • 12 case-study drafts per month
  • White-label custom subdomain and branding
  • Google Doc export
  • LinkedIn post copy per case study
  • Priority draft generation

Model: Monthly subscription with credit-metered drafts; no seats, no usage overage, no setup fee.

First revenue offer: One free completed case study on a real client: send the interview link, receive the 700-word draft, 5 quotes and LinkedIn copy within 24 hours of completion. If the edit pass takes under 20 minutes, Solo is $79/month.

Expected ACV: $119

Upsell path: Writers who hit the 3-draft ceiling in month one move to Studio at $199 for white-label plus 12 drafts; agencies start on Studio directly. No new tiers inside the first 30 days.

Target customers
20
Revenue model
$3,980
Costs
$140
Profit model
$3,840
Section 4 of 8

First-customer acquisition

Primary channel

LinkedIn DMs to self-identified freelance case-study writers and small content-agency owners.

Where to find them

  • LinkedIn search: case study writer
  • LinkedIn search: B2B content strategist freelance
  • LinkedIn: owners of 2-10 person content agencies listing case studies
  • Superpath Slack case-study and freelance channels
  • X search for writers posting about case-study backlogs
  • People who published a case study for a named B2B SaaS in the last 90 days

Exact outreach motions

  • Day 6 build a 60-row sheet: name, role, agency size, a case study they published, likely hours per study
  • Day 7 send 20 DMs with the free-run offer and no attachment
  • Day 8 send 20 more plus the day-3 follow-up on unanswered threads
  • Day 9 send the last 20 plus a Superpath channel post of the same offer
  • Day 10-11 deliver finished free-run drafts within 24 hours and follow with the edit-diff screenshot and Stripe link
  • Day 13 pitch 10 agency owners the Studio white-label angle specifically

Message template

Hey [name] - saw you write case studies for B2B SaaS. The bottleneck is usually booking the customer interview, not the writing. I built a tool that runs that interview from a branded link and hands back a 700-word draft plus pull quotes in your template. Want a free run on one of your clients this week? You send the link, I deliver the finished draft within 24h of them finishing.
Week-one goal: Product live with working Stripe checkout by day 5; 60-name prospect sheet built by day 6; 20 DMs sent by day 7; at least 2 replies agreeing to a free run.
Section 5 of 8

30-day execution roadmap

MVP deadline: day 5First sale target: day 1130 days planned
Week 1 · Build & validate · 7 days
Day 1Skeleton

Repo, schema, auth

  • Create Next.js app on Vercel and Supabase project
  • Define writers, interviews, turns, drafts and credits tables
  • Add magic-link auth and dashboard shell
  • Deploy and confirm the production URL

Deliverable: Deployed app with auth, empty dashboard and migrated schema.

Day 2Interview engine

12-question AI interviewer

  • Encode the fixed case-study script as JSON
  • Build the OpenAI follow-up loop that demands one concrete detail per answer
  • Stream answers to the public interview route and persist each turn
  • Add optional voice note via Whisper

Deliverable: A stranger completes a 12-question interview end-to-end with a stored transcript.

Day 3Draft generation

The sellable artifact

  • One structured JSON call returning draft, 5 quotes, 3 stats, headline and LinkedIn post
  • Store each quote with transcript character offsets
  • Render the draft in one selectable template
  • Test against two seeded transcripts

Deliverable: One complete draft package generated from a real transcript.

Day 4Editor and export

Verification and branding

  • Inline quote highlighting plus verify-quote toggle
  • Flag sentences with no transcript span
  • Export to Markdown, PDF and .docx
  • Subdomain branding page with logo and colours

Deliverable: Closed loop from branded link to exported, branded file.

Day 5Billing and launch gateMVP shipped

Take money

  • Stripe Checkout live mode with Solo and Studio prices
  • Webhook updates subscription status and decrements credits
  • Resend emails: invite, completion notice, draft ready
  • Rehearse the full paid flow once with a founder card

Deliverable: Live paid product; MVP deadline met.

