Score C · 6.08/10Free to read$100K Hackathon · November 16, 2026 → December 16, 2026Nov 16, 2026

Block Ledger

Block Ledger turns a specialty-crop grower's QuickBooks or farm-accounting export plus a simple block list into a per-block profitability ranking and a next-season input budget, delivered as a PDF and Excel packet the same day.

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

Decide in 20 seconds

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

Speculative6.08/10
MVP build (solo)
5 days
Launch price
$349
Sales to $1,000 revenue
3 customers
First paying customer
Direct outreach to independent farm bookkeepers and small ag-focused CPA firms in California's Central Valley, offering a free packet on one redacted client export in exchange for a twenty-minute reaction call.
Who pays

Owner-operators and farm managers of 300-2,500 acre almond, pistachio or walnut operations in California's Central Valley, and the independent farm bookkeepers and small ag CPA firms who close their books during the November-December year-end window.

Biggest execution risk

Market demand 4.7/10

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

Why this could work in 30 days: one founder can ship the MVP in 5 days; the plan targets a first sale on day 12; 3 sales at $349 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

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

0/11
Entry price
$349
Solo build
5 days
30d base revenue
$6,282
30d hackathon profit
$6,094
Section 1 of 8

Product opportunity and positioning

Target customer

Owner-operators and farm managers of 300-2,500 acre almond, pistachio or walnut operations in California's Central Valley, and the independent farm bookkeepers and small ag CPA firms who close their books during the November-December year-end window.

Why now

Harvest accounting is being closed for year-end loan renewals and next season's input budget in exactly this November-December window, and the export the tool needs is already sitting on the grower's or bookkeeper's desktop.

Problem

The books show $180,000 spent on fertilizer, sprays, custom harvest and water, but not which blocks consumed it, because costs are coded by vendor and account, so growers allocate by acreage and keep overfunding their losing blocks.

Solution

The grower uploads an accounting export and a block list; an LLM proposes a cost driver per vendor and account; the grower accepts or overrides each mapping on one screen; the engine outputs a ranked per-block profit PDF, a below-break-even flag list and an Excel next-season input budget by block after Stripe payment.

Differentiation · Not a farm-management system or a peer benchmark: it is a same-day cost-allocation artifact with a reusable mapping profile, replacing a two-day bookkeeper spreadsheet at $100/hour.

Product opportunity and positioning

A Central Valley almond, pistachio or walnut operation closing its books in November knows one number: whole-farm profit. It does not know which blocks produced it. Custom harvest is invoiced per acre by a contractor, water arrives as district assessments, sprays are invoiced by the shop, and all of it lands in QuickBooks coded by vendor and account. Allocating that spend by acreage is the default, and it quietly subsidises the losing blocks with the winning ones, then overfunds them again next season.

Block Ledger sells one artifact that closes that gap: a per-block profitability packet built from an accounting export the grower already produces for tax and lender purposes. No new data entry. No farm-management migration. No season-long onboarding. The only friction is a fifteen-minute review where the grower confirms how each vendor and account should be spread across blocks — per acre, per tree, per hour, per water acre-foot, or a custom driver.

The product was scored 60.8/100 by JEV (band C, confidence 0.73). The criteria that decide a 30-day contest are the strong ones here: speed to first revenue at 8.03/10 and reachable customers at 6.99/10, because a bookkeeper can hand over a redacted export this week and the deliverable is a named, dated file. The weak criteria are market demand (4.68) and monetization (4.68), and this plan attacks both rather than ignoring them: the buyer is narrowed to the bookkeeper or small ag CPA firm inside the six-week year-end close, where the packet is a billable deliverable rather than a discretionary tool, and revenue is stacked with a five-packet advisor pack instead of depending on single $349 charges.

Positioning is deliberately narrow. This is not a farm OS, not a peer-benchmarking network, not a processor-PDF yield scorecard. Those require other farms' data, field boundaries, or a season of adoption. Block Ledger needs one grower's own export and produces an internal ranking plus a forward-looking budget, so it is useful to the first customer on the first day with zero network effect. The honest alternative today is a bookkeeper's spreadsheet: two days of billable time, no saved mapping, and a number the grower cannot re-run next year without paying again.

Why the timing works: harvest accounting is finalised in November and December, year-end lender renewals are being prepared, and next season's input budget is being set. The export exists. The question is live. A packet delivered the same afternoon is worth $349; the same packet in March is a curiosity.

2. Smallest sellable MVP

Five build days, one path: upload accounting export, confirm the column mapping, confirm the allocation drivers, pay, receive the PDF and the Excel budget. It is sellable on day 5 because the output is an artifact a grower can hand to a lender or a partner, not a dashboard that requires belief. Everything else is cut: no live QuickBooks OAuth, no yield or grade tracking, no processor settlement parsing, no peer benchmarking, no GIS, no payroll, no mobile app, no multi-user roles.

