Score B · 6.83/10Free to read$100K Hackathon · November 16, 2026 → December 16, 2026Nov 16, 2026

Figurelock Cost Volume

Turns a founder's effort plan and rates into an agency-formatted SBIR/STTR cost workbook plus a budget justification narrative where every dollar figure is locked to a computed spreadsheet cell.

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.

Promising6.83/10
MVP build (solo)
5 days
Launch price
$149
Sales to $1,000 revenue
7 customers
First paying customer
Direct LinkedIn and X outreach to founders posting publicly about an upcoming SBIR/STTR deadline or an indirect rate question.
Who pays

Founder or ops/finance lead at a 2-15 person US deeptech, biotech or dual-use company preparing a first or second SBIR/STTR Phase I submission, typically 5-21 days before a deadline, with a technical volume drafted and no cost volume.

Biggest execution risk

Monetization potential 4.9/10

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 5 days; the plan targets a first sale on day 12; 7 sales at $149 clears $1,000 of collected revenue; strongest signal is speed to first revenue at 8.2/10.

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

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

0/10
Entry price
$149
Solo build
5 days
30d base revenue
$3,430
30d hackathon profit
$3,009
Section 1 of 8

Product opportunity and positioning

Target customer

Founder or ops/finance lead at a 2-15 person US deeptech, biotech or dual-use company preparing a first or second SBIR/STTR Phase I submission, typically 5-21 days before a deadline, with a technical volume drafted and no cost volume.

Why now

SBIR/STTR solicitations run on recurring cycles, so at any moment a cohort of founders sits one to three weeks from a deadline with the technical volume done and the cost volume untouched — a bounded artifact with a pass/fail outcome and a price inside founder signature authority.

Problem

Non-financial founders must convert an effort plan into person-months and loaded rates, assert an indirect rate with a pool/base schedule, and keep the spreadsheet arithmetic in agreement with separately typed justification prose, all inside the agency's required section order and columns.

Solution

One intake form produces three linked artifacts in one pass: an .xlsx cost workbook in the selected agency's section order, a .docx budget justification where every dollar sentence carries a cell-reference token, and a consistency report PDF; a validator blocks export whenever a stated figure disagrees with its computed cell.

Differentiation · Generic Excel templates and consultant decks do not check themselves. Figurelock is the only one of the three outputs that refuses to export a drifted number, which is the exact failure reviewers flag.

1. Product opportunity and positioning

The SBIR/STTR cost volume is the artifact nobody wants to own. The technical founder has spent weeks on the technical volume and now has 5 to 21 days to produce a spreadsheet that converts an effort plan into person-months, loaded rates and an indirect pool/base schedule, in the agency's required section order and columns — plus a justification narrative whose numbers must agree with that spreadsheet. The two drift apart because they are authored in different programs. A reviewer who finds a $4,000 discrepancy between the workbook and the prose does not ask which one is right.

Figurelock Cost Volume sells the reconciliation, not the spreadsheet. The intake is one form: roles, hours, base rates, fringe, indirect pool and base, materials, travel, consultants, subcontracts, equipment, period of performance. The output is three artifacts generated from that single input set: an agency-formatted .xlsx workbook with a computed rate buildup and totals row, a .docx justification narrative in which every dollar sentence embeds the cell reference it defends, and a consistency report PDF that lists each line item, its share of total cost, and the sentence defending it. The validator is the product: export is blocked when a narrative number differs from its computed cell, or when a line item above a configurable share of total cost has no justification.

The buyer is narrow on purpose. A founder or ops lead at a 2-15 person US deeptech, biotech or dual-use company, preparing a first or second Phase I submission, who already has a technical volume drafted and a deadline inside three weeks. They are not shopping for a platform; they are shopping for the one document blocking the submission. The secondary buyer is the independent SBIR consultant preparing several budgets per quarter, who needs the same artifact ten times with different client names.

This is not the discovery-and-drafting platform and it is not a technical-volume ghostwriter. It deliberately ignores opportunity matching and Volume 2, and it is not tied to a single agency topic. It monetises the one downstream artifact those tools leave behind.

The JEV score of 68.3 (band B) is honest about where this is weak: monetization potential scored 4.89. A one-time $149-$449 charge with no repeat path is the constraint. The plan attacks that directly with a consultant bundle that converts single purchases into repeat generations, and a re-export diff that gives buyers a reason to return for their next submission cycle. Speed to first revenue scored 8.19 and is the reason this wedge is worth 30 days: the buyer already knows they need this document this week, and the price clears without procurement.