Day 6Prospect list

60 named buyers

  • Build 60-row sheet with name, role, agency size, published case study, LinkedIn URL
  • Draft three DM variants: free run, agency white-label, backlog reply
  • Warm the account with genuine comments on 15 target posts

Deliverable: 60-name target sheet plus three tested message variants.

Day 7First outbound

20 DMs and one public thread

  • Send 20 DMs with the free-run variant
  • Publish a build-in-public X thread describing the loop
  • Log every send and reply in the sheet

Deliverable: 20 DMs sent, one X thread live, reply log started.

Week 2 · Launch & first sale · 7 days
Day 8Outbound volume

Second batch

  • Send 20 more DMs
  • Run day-3 follow-ups on all unanswered threads
  • Reply to every response within two hours

Deliverable: 40 DMs sent with follow-ups logged.

Day 9Outbound completion

Finish the list

  • Send the final 20 DMs
  • Post the same offer in a Superpath channel
  • Book every positive reply into a concrete free run

Deliverable: 60 DMs sent and a Superpath post published.

Day 10Free runs

Deliver proof

  • Send interview links for the first accepted free runs
  • Deliver the finished draft within 24 hours of completion
  • Send the edit-diff screenshot with the Stripe link

Deliverable: First completed client case study delivered.

Day 11First saleFirst sale target

Convert the warmest

  • Follow up every warm reply with the finished sample
  • Offer a 20-minute walkthrough of the loop
  • Resend the Stripe link to the three warmest threads

Deliverable: Target: first paid Solo subscription.

Day 12Quality pass

Prompt fixes from real interviews

  • Review the first five real transcripts for vague-answer leaks
  • Tighten the follow-up prompts and the template output
  • Redeploy and re-run the affected drafts

Deliverable: Prompt set v1.1 deployed with before/after comparison.

Day 13Agency motion

Studio tier

  • Build a 10-name list of small content-agency owners
  • Pitch the white-label angle with a branded subdomain sample
  • Set Studio at $199 in the conversation, not in a discount

Deliverable: 10 agency pitches sent.

Day 14Week 2 review

Kill the worst message

  • Tally reply and conversion rate per DM variant
  • Cut the weakest variant and rewrite the opener
  • Reconcile all payments in the ledger

Deliverable: Metrics sheet and the single winning message chosen.

Week 3 · Repeat & systematize · 7 days
Day 15Public demo

Organic top of funnel

  • Publish an X thread showing one interview to one draft
  • Record a 90-second Loom of the loop
  • Pin the offer with the Loom attached

Deliverable: Public demo thread plus Loom link.

Day 16Second cohort

30 new names

  • Build 30 new prospects who wrote case studies in the last quarter
  • Send variant B to all 30
  • Cross-post the Loom to Superpath

Deliverable: 30 new prospects messaged.

Day 17Free run delivery

Three more drafts

  • Deliver free runs 3 to 5 within the 24-hour SLA
  • Capture each writer's edit diff
  • Ask each one what they would have charged for the study

Deliverable: Three additional delivered drafts with edit diffs.

Day 18Conversion

Free run to paid

  • Ask every free-run recipient for the $79 decision today
  • Send a one-page hours-saved calculation
  • Request one intro to another writer from each yes

Deliverable: Conversion decisions logged and two intro asks sent.

Day 19Studio readiness

White-label polish

  • Verify custom subdomain, logo and colour flow end-to-end
  • Polish Google Doc export formatting
  • Document the Studio checklist for a walkthrough

Deliverable: Studio tier demonstrably white-label on a live client-facing link.

Day 20Pricing evidence

Hold the price

  • Ask five writers whether 3 or 6 drafts is the binding limit
  • Record objection reasons verbatim
  • Do not discount; adjust the free-run framing instead

Deliverable: Pricing evidence notes and a decision memo.