The one feature that must not be cut is the explicit unallocated residual line. If the export is too coarse to split custom harvest credibly, the packet says so and shows the money it could not place. A packet that hides an allocation gap is worse than no packet.

3. Technical implementation plan

Next.js on Vercel, Supabase for Postgres, auth and file storage, SheetJS for CSV/XLSX parsing, an LLM API to propose a driver and a plain-language reason per vendor and account, React-PDF for the packet, ExcelJS for the budget workbook, Stripe Checkout for payment, Resend for delivery email.

Flow: file lands in Supabase storage, SheetJS normalises rows, the LLM returns a driver proposal per distinct vendor/account pair, the grower edits it on one screen, the accepted mapping is persisted per farm and reused on the next upload. The allocation engine computes revenue, allocated cost, net and net-per-acre per block. The packet renders before the paywall so the grower sees the ranked table and the flagged blocks; the full PDF and Excel download unlock on the Stripe webhook. Every paid packet is emailed with signed links.

The hardest technical risk is the mapper proposing a confident wrong driver on a messy export. Mitigation: show the reason next to every proposal, require explicit accept or override, print the unallocated residual, and keep a hand-written fallback driver list the grower can pick from when the LLM output is unusable.

4. Pricing and monetization

One price: $349 per packet, one-time, paid by the grower or by the bookkeeper who rebills it. The advisor pack is five packets for $1,395, which is the lever against the weak monetization score: one engaged bookkeeper closing dozens of grower books becomes several packets in one relationship instead of one.

Stripe at 2.9% + $0.30 is roughly 3.3% on a $349 charge. Delivery cost per packet is a few dollars of LLM and rendering. Two packets cover the entire month's tooling; everything after that is margin. There is no subscription, so the 30-day number depends on new packets, not expansion — which is why the outreach motion runs every single day, not just at launch.

5. First-customer acquisition

One channel: direct outreach to independent farm bookkeepers and small ag-focused CPA firms in Fresno, Tulare, Kern, Stanislaus and Merced counties, sourced from public firm directories and CPA society listings. The offer is a free packet for one redacted client export, no signup, returned within the week, followed by a twenty-minute reaction call and a fix shipped inside 48 hours. In parallel, the same offer goes into California almond and pistachio grower groups.

6. 30-day roadmap

Days 1-5 build the whole product and go live with Stripe. Days 6-7 produce the outreach list and a redacted sample packet built from synthetic data, clearly labelled as synthetic. Days 8-14 run the first free packets, capture objections, ship the first fix, and push for the first paid packet around day 12; week two ends with a funnel sheet showing sends, replies, packets and payments. Days 15-21 double outreach volume and pitch the advisor pack to the two most engaged bookkeepers. Days 22-28 switch to batch days, prove mapping reuse, collect buyer quotes, ask every buyer for one referral, and send a last-call to warm leads before December 16. Days 29-30 reconcile Stripe against payment records, refund any dissatisfied buyer, and freeze the product.

7. Revenue and profit scenarios

All three scenarios are projections built from stated assumptions, not verified revenue; no customers exist at the start.

Conservative: 6 packets at $349 for $2,094 gross, about $63 in processing fees, no refunds, no chargebacks, no ad spend, roughly $30 of hosting, domain and LLM cost. Base: 18 packets for $6,282 gross, about $188 in fees, roughly $75 of operating cost. Upside: 40 packets for $13,960 gross, one refund, about $417 in fees, roughly $95 of operating cost.

Hackathon profit deducts refunds, chargebacks, processing fees and paid marketing but not hosting, domains, tooling or LLM delivery cost. Business profit deducts operating cost as well. In every scenario paid marketing is zero, because the channel is organic outreach and free group posts.

8. Main risks and validation experiments

Risk one: the export is too coarse to split custom harvest or water credibly. Test it in week one by having three bookkeepers send a redacted export and measuring the share of cost dollars that map to a driver; if most of the money lands in the residual, the wedge is thinner than assumed.

Risk two: the LLM proposes a visibly wrong driver and destroys trust in the whole packet. Test by running ten messy real-world export shapes through the mapper and counting how many proposals a bookkeeper would accept unchanged. Below seven in ten, the review screen needs restructuring before more outreach.

Risk three: the bookkeeper absorbs the work instead of the grower paying $349. Test with the second and third bookkeeper calls by asking directly who pays, then pushing the rebill framing as the default.

Risk four: year-end timing slips and buyers defer past December 16. Test by asking every free-run user for a purchase date on the reaction call; if most defer, the packet needs a cost line that improves the current lender conversation rather than next season's budget.