Section 2 of 8

Smallest sellable MVP

An intake form that outputs one agency-formatted .xlsx cost workbook with a computed rate buildup, plus a consistency report PDF, sold one-time at $149 through Stripe.

Must-have features

  • Intake form covering roles, hours, base rates, fringe, indirect pool/base, materials, travel, consultants, subcontracts, equipment, period of performance
  • Rate buildup engine computing loaded hourly rates, fringe percentage and an indirect rate with explicit pool/base schedule
  • Agency formatter exporting .xlsx in a selectable SBIR/STTR section order with required columns and a totals row
  • Figure-locked .docx justification narrative with a source-cell token on every dollar sentence
  • Export validator blocking the .docx on a narrative-vs-cell mismatch or an unjustified line item above a share threshold
  • Consistency report PDF listing each line item, percentage of total, and its defending sentence
  • One-click re-export with a diff of changed cells and affected narrative sentences
  • Stripe Checkout with token-gated download links tied to a TrustMRR-connected account

Explicitly out of scope

  • Technical volume or any narrative outside the budget justification
  • Opportunity discovery, topic matching or SAM.gov registration help
  • Accounting or general-ledger integrations
  • Multi-user collaboration, comments or approval workflows
  • Phase II commercialization plans or post-award invoicing
  • Any human review, consulting call or done-for-you service

Demo flow (what the buyer sees)

  1. Buyer opens the intake form and enters roles, hours and base rates
  2. Buyer enters fringe and an indirect pool/base schedule
  3. Buyer adds materials, travel, consultants, subcontracts and equipment lines
  4. Buyer selects the agency section order and generates
  5. System computes loaded rates and totals, formats the .xlsx workbook
  6. System writes the justification narrative with cell-reference tokens and runs the validator
  7. Consistency report PDF is produced alongside the workbook
  8. Buyer downloads all three artifacts from token-gated links
Build plan

Technical implementation plan

Stack

Next.js on VercelSupabase Postgres for intake data and export recordsexceljs for the .xlsx cost workbookdocxtemplater for the figure-locked .docx narrativeAnthropic or OpenAI API for justification prose from structured line itemsStripe Checkout plus TrustMRR-connected payment accountServer-side validator comparing narrative numbers to computed cells

Architecture

A single Next.js app on Vercel renders the intake form and posts a structured budget object to a server route. A pure rate-buildup module computes loaded hourly rates, fringe and the indirect rate from the pool/base inputs, returning a flat array of line items each with a deterministic cell address. exceljs writes the workbook from that array using a per-agency template definition. An LLM call receives only the line-item array and produces justification sentences that must reference a cell token; a server-side validator re-parses every numeric token in the generated prose, compares it to the computed cell value, and refuses the export on mismatch. docxtemplater fills a .docx template. Supabase stores the intake object and each export record so re-export can diff changed cells. Stripe Checkout gates the download routes behind a signed token tied to a paid session.

Build steps

  1. Create Next.js app, Supabase schema for budgets and exports, and deployment to Vercel
  2. Implement the rate-buildup module with unit tests on fringe, loaded rate and pool/base arithmetic
  3. Implement exceljs workbook writer against a single agency section-order definition
  4. Add the second agency template definition and make the section order selectable
  5. Build the LLM justification generator that emits sentences with embedded cell tokens
  6. Implement the validator: parse numeric tokens, compare to computed cells, block on mismatch
  7. Generate the consistency report PDF from the line-item array and matched sentences
  8. Add docxtemplater narrative export and token-gated download routes
  9. Wire Stripe Checkout, webhook, and export record creation
  10. Add one-click re-export with a cell-diff and affected-sentence list

AI components

  • Justification narrative generation from structured line items, constrained to a fixed sentence pattern with cell tokens
  • A fallback deterministic sentence builder used whenever the LLM output fails validation twice
Hardest technical risk: The validator's number parsing: LLM prose will express the same figure as $12,480, 12.48k or 12,480 dollars, and a false mismatch blocks a legitimate export while a missed mismatch silently breaks the core promise. Mitigation is to constrain generation to a fixed sentence pattern with an explicit token and to normalise aggressively before comparison.
Section 3 of 8