Day 21Week 3 review

Books and pipeline

  • Reconcile TrustMRR against Stripe payouts
  • Count live subscriptions and credits consumed
  • Rank the open pipeline by likelihood to pay

Deliverable: Reconciled revenue ledger and prioritised pipeline.

Week 4 · Scale what works · 7 days
Day 22Outbound batch

Backlog posters

  • Target 30 writers publicly posting about case-study backlogs
  • Send variant C referencing their post specifically
  • Follow up once within 48 hours

Deliverable: 30 new DMs and a follow-up queue.

Day 23Referrals

Warm intros

  • Ask every active user for one intro
  • Offer one extra draft for a named intro that signs up
  • Send the ask to all current subscribers

Deliverable: Referral asks sent to 100% of users.

Day 24Reliability and cost

Unit economics

  • Add retry and failure email on generation errors
  • Audit tokens per interview and per draft
  • Cap interviews per credit to protect margin

Deliverable: Cost-per-draft log confirming delivery under $2.

Day 25Delivery SLA

Zero overdue drafts

  • Clear every outstanding free-run draft
  • Confirm 24-hour SLA held on each one
  • Send a short completion note with each export

Deliverable: Zero overdue drafts.

Day 26Close-out

Non-repliers and declines

  • Send the final day-7 follow-up to every silent thread
  • Ask every decline one question about the blocker
  • Archive dead threads so the sheet stays actionable

Deliverable: All threads closed with a stated reason.

Day 27Organic asset

Template landing page

  • Publish a public case-study interview template page with the offer
  • Post it on X and in Superpath
  • Add the product link to every earlier post

Deliverable: Live landing page with a working checkout link.

Day 28Sales sprint

Every warm lead

  • Message every warm lead with a 48-hour framing
  • Offer a live 15-minute free run on their next study
  • Log each outcome before midnight

Deliverable: All warm leads contacted and dispositioned.

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

Month two

  • Check in with each subscriber on their first month
  • Confirm renewal expectation explicitly
  • Escalate any credits unused as a signal of weak fit

Deliverable: Renewal confirmations for every active subscriber.

Day 30Close

Books and retro

  • Reconcile TrustMRR, refunds and fees line by line
  • Export the final ledger and profit statement
  • Write a one-page retro with the next 30-day hypothesis

Deliverable: Final reconciled profit statement and retro document.

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/Base/Upside: blended ARPU is $79 Solo-only in the conservative case, $119 in base (six Solo, three Studio) and $121 upside (thirteen Solo, seven Studio).
  • All scenarios: processor is Stripe at roughly 4.5% effective (2.9% + $0.30), stated as an assumption not a contract.
  • All scenarios: paid marketing is $0 because the acquisition channel is organic LinkedIn DMs and community posts; there is no ad budget.
  • Conservative: December reply rates depress results; 3 subscriptions from 60 DMs and 5 completed free runs.
  • Base: 15% DM reply rate holds, free runs convert at roughly one in three, 9 subscriptions inside the window.
  • Upside: one agency takes Studio early and refers two peers; 20 subscriptions with one $79 refund and no chargebacks.
  • Operating cost covers Vercel, Supabase, Resend, domain and OpenAI/Whisper inference at roughly $1-2 per drafted case study.
  • All figures are 30-day projections for the contest window only, not run-rate or verified revenue.

Conservative

3 customers
$237
collected revenue · avg payment $79
Refunds
$0
Chargebacks
$0
Payment fees
$11
Paid marketing
$0
Hackathon profit
$226
Operating costs
$18
Business profit
$208

Base case

9 customers
$1,071
collected revenue · avg payment $119
Refunds
$0
Chargebacks
$0
Payment fees
$48
Paid marketing
$0
Hackathon profit
$1,023
Operating costs
$60
Business profit
$963

Optimistic

20 customers
$2,420
collected revenue · avg payment $121
Refunds
$79
Chargebacks
$0
Payment fees
$109
Paid marketing
$0
Hackathon profit
$2,232
Operating costs
$120
Business profit
$2,112

Model your own numbers

Runs in your browser · nothing is saved
Collected revenue
$711
Payment fees (3.9%)
$28
Paid marketing
$0
Hackathon profit
$683
Business profit
$623

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 built from the stated assumptions. They are not verified revenue, existing customers, traction or guarantees of any kind.