9. First 24 hours

Hour 1-2: create the repo, deploy a Next.js skeleton to Vercel, provision Supabase with auth and a storage bucket. Hour 3-5: build login, the upload form and the farm record. Hour 6-8: wire SheetJS parsing and the manual column-mapping fallback so a messy export renders as a normalised table. Hour 9-11: draft the LLM mapping prompt against two synthetic exports and inspect the proposed drivers by hand. Hour 12-14: build the one-screen driver review and persist the mapping profile. Output by hour 14: a deployed app that takes a login, a file and a block list, and shows proposed drivers a human can edit.

Section 2 of 8

Smallest sellable MVP

An upload-and-pay workflow that takes one accounting export and one block list and returns a per-block profit PDF plus an Excel next-season budget, generated before the paywall and released after a $349 Stripe payment.

Must-have features

  • CSV/XLSX upload with column auto-detection and a manual column-mapping fallback
  • Block list intake form plus bulk CSV import of block name, acres, planted year and variety
  • LLM-assisted allocation mapper proposing one cost driver per vendor/account with a plain-language reason
  • One-screen review where the grower accepts or overrides every proposed driver
  • Mapping profile persisted per farm and reused on later uploads
  • Allocation engine computing revenue, allocated cost, net and net-per-acre per block
  • PDF packet with ranked block table, below-break-even flag list, unallocated residual line and methodology footnote
  • Excel export with a per-block next-season input budget tab
  • Stripe Checkout releasing the finished files only after payment, with the packet rendered before the paywall
  • Resend delivery email carrying signed download links

Explicitly out of scope

  • No live QuickBooks or Xero OAuth sync in the first 30 days; file upload only
  • No yield, grade or processor settlement parsing
  • No peer benchmarking against other farms or county averages
  • No field boundary maps, satellite imagery or GIS layers
  • No payroll, invoicing, inventory or farm-management features
  • No multi-user roles, approvals or audit trail
  • No mobile app
  • No 24/7 support promise, phone line or chat widget

Demo flow (what the buyer sees)

  1. Upload a QuickBooks or bookkeeper CSV/XLSX export
  2. Confirm auto-detected columns or map them manually
  3. Paste or import the block list
  4. Review the proposed cost drivers per vendor and override where wrong
  5. Generate the packet and see the ranked block table and flagged blocks on screen
  6. Pay $349 through Stripe Checkout
  7. Receive the PDF packet and Excel budget by email with signed links
Build plan

Technical implementation plan

Stack

Next.js on VercelSupabase Postgres, auth and file storageSheetJS for CSV/XLSX parsingLLM API (Claude or GPT class) for column and driver proposals@react-pdf/renderer for the packet PDFExcelJS for the budget workbookStripe Checkout plus webhookResend for delivery email

Architecture

A single Next.js app on Vercel talks to Supabase for auth, Postgres and file storage. An uploaded export is parsed by SheetJS into normalised transaction rows; distinct vendor/account pairs are sent to the LLM, which returns a proposed cost driver and a short reason for each. The grower reviews and edits those proposals on one screen, and the accepted set is saved as a reusable mapping profile for that farm. A server-side allocation engine applies the drivers across blocks, computes revenue, allocated cost, net and net-per-acre, and records any unallocatable spend as an explicit residual. The packet renders as PDF and Excel before the paywall so a summary is visible; Stripe Checkout takes payment and the webhook unlocks full-file download and triggers a Resend email with signed links.

Build steps

  1. Create the Next.js repo, deploy to Vercel, provision Supabase with auth and a private storage bucket
  2. Build login, farm record creation and the upload form
  3. Parse uploads with SheetJS, add column auto-detection plus a manual mapping fallback
  4. Build the block list form and bulk CSV import with validation on acres and planted year
  5. Write the LLM mapping prompt and a JSON schema for driver, reason and confidence
  6. Build the one-screen driver review and persist the accepted mapping profile in Postgres
  7. Implement the allocation engine including per-acre, per-tree, per-hour, per-acre-foot and custom drivers, plus the unallocated residual
  8. Render the PDF packet with the ranked table, flag list, residual line and methodology footnote
  9. Render the Excel workbook with the per-block next-season input budget tab
  10. Integrate Stripe Checkout and the webhook that unlocks downloads
  11. Add Resend delivery email with signed, expiring links
  12. Run three synthetic exports end to end, then flip to live payments

AI components

  • Column detection assistance for exports whose headers are cryptic or duplicated
  • Per-vendor and per-account allocation driver proposal with a plain-language reason and confidence
  • Natural-language explanation text inside the packet's methodology footnote
Hardest technical risk: The LLM confidently proposing a wrong allocation driver on a messy real export, which invalidates every number downstream; mitigated by showing the reason, forcing explicit accept or override, printing the unallocated residual, and offering a fixed driver list as a fallback.
Section 3 of 8