Pricing and monetization

Workbook

Entry offer
$149 one-time

For Founder who only needs a reconciled spreadsheet and consistency report

  • Agency-formatted .xlsx cost workbook
  • Rate buildup with pool/base schedule
  • Consistency report PDF

Workbook + Justification

$249 one-time

For Founder submitting within 30 days who needs the narrative too

  • Everything in Workbook
  • Figure-locked .docx budget justification
  • Export validator with cell-level mismatch flags
  • Two re-exports after edits

Consultant Bundle

$449 one-time

For Independent SBIR consultant preparing several budgets per quarter

  • Ten workbook plus justification generations
  • Client name substitution on export
  • Saved rate templates per client

Model: One-time per-submission purchases in three tiers, with a consultant bundle that pre-buys ten generations.

First revenue offer: Send the founder their own effort plan converted into a formatted workbook plus consistency report for free, once, then offer the $249 Workbook + Justification for the narrative that matches it.

Expected ACV: $249

Upsell path: Every buyer receives one re-export credit at delivery; the second submission cycle is sold as a returning-customer $199 top-up, and consultants are moved to the $449 bundle after their second single purchase.

Target customers
32
Revenue model
$7,968
Costs
$140
Profit model
$7,828
Section 4 of 8

First-customer acquisition

Primary channel

Direct LinkedIn and X outreach to founders posting publicly about an upcoming SBIR/STTR deadline or an indirect rate question.

Where to find them

  • LinkedIn posts and comments containing SBIR, STTR, cost volume or indirect rate in the last 30 days
  • X search for the same terms filtered to founder bios
  • SBIR/STTR and dual-use founder Slack and Discord communities
  • Deeptech and biotech accelerators whose current cohort is between demo day and first Phase I
  • Independent SBIR consultants listed on public grant-consulting directories

Exact outreach motions

  • Send a two-line message referencing the specific deadline date in their post, offering to convert their effort plan into a formatted cost workbook once, free
  • Reply inside community threads where someone asks about indirect rates with the pool/base explanation and no link, then DM
  • Post one anonymised before/after per week showing a drifted budget and justification against the reconciled version
  • Day-3 follow-up referencing their deadline date, no pitch, one line
  • Day-7 follow-up with a one-screenshot consistency report from a comparable company shape, anonymised
  • Ask every delivered free workbook for a two-sentence reaction, and ask that reaction's author for the paid narrative

Message template

Saw your post about the [date] SBIR deadline. I built a tool that turns an effort plan into an agency-formatted cost workbook and a justification where every number is tied to its cell. Send me your roles/hours/rates and I will generate one for you free, once. If it reconciles, the narrative version is $249.
Week-one goal: 150 qualified names on the list, 70 messages sent, 8 replies with an effort plan, 3 free workbooks delivered.
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

Ship the skeleton

  • Scaffold Next.js app and deploy to Vercel
  • Create Supabase tables for budgets and exports
  • Freeze the intake field list in one spec file

Deliverable: Deployed skeleton with empty intake form

Day 2Engine

Rate buildup and workbook

  • Implement fringe, loaded rate and pool/base math
  • Unit test the arithmetic on three hand-checked budgets
  • Write the exceljs workbook for agency template A

Deliverable: Working .xlsx export from hardcoded input

Day 3Format and validation

Second template plus validator core

  • Add agency template B and section-order selector
  • Build the numeric token parser and cell comparison
  • Block export on mismatch with a visible error list

Deliverable: Validator that rejects a deliberately drifted sentence

Day 4Narrative

Justification and reports

  • Generate justification sentences with cell tokens
  • Fall back to deterministic builder after two failed validations
  • Render the consistency report PDF

Deliverable: Three linked artifacts from one input set

Day 5MonetizeMVP shipped

Checkout and launch

  • Wire Stripe Checkout and webhook to a TrustMRR-connected account
  • Gate download routes behind signed tokens
  • Publish landing page with the $149 and $249 tiers

Deliverable: Live product accepting payment

Day 6Hardening

Self-test at speed

  • Run three synthetic companies end to end
  • Fix every validator false positive found
  • Time a full generation run

Deliverable: Known-good build with timing baseline

Day 7Pipeline

Build the list

  • Collect 150 founders with deadlines in the next 30 days
  • Write the day-3 and day-7 follow-up copy
  • Prepare the anonymised before/after asset