Section 7 of 8

Main risks and validation experiments

Risk 1The writer's end customer never completes the link-based interview, starving the pipeline of drafts.

Mitigation

Cap at 12 questions, allow save-and-resume, send one reminder at 24 hours, draft from partial transcripts.

Validation experiment

Send 10 real interview links during the first five free runs and measure completion rate.

Kill criteria

If completion is under 30% across 10 links, drop to a 5-question voice-note-only interview before scaling outreach.

Risk 2Drafts need heavy rewriting, so the writer does not believe the time saving.

Mitigation

Generate quote-first from transcript spans and lock the output to the writer's own sample structure.

Validation experiment

Measure word-level edits on the first five delivered drafts.

Kill criteria

If median edits exceed 40% of the draft, fix prompts before sending DM batch two.

Risk 3December holiday season suppresses reply rates and slows the 30-day window.

Mitigation

Ship the MVP on day 5, front-load outbound into days 7-10, and lean on agency buyers who work through December.

Validation experiment

A/B two DM openers across the first 40 sends and compare replies.

Kill criteria

If 60 DMs produce under 10% replies, switch the primary channel to Superpath and X DMs on day 12.

Risk 4Freelancers resist because the interview itself feels billable.

Mitigation

Pitch more case studies in the same hours, never replacement; show the hours-saved arithmetic on their own study.

Validation experiment

Ask five writers directly whether the interview is billed separately, and record the answers verbatim.

Kill criteria

If 8 of 10 free-run users decline at $79 citing price, move to a $29 per-draft credit pack.

Risk 5Inference cost per interview exceeds plan value when writers send links that never become drafts.

Mitigation

Cap interviews per credit at 3, cap turns at 20, use a cheaper model for follow-ups and the stronger model only for drafting.

Validation experiment

Log tokens and cost for the first 20 interviews and 10 drafts.

Kill criteria

If cost per drafted case study exceeds $5, restrict Studio drafts and renegotiate the plan limits.

Section 8 of 8

Hour-by-hour: the first 24 hours

Hours 1-2

  • Create or confirm Vercel, Supabase, OpenAI, Resend and Stripe accounts
  • Connect Stripe to TrustMRR and confirm a test charge succeeds
  • Register the domain and point it at Vercel

Output: All accounts live, payment account connected, domain resolving.

Hours 3-5

  • Scaffold the Next.js repo and push to production
  • Write and apply the Postgres schema for writers, interviews, turns, drafts and credits
  • Add magic-link auth and an empty dashboard

Output: Deployed skeleton with auth and migrated schema.

Hours 6-8

  • Encode the 12-question case-study script as JSON
  • Build the streaming interview route with turn persistence
  • Test one full interview locally and inspect the transcript

Output: A locally completed 12-question interview with a stored transcript.

Hours 9-10

  • Start the 60-name LinkedIn prospect sheet
  • Fill the first 15 rows with role, agency size and a published case study
  • Note likely hours per case study for each

Output: 15-row prospect sheet with fields defined.

Hours 11-12

  • Write the three DM variants and the X build-in-public thread
  • Stage the day-7 send list and the follow-up schedule
  • Confirm no outreach is sent before the product is live

Output: Message copy staged with a dated send and follow-up calendar.

Complete report

Full 30-day launch report

Read the full generated report

Proofwright - 30-Day Public Report (Ship or Die $100K Hackathon)

Concept: Proofwright v1. Operator: one solo founder. Contest window: Monday 16 November 2026, 13:00 UTC to Wednesday 16 December 2026, 13:00 UTC. Every number in this report is a projection built from its own stated assumptions. Nothing here describes verified customers, sales, revenue or traction.