Pricing and monetization

Block Profitability Packet

Entry offer
$349 one-time

For Owner-operator or farm manager of a single specialty-crop operation running one season's analysis

  • Per-block profitability ranking PDF
  • Below-break-even flag list with the cost lines driving the loss
  • Next-season input budget per block in Excel
  • Saved allocation mapping profile for reuse on the next upload

Advisor Pack (5 packets)

$1,395 one-time

For Independent farm bookkeeper or small ag CPA firm closing multiple grower books in the year-end window, rebilling each packet to the grower

  • Five Block Profitability Packets
  • One shared advisor login
  • Priority 24-hour turnaround on each client file
  • Saved mapping profile per client farm

Model: One-time $349 per Block Profitability Packet, with a five-packet advisor pack at $1,395 for bookkeepers and CPA firms who rebill it to grower clients.

First revenue offer: I built a tool that takes a grower's QuickBooks export and a block list and returns per-block profit plus next season's input budget by block, as PDF and Excel. Send me one client's export with the names redacted and I'll return the packet free this week, no signup. If it is useful, the paid version is $349 and your firm can rebill it.

Expected ACV: $349

Upsell path: The advisor pack converts one engaged bookkeeper from a single $349 packet into $1,395, and repeat runs on the same farm next season reuse the saved mapping, making the second purchase a low-friction re-order the bookkeeper can bill at a lower internal cost.

Target customers
40
Revenue model
$13,960
Costs
$510
Profit model
$13,450
Section 4 of 8

First-customer acquisition

Primary channel

Direct outreach to independent farm bookkeepers and small ag-focused CPA firms in California's Central Valley, offering a free packet on one redacted client export in exchange for a twenty-minute reaction call.

Where to find them

  • California Society of CPAs member listings filtered to Fresno, Tulare, Kern, Stanislaus and Merced counties
  • Public accounting-firm directories and Google Maps listings for bookkeeping firms in Visalia, Fresno, Bakersfield, Modesto and Merced
  • California almond and pistachio grower groups on Facebook
  • Central Valley farm-management association member lists and county farm bureau directories
  • LinkedIn profiles titled farm bookkeeper, ag accountant or controller in the Central Valley
  • Referrals from every free-run grower asked for one other grower's name

Exact outreach motions

  • Email batch 1 of 15 bookkeepers: free packet on one redacted client export, sample packet attached, no signup required
  • Post the same offer in three California almond and pistachio grower groups with a redacted sample attached
  • Send follow-up one at day 4 and follow-up two at day 9 with the sample packet, then stop
  • Run free packets and ask only what is wrong or missing, then ship the fix within 48 hours and send the corrected version back
  • Call five small ag CPA firms and pitch the packet as a year-end deliverable they can rebill to the grower
  • Email batch 2 of 20 new firms after the first fix lands, referencing the corrected packet

Message template

Subject: per-block profit from your grower's QuickBooks export

I built a small tool that takes a grower's QuickBooks (or bookkeeping) export plus a block list and returns a per-block profit ranking and next season's input budget by block, as a PDF and an Excel file. It runs the same afternoon. Send me one client's export with names redacted and I will return the packet free this week, no signup. If it holds up, it is $349 and your firm can rebill it as a year-end deliverable. Reply with a file and I will run it today.
Week-one goal: 50 bookkeeper and CPA firms on a tracked list, 15 emails sent, 3 grower-group posts live, 3 free packets delivered and 2 reaction calls booked.
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, auth, upload

  • Create Next.js repo and deploy to Vercel
  • Provision Supabase auth and private storage bucket
  • Build login, farm record and upload form

Deliverable: Deployed app accepts a login and a file upload.

Day 2Ingest

Parse export and block list

  • Parse CSV/XLSX with SheetJS into normalised rows
  • Add column auto-detection plus manual mapping screen
  • Build block list form and bulk CSV import with validation

Deliverable: Uploaded export renders as a table; block list saved.

Day 3Mapping

LLM driver proposals

  • Write the LLM prompt and JSON schema for driver, reason, confidence
  • Build the one-screen review with accept and override
  • Persist the accepted mapping profile per farm

Deliverable: Mapping profile saved and reloadable.

Day 4Output

Allocation engine and documents

  • Implement per-acre, per-tree, per-hour, per-acre-foot and custom drivers
  • Add the unallocated residual line
  • Render the PDF packet and the Excel budget tab

Deliverable: PDF packet and Excel budget generated from a test export.

Day 5LaunchMVP shipped

Paywall and go live

  • Wire Stripe Checkout and the unlocking webhook
  • Send delivery email with signed links via Resend
  • Run three synthetic exports end to end, then flip live

Deliverable: Block Ledger live and able to take a $349 payment.