Deliverable: 150-row outreach sheet with deadline dates

Week 2 · Launch & first sale · 7 days
Day 8Outreach

First 30 sends

  • Send 30 personalised messages
  • Post teardown 1 in two SBIR communities
  • Log every reply in the sheet

Deliverable: 30 sends, first replies logged

Day 9Outreach

Scale and deliver

  • Send 40 more messages
  • Generate free workbooks for every replier
  • Ship each workbook within 24 hours

Deliverable: Free workbooks delivered

Day 10Conversion

Free to paid

  • Day-3 follow-up on all unanswered threads
  • Offer $249 narrative to every free recipient
  • Fix any validator complaint raised in delivery

Deliverable: First paid checkout link sent

Day 11Outreach

Consultant segment

  • Build a 50-name consultant list
  • Send the $449 bundle offer
  • Second community teardown post

Deliverable: 50 consultant messages sent

Day 12RevenueFirst sale target

Close the first sale

  • Follow up on every open thread with a deadline under 10 days
  • Close at least one paid order
  • Confirm TrustMRR records the payment

Deliverable: First verified sale

Day 13Delivery

Earn the testimonial

  • Deliver the paid narrative with re-export credits
  • Ask for a two-sentence reaction
  • Log the exact objection pattern that almost blocked the sale

Deliverable: Delivered order plus written reaction

Day 14Product

Re-export diff

  • Ship one-click re-export with cell diff
  • Show affected narrative sentences per changed cell
  • Publish changelog note to buyers

Deliverable: Re-export diff live

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

Second wave

  • Send 40 messages to new names
  • Follow up on consultant thread
  • Post third teardown showing a reconciled budget

Deliverable: Pipeline refilled to 60 open threads

Day 16Accuracy

Template correctness

  • Audit template A and B section order against published instructions
  • Fix any column or totals-row deviation
  • Document assumptions on the landing page

Deliverable: Verified template definitions

Day 17Conversion

Second and third sales

  • Call every day-7 follow-up due today
  • Offer the $149 tier to non-responders who only need the workbook
  • Close with a deadline-anchored message

Deliverable: Cumulative paid orders logged

Day 18Content

Proof asset

  • Publish a before/after of a drifted budget
  • Add the intake to a single self-serve link in the post
  • Track click-to-intake rate

Deliverable: Self-serve funnel with measured click rate

Day 19Retention

Repeat path

  • Offer returning-customer $199 top-up to all buyers
  • Move any second-time buyer to the $449 bundle
  • Record which tier converts best

Deliverable: Repeat offer sent to full buyer list

Day 20Review

Midpoint audit

  • Compute verified profit so far from Stripe and TrustMRR
  • Rank channels by paid conversion, not replies
  • Kill the worst-performing message variant

Deliverable: Midpoint profit and channel ranking

Day 21Outreach

Biotech segment

  • Build 40-name biotech and life-sciences list
  • Send the same offer with a biotech-shaped example
  • Follow up on all open consultant threads

Deliverable: Second segment in pipeline

Week 4 · Scale what works · 7 days
Day 22Product

Remove the top blocker

  • Fix the single most common validator complaint
  • Reduce intake to the fields actually used
  • Ship the shorter form

Deliverable: Lower-friction intake live

Day 23Conversion

Deadline week push

  • Message every open thread whose deadline is under 14 days
  • Attach an anonymised consistency report
  • Close same day where possible

Deliverable: Deadline-week closes

Day 24Referral

Buyer-led leads

  • Ask every buyer for one founder with the same deadline
  • Send referral a free workbook offer
  • Log referral-sourced threads separately

Deliverable: Referral threads opened

Day 25Outreach

Fourth wave

  • Send 40 new messages
  • Post fourth community teardown
  • Re-run the best-converting message verbatim

Deliverable: Pipeline at 40 open threads

Day 26Product

Template depth

  • Add a third agency section-order definition
  • Test all three templates on one synthetic budget
  • Publish which agencies are supported

Deliverable: Three supported agency formats

Day 27Revenue

Bundle push

  • Offer the $449 bundle to every consultant contact
  • Offer it to any buyer with two or more purchases
  • Confirm delivery cost per generation stays near $2

Deliverable: Bundle offers out, unit cost confirmed

Day 28Finance

Reconcile

  • Reconcile Stripe, fees, refunds and TrustMRR
  • Issue any refund before the window closes
  • Record net hackathon profit