JEV scores this concept 69.1/100 (band B). Speed to first revenue is the strongest sub-score and operational simplicity the weakest, so the plan front-loads a sellable artifact and deliberately removes every operational surface that is not the interview-to-draft loop.

1. Product opportunity and positioning

The buyer is one person under deadline: a solo B2B case-study writer, or the owner of a two-to-ten person content shop selling case studies at a fixed price. Their day is not blocked by writing. It is blocked by booking a customer interview through a client, then through that client's customer, sometimes twice. Every slipped booking is unbilled time and a slipping deadline, and the case study is the least billable artifact in the engagement.

Proofwright compresses that funnel into a link. The writer brands it, sends it, and the customer answers twelve questions in a browser at 11pm if that is when they have five minutes. The AI moderator refuses vague answers and re-asks until it captures a number, an alternative the customer tried, or the objection someone had to overcome - the exact material a case study cannot be written without. The output is not a chat log. It is a 700-word draft in the writer's template, five pull quotes, three stat callouts, a headline and a LinkedIn post.

The defensible mechanic is verification, not generation. Generic AI writers produce plausible prose that a professional must fact-check line by line, which eats the time saved. Here every pull quote carries a span back to the transcript and the editor flags any sentence with no source, so the writer does one edit pass instead of a truth audit. That is what turns an AI draft from a liability into something a professional will put their name on. White-label is the second lever: the client's customer never sees Proofwright, only the writer's logo on a subdomain, so the writer can resell the speed as their own process.

The riskiest assumption is not demand. It is whether the writer's end customer will finish a link-based interview at all.

2. Smallest sellable MVP

The sellable unit is one finished case-study package from one real interview, delivered inside 24 hours.

Present on day five: a branded interview link; a twelve-question text interview with follow-up logic and a twenty-turn cap; a structured draft generator; an editor with quote highlighting and a verify-quote toggle; Markdown, PDF and Google Doc export; Stripe Checkout with credit metering; a dashboard showing sent, in progress, completed and draft ready.

Deliberately absent: outbound phone calls, CRM integrations, team seats, template marketplace, analytics. Each adds days and none is why someone pays $79 this week. The only voice feature is an optional browser voice note transcribed server-side, because some customers will not type but will talk for forty seconds.

3. Technical implementation plan

Next.js on Vercel, Supabase for Postgres, auth and storage, OpenAI for follow-ups and drafting, Whisper for voice notes, Stripe Checkout with a webhook, Resend for the three transactional emails.

Four build days plus one billing day. Day one is schema and auth. Day two is the interview engine: the script lives as JSON, the model sees the last three turns plus the question intent, and cannot advance until the answer contains a concrete detail or the customer has been re-asked twice. Day three is generation: one structured JSON call returning draft, quotes, stats, headline and social copy, with quote character offsets stored against transcript positions so highlighting and verification are deterministic rather than a second model guess. Day four is the editor, toggle and exports plus subdomain branding. Day five is live Stripe, credits, gating and email.

Hardest technical risk is a runaway or truncated interview. Mitigation is a hard twenty-turn cap, a finish-early escape, and drafting from partial transcripts, because a 70% interview that yields a usable draft beats a perfect interview nobody completes.

4. Pricing and monetization

Two monthly plans. Solo at $79 covers 3 drafts. Studio at $199 covers 12 drafts, the white-label subdomain, Google Doc export, LinkedIn copy and priority generation. Credits rather than seats, because the unit of value is a finished draft.

The week-one offer is a free completed case study on a real client, with the finished artifact used as proof. Conversion is a screenshot of the edit diff plus arithmetic: three case studies at roughly four hours saved is most of a working week returned for $79.