Day 6Proof asset

List and sample packet

  • Build a 50-row list of bookkeepers and CPA firms with name, firm, county, contact
  • Generate a redacted sample packet from synthetic data and label it synthetic
  • Draft the outreach email and two follow-ups

Deliverable: 50-row outreach list and sample packet PDF.

Day 7First outreach

Send and log

  • Send 15 emails with the sample packet attached
  • Post the offer in three almond and pistachio grower groups
  • Open a reply tracker with send, reply, packet and paid columns

Deliverable: 15 sends, 3 group posts, tracker live.

Week 2 · Launch & first sale · 7 days
Day 8Free runs

Deliver and book calls

  • Run free packets for the first three respondents
  • Book two twenty-minute reaction calls
  • Note turnaround time per packet

Deliverable: 3 packets delivered, 2 calls booked.

Day 9Feedback

Objections

  • Run the two reaction calls and take verbatim notes
  • Rank the top five objections by severity
  • Pick the single fix with the largest trust impact

Deliverable: Ranked objection list and one chosen fix.

Day 10Fix one

Ship and redeploy

  • Implement the chosen fix
  • Re-render the affected packets and resend corrected versions
  • Log the change in a one-line changelog

Deliverable: v1.1 deployed and corrected packets resent.

Day 11Volume

Second batch

  • Send 20 emails to the next firms on the list
  • Send follow-up one to all day-7 non-repliers with the sample attached
  • Add 10 new firms from CPA society listings

Deliverable: 20 new sends, 15 follow-ups, list at 60.

Day 12First saleFirst sale target

Convert a warm user

  • Offer the paid packet to every free-run user with a same-day turnaround
  • Send the Stripe checkout link directly
  • Target one completed payment

Deliverable: First paying customer, target day 12.

Day 13CPA angle

Lender-adjacent firms

  • Call five small ag CPA firms
  • Pitch the packet as a rebillable year-end deliverable
  • Log who pays: firm or grower

Deliverable: 5 CPA calls, 2 substantive conversations logged.

Day 14Review

Funnel math

  • Compute sends, replies, packets and paid conversion
  • Rewrite the subject line if reply rate is under 10 percent
  • Confirm every paid packet was delivered

Deliverable: Funnel sheet with rates and one changed message.

Week 3 · Repeat & systematize · 7 days
Day 15Scale outreach

Widen the list

  • Send 25 emails to remaining firms
  • Add 20 new firms from county farm bureau and directory sources
  • Post the offer in three more grower groups

Deliverable: 45 sends and 3 new group posts.

Day 16Batch ops

Free-run throughput

  • Run four free packets in one block of the day
  • Write a one-page ops checklist for packet generation
  • Time each run to find the slowest step

Deliverable: 4 packets delivered and an ops checklist.

Day 17Advisor pack

Bookkeeper volume

  • Pitch the five-packet $1,395 advisor pack to the two most engaged bookkeepers
  • Offer priority 24-hour turnaround as the differentiator
  • Log the response and any price objection

Deliverable: At least one advisor pack proposal sent.

Day 18Fix two

Second-ranked objection

  • Implement the second-most-requested change
  • Verify the change against two previously delivered packets
  • Redeploy and note it in the changelog

Deliverable: v1.2 deployed.

Day 19Public proof

Redacted examples

  • Post a redacted packet excerpt in three grower groups and forums
  • Answer every comment the same day
  • Track which post drove a file upload

Deliverable: 3 posts live with replies answered.

Day 20Follow-up close

Final touches

  • Send follow-up two to batch-two non-repliers
  • Mark dead leads closed and stop contacting them
  • Summarise objections heard since day 9

Deliverable: Follow-up log closed with a response tally.

Day 21Review

Revenue decision

  • Count verified payments to date
  • If under eight sales, shift all effort to bookkeeper batch runs
  • Update the packet turnaround promise to match capacity

Deliverable: Decision memo on channel mix for the final nine days.

Week 4 · Scale what works · 7 days
Day 22Batch day

Bookkeeper client files

  • Run five client files for the most engaged bookkeeper
  • Deliver all five within 24 hours
  • Collect one written reaction per file

Deliverable: 5 packets delivered under the advisor pack.

Day 23Reuse proof

Second run on one farm

  • Run a second upload on an existing farm using the saved mapping
  • Time upload to finished packet
  • Record the reuse time in the ops checklist

Deliverable: Measured minutes from upload to packet on a repeat run.

Day 24Buyer language

Why they paid

  • Ask three buyers what the packet replaced
  • Record verbatim quotes about lender questions and spreadsheet hours
  • Rewrite one outreach line using their words

Deliverable: 3 verbatim quotes and one rewritten outreach line.