Deliverable: Reconciled profit statement

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

Double down

  • Rebuild the message that produced most paid orders
  • Send it to the remaining uncontacted names
  • Cut every channel that produced zero paid orders

Deliverable: Final outreach wave on the winning variant

Day 30Close

Final day

  • Close every open thread with a deadline under 21 days
  • Publish the 30-day result summary without inventing traction
  • Decide continuation on repeat-purchase evidence alone

Deliverable: Final numbers and go/no-go memo

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

  • Base: blended one-time ARPU lands near $245 as buyers mix the $149, $249 and $449 tiers.
  • Base: 14 paid buyers complete checkout inside the 30-day window.
  • Conservative: only 4 buyers convert from free-workbook offers.
  • Upside: 32 buyers, including roughly 4 consultant bundles at $449.
  • Payment fees assumed at 5% of collected revenue (Stripe 2.9% + $0.30 plus international and dispute buffer).
  • Ad spend is paid community sponsorship only in the upside case; all outreach is organic and costs $0.
  • Operating costs cover Vercel, Supabase, domain and LLM generation at roughly $1-2 per generation.
  • Conservative assumes zero refunds; base assumes one $249 refund; upside assumes two $249 refunds on missed deadlines.
  • Conservative and base assume zero chargebacks; upside assumes one $249 chargeback.
  • All figures are projections derived from these stated assumptions, not verified revenue or existing customers.

Conservative

4 customers
$848
collected revenue · avg payment $212
Refunds
$0
Chargebacks
$0
Payment fees
$42
Paid marketing
$0
Hackathon profit
$806
Operating costs
$48
Business profit
$758

Base case

14 customers
$3,430
collected revenue · avg payment $245
Refunds
$249
Chargebacks
$0
Payment fees
$172
Paid marketing
$0
Hackathon profit
$3,009
Operating costs
$90
Business profit
$2,919

Optimistic

32 customers
$8,480
collected revenue · avg payment $265
Refunds
$498
Chargebacks
$249
Payment fees
$424
Paid marketing
$150
Hackathon profit
$7,159
Operating costs
$260
Business profit
$6,899

Model your own numbers

Runs in your browser · nothing is saved
Collected revenue
$2,086
Payment fees (3.9%)
$81
Paid marketing
$0
Hackathon profit
$2,005
Business profit
$1,915

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 built from stated assumptions, not verified revenue, existing customers or real traction.

Section 7 of 8

Main risks and validation experiments

Risk 1Founders treat the cost volume as consultant work and refuse a self-serve tool at any price.

Mitigation

Lead with a delivered free workbook on their own numbers rather than a product description; the artifact is the pitch.

Validation experiment

Send 20 messages offering a free converted workbook; count how many send back an effort plan.

Kill criteria

Fewer than 2 of 20 sends an effort plan after two waves — the buyer wants a human, not a tool.

Risk 2Agency cost-volume section orders and required columns differ or change, so a wrong template destroys trust on the first sale.

Mitigation

Ship only two agency templates at launch, audit both against published instructions on day 16, and state supported agencies explicitly on the landing page.

Validation experiment

Have one practicing SBIR consultant review a generated workbook against their own template for a named agency.

Kill criteria

A reviewer finds a structural deviation that cannot be fixed within one day — narrow to a single agency before selling again.

Risk 3The figure-locked narrative reads mechanically, so buyers rewrite it and the re-export loop loses value.

Mitigation

Constrain generation to a fixed sentence pattern with slots for scope detail, and ship a deterministic fallback that reads plainly.

Validation experiment

Ask five free-workbook recipients to rate narrative readability 1-5 and to state whether they edited it.

Kill criteria

Average readability below 3 or every buyer rewrites — the narrative tier is not defensible at $249.

Risk 4Refunds after a missed deadline or a rejected submission erase a large share of collected revenue.

Mitigation

State clearly that the tool produces a reconciled cost volume, not an award, and refund only within the window per published terms.

Validation experiment

Track the refund reason on every request; if reasons cluster on output quality, fix before selling more.

Kill criteria

Refund rate above 15% of paid orders — the promise is outrunning the product.

Risk 5Single-founder outbound cannot reach 150 qualified conversations inside 30 days.

Mitigation

Fifty sends per wave, four waves, templated logging, and a hard rule to spend no more than 15 minutes per prospect.

Validation experiment

Measure reply rate after wave one; if under 8%, rewrite the first line before wave two.