Blended expected revenue per account is about $119, assuming roughly two Solo accounts to every Studio. Revenue grows by upgrading writers who hit the 3-draft ceiling in month one, not by adding tiers inside the window.

5. First-customer acquisition

One channel: LinkedIn DMs to people who publicly describe themselves as case-study writers or content-agency owners. Around it: Superpath Slack channels, an X build-in-public thread, and follow-ups at day 3 and day 7 with the finished sample attached.

The 60-name sheet is a day-six deliverable, holding role, agency size, a case study they published and a note on what their current process likely costs them. Three variants go out: the free-run offer, the agency white-label angle, and a specific reply to anyone posting about a backlog.

Week-one target: product live with working checkout by day five, 60 names listed, 20 DMs sent, two free-run acceptances. Never claim results - send the artifact instead.

6. 30-day execution roadmap

Week 1 (days 1-5) builds to a live paid product; day 6 is the prospect sheet; day 7 sends the first twenty DMs. Week 2 (days 8-14) is outbound volume plus the first free runs, with the first paid conversion targeted for day 11 and prompt fixes on day 12. Week 3 (days 15-21) adds the agency and Studio motion, a public demo thread, referral asks, and a revenue reconciliation. Week 4 (days 22-28) is a fourth outbound batch, a reliability and cost audit, referral asks, and a sales sprint across every warm lead. Days 29 and 30 are renewal confirmations, final reconciliation and a written retro.

Gate rule: no outbound beyond day 7 until the loop has produced at least one complete draft package from a real transcript. If the loop is not complete by day 9, ship text-only without voice notes and keep going. The day-by-day table accompanying this report is the operative schedule.

7. Revenue and profit scenarios

Three projections, all hypotheses. Conservative: 3 Solo subscriptions at $79, $237 gross, no refunds or chargebacks, about $11 in processing, $0 paid marketing and $18 of hosting and inference. Base: 9 subscriptions blended at $119, $1,071 gross, $48 in processing, $60 operating. Upside: 20 subscriptions blended at $121, $2,420 gross, one $79 refund, $109 in processing, $120 operating.

The contest ranks hackathon profit: gross revenue minus refunds, chargebacks, processing fees and paid marketing spend. Hosting, domains and inference are excluded from that ranking number but are still real money, so business profit subtracts them too. Paid marketing is $0 in every scenario because the channel is organic DMs.

These are projections from stated assumptions, not verified revenue or existing customers.

8. Main risks and validation experiments

Completion risk: the end customer may never finish. Test ten real links inside the first five free runs; under 30% completion means cutting to a five-question voice-note interview.

Rewrite risk: the draft may need heavy editing. Measure word-level edits on the first five drafts; median edits above 40% means fixing prompts before the second DM batch.

Seasonality risk: December suppresses replies. Ship by day five, front-load outbound into days 7-10, and prefer agency buyers who work through the holidays. Under 10% reply on 60 DMs means switching channel on day 12.

Positioning risk: writers may hear this as replacing billable interview work. Pitch more case studies in the same hours, and ask five writers directly whether they bill the interview separately.

Cost risk: links that never become drafts can burn inference. Cap interviews per credit at three, cap turns at twenty, and use the cheaper model for follow-ups; above $5 of delivery cost per drafted study, tighten Studio limits.

9. First 24-hour action plan

Hours 1-2: open or confirm Vercel, Supabase, OpenAI, Resend and Stripe accounts, connect Stripe to TrustMRR, confirm a test charge, point the domain at Vercel.

Hours 3-5: scaffold the repo, apply the schema, add magic-link auth and an empty dashboard, deploy.

Hours 6-8: encode the twelve-question script, build the streaming interview route with turn persistence, run one full interview and inspect the transcript.

Hours 9-10: start the prospect sheet and fill the first fifteen rows with role, agency size and a published case study.

Hours 11-12: write the three DM variants and the X thread, stage the day-7 send list, and confirm nothing goes out before the product is live.

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 →