Day 25Referrals

Ask every buyer

  • Ask each buyer for one grower or bookkeeper name
  • Send intro emails the same day
  • Track referral source per payment

Deliverable: 5 referral asks sent with intros.

Day 26Lender test

Distribution spike

  • Contact four Central Valley ag lenders with a sample packet format
  • Ask whether they would request it from a borrower
  • Log replies and objections

Deliverable: 4 lender contacts and at least one reply.

Day 27Last call

Warm leads

  • Send a last-call email to every warm lead noting the December 16 deadline and year-end timing
  • Offer a same-day turnaround on any file received that day
  • Follow up by phone where a number exists

Deliverable: Last-call sent to roughly 20 warm leads.

Day 28Hardening

Delivery integrity

  • Batch-generate any pending packets
  • Match every payment to a delivered file
  • Fix any failed render or email bounce

Deliverable: Zero undelivered paid packets.

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

Verified numbers

  • Match payment records against Stripe charges
  • Refund any dissatisfied buyer and record it
  • Confirm zero chargebacks and zero paid ad spend

Deliverable: Reconciled revenue sheet for the contest window.

Day 30Wrap

Freeze and plan

  • Freeze the product and export the customer and mapping-profile list
  • Write a one-page next-quarter plan around advisor packs and repeat runs
  • List the top three weaknesses to fix before a next season

Deliverable: Final report, customer export and a one-page backlog.

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

  • All three scenarios are projections from stated assumptions, not verified revenue, and no customer exists at the start of the window.
  • Base assumption: 18 packets at the single $349 price; upside includes advisor packs measured as multiple packets.
  • Payment processing assumed at Stripe 2.9% plus $0.30 per charge, about 3.3 percent of gross on a $349 charge.
  • Paid marketing is zero in every scenario because the channel is organic email and free group posts; organic effort is not a cash cost.
  • Operating cost covers domain, hosting on free or near-free tiers, and LLM inference of roughly $2 to $2.50 per packet generated.
  • Conservative assumes 1 refund-free month and no chargebacks; upside assumes one refund on 40 packets as volume grows.
  • Delivery cost per packet stays under $3 because the packet is one PDF and one Excel workbook plus a single parsing and mapping pass.
  • No recurring revenue is assumed; each scenario counts new packets only.
  • Timing assumes harvest accounting is in the grower's books by the November-December window, which may fail for late-harvested walnut blocks.

Conservative

6 customers
$2,094
collected revenue · avg payment $349
Refunds
$0
Chargebacks
$0
Payment fees
$63
Paid marketing
$0
Hackathon profit
$2,031
Operating costs
$30
Business profit
$2,001

Base case

18 customers
$6,282
collected revenue · avg payment $349
Refunds
$0
Chargebacks
$0
Payment fees
$188
Paid marketing
$0
Hackathon profit
$6,094
Operating costs
$75
Business profit
$6,019

Optimistic

40 customers
$13,960
collected revenue · avg payment $349
Refunds
$349
Chargebacks
$0
Payment fees
$417
Paid marketing
$0
Hackathon profit
$13,194
Operating costs
$95
Business profit
$13,099

Model your own numbers

Runs in your browser · nothing is saved
Collected revenue
$6,282
Payment fees (3.9%)
$245
Paid marketing
$0
Hackathon profit
$6,037
Business profit
$5,962

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 figures are forward-looking scenario projections built from the stated assumptions. They are not verified revenue, existing customers or real traction.

Section 7 of 8

Main risks and validation experiments

Risk 1Accounting exports are too coarse to split custom harvest, water and sprays credibly, leaving most cost dollars unallocated.

Mitigation

Print an explicit unallocated residual line in every packet and offer a fixed driver list the grower can pick from when automated proposals fail.

Validation experiment

Have three bookkeepers send one redacted export each in week one and measure the share of cost dollars that map to a driver.

Kill criteria

If more than half of cost dollars land in the residual on three different real exports, the wedge is too thin and the concept needs a narrower cost category or a different buyer.

Risk 2The LLM proposes a visibly wrong allocation driver and the grower distrusts every number in the packet.

Mitigation

Show the reason next to each proposal, require explicit accept or override, and store the override so it is reused on later uploads.

Validation experiment

Run ten messy export shapes through the mapper and count how many proposals a bookkeeper would accept unchanged on a blind review.

Kill criteria

If fewer than seven in ten proposals are accepted unchanged after two prompt iterations, stop outreach and rebuild the review screen before selling again.

Risk 3Bookkeepers absorb the analysis as part of their existing fee, so the grower never pays the $349 separately.