Kill criteria

Under 3% reply rate across 100 sends — the channel is wrong, not the copy.

Risk 6Deadline-week support requests break the no-human-operations constraint.

Mitigation

Every delivery includes a self-serve re-export credit and a validator error list; no calls are offered at any tier.

Validation experiment

Count inbound support messages per delivery across the first ten orders.

Kill criteria

More than 2 support messages per delivery sustained — the product is not self-serve enough to scale solo.

Section 8 of 8

Hour-by-hour: the first 24 hours

Hours 1-2

  • Create the Next.js repository and deploy a blank app to Vercel
  • Create the Supabase project and two tables: budgets, exports

Output: Deployed skeleton URL and live database

Hours 3-5

  • Write the rate-buildup module: fringe, loaded hourly rate, indirect rate from pool and base
  • Write three hand-checked test budgets with expected totals

Output: Passing arithmetic tests with fixed expected values

Hours 6-8

  • Build the single-screen intake form against the field spec
  • Post the form payload to a server route that echoes the computed line items

Output: Working intake that returns computed line items

Hours 9-11

  • Wire exceljs to write a minimal .xlsx from the line-item array
  • Add a totals row and a rate-buildup sheet

Output: Downloadable workbook from real form input

Hours 12-14

  • Draft the 150-name outreach sheet structure with columns for name, company, deadline date, channel, status
  • Write the first outreach message and both follow-ups

Output: Outreach sheet template plus message copy ready to send

Complete report

Full 30-day launch report

Read the full generated report

Figurelock Cost Volume — 30-day launch report

Solo founder. Window 16 Nov 2026 13:00 UTC to 16 Dec 2026 13:00 UTC. One product, one channel, one price ladder.

1. Product opportunity and positioning

The SBIR/STTR cost volume is the artifact nobody wants to own. The technical founder has spent weeks on the technical volume and now has 5 to 21 days to turn an effort plan into person-months, loaded rates and an indirect pool/base schedule, in the agency's section order and columns, plus a justification narrative whose numbers must agree with that spreadsheet. The two drift apart because they are authored in different programs. A reviewer who finds a discrepancy does not ask which document is right.

Figurelock sells the reconciliation, not the spreadsheet. One intake produces three linked artifacts: an agency-formatted .xlsx workbook with a computed rate buildup and totals row, a .docx justification in which every dollar sentence embeds the cell reference it defends, and a consistency report PDF listing each line item, its share of total cost and the sentence defending it. Export is blocked when a narrative number disagrees with its computed cell.

The buyer is deliberately narrow: a founder or ops lead at a 2-15 person US deeptech, biotech or dual-use company, preparing a first or second Phase I, with the technical volume drafted and a deadline inside three weeks. Secondary: the independent SBIR consultant producing several budgets per quarter. This is not discovery, not matching, not technical-volume drafting, and not a single-agency topic tool. It monetises the one downstream artifact those categories leave behind.

The JEV score is 68.3, band B. Its own weakness note is monetization at 4.89: a one-time charge with no repeat path. This plan answers that with the $449 consultant bundle and a returning-customer top-up, while betting on the strongest sub-score, speed to first revenue at 8.19.

2. Smallest sellable MVP

The minimum sellable thing is one intake form that outputs a reconciled .xlsx workbook plus a consistency report PDF, sold once at $149. Everything else is provisional. The $149 tier exists because the workbook alone resolves the arithmetic drift, and a buyer with 9 days left will pay for that before they will evaluate a narrative generator.

The $249 tier adds the figure-locked .docx and the export validator. The $449 consultant bundle pre-buys ten generations. Build order follows the price ladder: workbook first, narrative second, bundle bookkeeping last.

Explicitly out of scope: technical volume, opportunity discovery, SAM.gov help, ledger imports, collaboration, post-award invoicing, and any human review or call.

3. Technical implementation plan

A single Next.js app on Vercel. Supabase Postgres holds the budget object and every export record so re-export can diff changed cells. The rate-buildup module is pure and unit-tested: it takes roles, hours, base rates, fringe, pool and base, and returns a flat line-item array where each item carries a deterministic cell address. exceljs writes the workbook from that array against a per-agency template definition; two agencies ship at launch, a third is added on day 26.