Mitigation

Lead the pitch with the rebill framing and price the advisor pack so the firm makes margin by reselling each packet.

Validation experiment

Ask each of the first five bookkeepers directly who would pay and what they would rebill it at.

Kill criteria

If four of five bookkeepers say the grower would not pay separately and would not rebill, move to selling direct to owner-operators instead.

Risk 4Year-end timing slips and buyers defer past December 16.

Mitigation

Reframe one section of the packet around the current lender renewal conversation rather than next season's budget so the value is immediate.

Validation experiment

On every reaction call, ask for a purchase date and record whether the buyer says this month, next month or after year-end.

Kill criteria

If most free-run users name a date after December 16, the contest window cannot be met through this channel and the offer needs a same-week trigger.

Risk 5A disputed allocation forces rework that destroys margin at a one-time $349 price.

Mitigation

Cap rework to one driver change per packet and state that limit in the delivery email.

Validation experiment

Track rework requests per paid packet across the first ten sales.

Kill criteria

If more than one in three paid packets needs a second full rebuild, stop selling until the driver review absorbs the disagreement up front.

Section 8 of 8

Hour-by-hour: the first 24 hours

Hours 1-2

  • Create the Next.js repo and deploy the skeleton to Vercel
  • Provision Supabase with auth and a private storage bucket

Output: Deployed app URL and a live Supabase project.

Hours 3-5

  • Build login, farm record creation and the upload form
  • Store an uploaded file against the farm record

Output: A working authenticated upload path.

Hours 6-8

  • Wire SheetJS to parse an uploaded CSV/XLSX
  • Add the manual column-mapping fallback so cryptic headers still parse

Output: An uploaded export rendered as a normalised transaction table.

Hours 9-11

  • Draft the LLM mapping prompt against two synthetic exports
  • Hand-inspect the proposed drivers and reasons and tighten the prompt

Output: A prompt and JSON schema that returns driver, reason and confidence.

Hours 12-14

  • Build the one-screen driver review with accept and override
  • Persist the accepted mapping profile per farm

Output: An end-to-end skeleton: login, file, block list, editable drivers, saved profile.

Hours 15-16

  • Draft the outreach email and the first 15 bookkeeper targets from public directories

Output: A 15-row outreach list and a sendable email draft.

Complete report

Full 30-day launch report

Read the full generated report

Block Ledger — 30-Day Launch Report

1. Product opportunity and positioning

A Central Valley almond, pistachio or walnut operation closing its books in November knows one number: whole-farm profit. It does not know which blocks produced it. Custom harvest is invoiced per acre by a contractor, water arrives as district assessments, sprays are invoiced by the shop, and all of it lands in QuickBooks coded by vendor and account. Allocating that spend by acreage is the default, and it quietly subsidises the losing blocks with the winning ones, then overfunds them again next season.

Block Ledger sells one artifact that closes that gap: a per-block profitability packet built from an accounting export the grower already produces for tax and lender purposes. No new data entry, no farm-management migration, no season-long onboarding. The only friction is a fifteen-minute review where the grower confirms how each vendor and account spreads across blocks — per acre, per tree, per hour, per water acre-foot, or a custom driver.

JEV scored this concept 60.8/100 (band C, confidence 0.73). The criteria that decide a 30-day contest are the strong ones: speed to first revenue at 8.03/10 and reachable customers at 6.99/10, because a bookkeeper can hand over a redacted export this week and the deliverable is a named, dated file. The weak criteria are market demand (4.68) and monetization (4.68), and this plan attacks both instead of ignoring them. The buyer is narrowed to the bookkeeper or small ag CPA firm inside the six-week year-end close, where the packet is a billable deliverable rather than a discretionary tool. Revenue is stacked with a five-packet advisor pack instead of depending on single $349 charges.

Positioning stays narrow. This is not a farm OS, not a peer-benchmark network, not a processor-PDF yield scorecard. Those need other farms' data, field boundaries or a season of adoption. Block Ledger needs one grower's own export and returns an internal ranking plus a forward-looking budget, so it is useful to the first customer on day one with zero network effect. The honest alternative today is the bookkeeper's spreadsheet: two days of billable time, no saved mapping, and a number the grower cannot re-run next year without paying again.

Timing matters more than features. Harvest accounting is finalised in November and December, lender renewals are being prepared, and next season's input budget is being set. The export exists and the question is live. A packet delivered the same afternoon is worth $349; the same packet in March is a curiosity.

2. Smallest sellable MVP

Five build days, one path: upload export, confirm column mapping, confirm allocation drivers, pay, receive PDF and Excel. It is sellable on day 5 because the output is an artifact a grower can hand to a lender or a partner, not a dashboard that requires belief. No live accounting sync, no yield tracking, no processor parsing, no benchmarking, no GIS, no payroll, no mobile app, no multi-user roles.

The feature that cannot be cut is the explicit unallocated residual line. If an export is too coarse to split custom harvest credibly, the packet says so and shows the money it could not place. A packet that hides an allocation gap is worse than no packet.

3. Technical implementation plan

Next.js on Vercel, Supabase for Postgres, auth and file storage, SheetJS for parsing, an LLM API to propose a driver and a reason per vendor and account, React-PDF for the packet, ExcelJS for the budget workbook, Stripe Checkout for payment, Resend for delivery.

The file lands in storage, SheetJS normalises the rows, the LLM returns a driver proposal per distinct vendor and account pair, and the grower edits it on one screen. The accepted mapping is persisted per farm and reused on the next upload. The allocation engine computes revenue, allocated cost, net and net-per-acre per block. The packet renders before the paywall so the ranked table and flagged blocks are visible; the full PDF and Excel unlock on the Stripe webhook and are emailed with signed links.

The hardest technical risk is a confident wrong driver on a messy export. Mitigation: show the reason for every proposal, force explicit accept or override, print the residual, and keep a hand-written driver list as fallback.

4. Pricing and monetization

One price: $349 per packet, one-time, paid by the grower or by the bookkeeper who rebills it. The advisor pack is five packets for $1,395 and is the direct answer to the weak monetization score: one engaged bookkeeper becomes several packets inside one relationship instead of one.

Stripe at 2.9% plus $0.30 is roughly 3.3% on a $349 charge. Delivery is a few dollars of inference and rendering per packet. Two packets cover the month's entire tooling cost; everything after is margin. There is no subscription, so the 30-day number depends on new packets rather than expansion, which is why outreach runs daily rather than only at launch.

5. First-customer acquisition

One channel: direct outreach to independent farm bookkeepers and small ag-focused CPA firms in Fresno, Tulare, Kern, Stanislaus and Merced counties, sourced from public firm directories and CPA society listings. The offer is a free packet on one redacted client export, no signup, returned within the week, followed by a twenty-minute reaction call and a fix shipped within 48 hours. The same offer goes into California almond and pistachio grower groups in parallel.

6. 30-day roadmap

Days 1-5 build the whole product and go live with Stripe. Days 6-7 produce the outreach list and a redacted sample packet built from synthetic data, clearly labelled as synthetic. Days 8-14 run the first free packets, capture objections, ship the first fix, and push for a first paid packet around day 12; week two closes with a funnel sheet. Days 15-21 double outreach volume and pitch the advisor pack to the two most engaged bookkeepers. Days 22-28 run batch days, prove mapping reuse, collect buyer quotes, ask every buyer for one referral, and send a last-call before December 16. Days 29-30 reconcile Stripe against payment records, refund any dissatisfied buyer, and freeze the product.

7. Revenue and profit scenarios

All three scenarios are projections from stated assumptions, not verified revenue. No customer exists at the start of the window.

Conservative: 6 packets at $349 for $2,094 gross, about $63 in processing fees, no refunds, no chargebacks, no ad spend, and roughly $30 of hosting, domain and inference cost. Base: 18 packets for $6,282 gross, about $188 in fees and about $75 of operating cost. Upside: 40 packets for $13,960 gross with one refund, about $417 in fees and about $95 of operating cost.

Hackathon profit deducts refunds, chargebacks, processing fees and paid marketing, but not hosting, domains, tooling or inference delivery cost. Business profit deducts operating cost as well. Paid marketing is zero in every scenario because the channel is organic outreach and free group posts.

8. Main risks and validation experiments

Coarse exports. Test in week one by having three bookkeepers send a redacted export and measuring what share of cost dollars maps to a driver. If most of the money lands in the residual across three real files, the wedge is too thin.

Wrong drivers. Blind-review the mapper's proposals on ten messy export shapes and count how many a bookkeeper would accept unchanged. Under seven in ten after two prompt iterations, rebuild the review screen before selling more.

Who pays. Ask each of the first five bookkeepers directly whether the grower pays separately and what they would rebill. If four of five say no on both counts, switch to selling direct to owner-operators.

Timing slip. Ask every free-run user for a purchase date and record it. If most name a date after December 16, reframe the packet around the current lender renewal so the value is immediate.

9. First 24 hours

Hours 1-2: repo, Vercel deploy, Supabase auth and storage. Hours 3-5: login, farm record, upload path. Hours 6-8: SheetJS parsing plus manual column fallback. Hours 9-11: LLM mapping prompt tested against two synthetic exports. Hours 12-14: one-screen driver review with persisted profile. Hours 15-16: outreach email drafted and fifteen bookkeeper targets listed. By hour 14 the app takes a login, a file and a block list, and shows editable drivers.

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 →