The LLM call receives only the line-item array and produces justification sentences constrained to a fixed pattern with an embedded cell token. The validator then re-parses every numeric token in the produced prose, normalises formats such as $12,480, 12.48k and 12,480 dollars, compares against the computed cell, and refuses export on mismatch. A deterministic sentence builder is the fallback after two failed validations, so a generation never dead-ends. docxtemplater fills the .docx. Stripe Checkout gates signed, tokenised download routes.

The hardest risk is the parser itself: a false mismatch blocks a legitimate export, a missed mismatch breaks the core promise. Constrain generation and normalise aggressively.

4. Pricing and monetization

Three one-time tiers: Workbook $149, Workbook + Justification $249, Consultant Bundle $449 for ten generations. One-time removes the subscription objection from a buyer who submits twice a year. The repeat path is the returning-customer $199 top-up offered at delivery and the bundle offered after any second single purchase. Expected ACV on a mixed order book is around $245.

5. First-customer acquisition

One channel: direct LinkedIn and X outreach to founders posting publicly about an SBIR/STTR deadline or an indirect rate question. Build a 150-name sheet with deadline dates in week one. The offer is not a demo; it is a free converted workbook on their own effort plan, delivered within 24 hours, followed by the $249 narrative ask. Three follow-ups: day 3, day 7, and one deadline-anchored nudge. Community teardowns, one per week, carry the self-serve link for readers who will not reply to a DM.

6. 30-day roadmap

Week 1 (days 1-7), build and arm. Days 1-5 ship the MVP: skeleton, rate engine, two agency templates, validator, narrative, checkout, landing page. Day 5 is the MVP deadline. Days 6-7 harden against three synthetic budgets and build the 150-name list.

Week 2 (days 8-14), first revenue. Waves of 30 and 40 messages on days 8-9, free workbooks for every replier, $249 offer on day 10, first sale target day 12, deliver plus testimonial ask on day 13, re-export diff shipped day 14.

Week 3 (days 15-21), volume and correction. Second outreach wave, consultant bundle push, template audit on day 16, midpoint profit audit on day 20 that ranks channels by paid conversion and kills the worst message. Biotech segment added day 21.

Week 4 (days 22-28), deadline-week pressure. Shorten the intake, push every thread with a deadline under 14 days, harvest referrals, add a third agency template, reconcile Stripe against TrustMRR before the window closes.

Days 29-30, close. Rebuild only the winning message, cut every channel with zero paid orders, publish real numbers, decide continuation on repeat-purchase evidence.

7. Revenue and profit scenarios

All three scenarios are projections from stated assumptions, not verified revenue.

Conservative: 4 customers at $212 blended, $848 gross. Base: 14 customers at $245, $3,430 gross, one refund. Upside: 32 customers at $265, $8,480 gross, including four consultant bundles, two refunds, one chargeback, and $150 of paid community sponsorship.

Payment fees are modelled at 5% of collected revenue. Operating costs stay between $48 and $260 and cover Vercel, Supabase, domain and LLM generation at roughly $1-2 per run; they never approach the sale price, so no order is sold below cost. Ranking uses hackathon profit, which excludes hosting and inference; business profit subtracts them.

8. Risks and validation

The dominant risk is that founders treat the cost volume as consultant work and will not touch a self-serve tool. Test: 20 offers of a free converted workbook, count effort plans returned. Below 2 of 20 after two waves, the buyer wants a human.

Second, template correctness. A wrong column destroys trust on the first sale. Test: one practicing consultant reviews a generated workbook against their own template. Any deviation unfixable in a day means narrowing to a single agency.

Third, narrative readability. If every buyer rewrites the prose, the $249 tier is indefensible. Test: five recipients rate readability 1-5 and report whether they edited. Below 3 average, drop the tier.

Fourth, refunds after missed deadlines, monitored against a 15% threshold. Fifth, outbound reach, monitored at 3% reply rate across 100 sends. Sixth, support load above two messages per delivery, which breaks the solo constraint.

9. First 24 hours

Hours 1-2: repository, Vercel deploy, Supabase tables. Hours 3-5: rate-buildup module with three hand-checked test budgets. Hours 6-8: intake form returning computed line items. Hours 9-11: exceljs workbook with totals row and rate-buildup sheet. Hours 12-14: outreach sheet structure and three message variants. Nothing is researched that is not tested against a real download by hour 11.

All financial scenarios are projections derived from stated assumptions. No customers, revenue, partnerships or traction are claimed.

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

Related ideas

All ideas →