Open Course Kit
Turns a syllabus into a license-verified open course pack plus a textbook cost-savings memo, so a professor can drop paid textbooks before the semester starts.
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.
Individual US college professors, adjuncts and instructional designers at community colleges and small private universities finalizing a spring-semester syllabus in late November or December under pressure to cut course material costs.
Operational simplicity 5.1/10
Why this could work in 30 days: one founder can ship the MVP in 5 days; the plan targets a first sale on day 12; 11 sales at $99 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
12 concrete actions. Tick them off as you go — nothing is saved or sent.
Product opportunity and positioning
Target customer
Individual US college professors, adjuncts and instructional designers at community colleges and small private universities finalizing a spring-semester syllabus in late November or December under pressure to cut course material costs.
Why now
Spring syllabi lock in during late November and December, affordability committees are asking for cost reductions now, and a single instructor can approve $99 without a purchase order.
Problem
Replacing a $180 textbook means a weekend of scattered OER searching, guessing at Creative Commons licenses, rebuilding the reading schedule by hand, and then justifying the swap to a department — so most instructors give up and re-assign the paid book.
Solution
The professor uploads a PDF or DOCX syllabus; the pipeline extracts every required reading, matches each to open candidates with source URL and license string, flags no-match weeks with a cheap used-book fallback, and emails a DOCX/PDF Course Pack plus a Textbook Cost Savings Memo after a $99 Stripe checkout.
1. Product opportunity and positioning
Open Course Kit converts one syllabus into two documents a professor already owes someone: a week-by-week Course Pack with license receipts, and a one-page Textbook Cost Savings Memo formatted for a department affordability committee. The buyer is a single US community-college or small-private-university instructor finalizing a spring syllabus in late November or December — usually an adjunct or instructional designer with no procurement process and a $99 discretionary budget.
The JEV profile is explicit about where this concept is strong and weak. Speed to first revenue scores 8.23 normalized and solo build feasibility 7.18; market demand scores 5.13 and operational simplicity 5.12. The plan therefore does not attempt to create a market. It converts a small, known, deadline-driven set of individuals, one email at a time, while keeping operations manual behind an automated pipeline.
Positioning against three alternatives. Manual OER hunting costs a weekend of scattered searches and copyright guessing. A generic AI chat returns unverifiable links with no license column, which cannot be published in a syllabus. Campus library OER services are reliable but appointment-based and slow during syllabus season. Open Course Kit's differentiator is the verification artifact itself: every matched reading carries a source URL and a named license, every unmatched week carries an explicit fallback, and the buyer receives an editable DOCX plus a PDF they can attach to an email.
The wedge stays narrow on purpose. No LMS or SIS integration, no student accounts, no rentals or buyback, no institutional procurement, no rights negotiation, no marketplace. Every one of those adds a second stakeholder and turns a one-person purchase into a committee decision the founder cannot close inside 30 days.
Commercially the shape is favourable for a hackathon ranked on verified profit: one-time $99 charges, no inference subsidy, sub-$3 delivery cost per pack, and a buyer who can pay the same day they receive the sample. The binding constraint is not demand creation — it is reaching enough individual instructors during the two-week syllabus window, which is why acquisition, not engineering, consumes days 7 through 30.
Smallest sellable MVP
One end-to-end pipeline: upload PDF/DOCX syllabus → parsed week-by-week reading list → matched open resources with license and source URL → DOCX + PDF Course Pack → one-page savings memo → emailed download link after a $99 Stripe payment.
Must-have features
- Syllabus upload (PDF/DOCX) with text extraction and LLM structured parse into weeks, titles, authors, editions, ISBNs
- OER matching via OpenStax and OER Commons search with an LLM ranking pass and explicit no-match flag
- License receipt table capturing source URL and license string (CC BY, CC BY-SA, CC BY-NC, public domain) plus a NonCommercial warning
- Course Pack renderer producing DOCX and PDF with cover page, per-week table, hyperlinks, license column and attribution lines
- One-page Textbook Cost Savings Memo PDF with prior retail price, new cost, per-section savings and methodology footnote
- Stripe Payment Link checkout with webhook-triggered generation job
- Tokenized job status page showing intake, matching, verification, generation
- Expiring download email via Resend
- Per-run token and API cost ledger
Explicitly out of scope
- LMS, SIS or bookstore integration
- Student accounts, rentals, fulfillment or buyback
- Institutional procurement, SSO or district contracts
- Publisher rights negotiation or a licensing clearinghouse
- Marketplace of user-uploaded resources
- Live chat or 24/7 support promise
- Department-wide or campus-wide rollout tooling in v1
Demo flow (what the buyer sees)
- Buyer clicks the Stripe Payment Link and pays $99
- Stripe webhook creates a job and emails a tokenized upload link
- Buyer uploads the syllabus PDF or DOCX
- Pipeline parses readings and runs repository searches plus LLM ranking
- License receipt builder records URL and license per accepted match and flags no-match weeks
- Course Pack renders to DOCX and PDF; savings memo renders to PDF
- Buyer gets an email with expiring download links and a status page URL
Technical implementation plan
Stack
Architecture
A Stripe Payment Link is the only public commerce surface. Its webhook creates a job row in Supabase and emails a tokenized upload URL. Upload writes the file to Supabase Storage and enqueues a processing run in a Vercel background route. That run extracts text, calls the LLM to produce a structured week-by-week reading list, queries repository APIs per reading, ranks candidates with a second LLM pass, and writes matches, licenses and URLs back to Postgres. A renderer assembles DOCX and PDF from the stored job data. Render results and expiring signed URLs are sent through Resend, and the tokenized status route reads job stage from Postgres. A cost ledger table records tokens and API calls per run so unit delivery cost stays visible.
Build steps
- Day 1: create Supabase project, tables (jobs, readings, matches, files, cost_ledger) and storage buckets
- Day 1: scaffold Next.js on Vercel with upload, status and download routes
- Day 2: implement PDF/DOCX text extraction and LLM structured parse with a strict JSON schema and retry
- Day 3: implement OpenStax and OER Commons search clients, per-reading candidate retrieval, LLM ranking and no-match flags
- Day 3: implement license capture and classification from source pages with a manual override column
- Day 4: implement DOCX template and PDF render for the Course Pack, including attribution lines and hyperlinks
- Day 4: implement the one-page savings memo PDF from parsed retail price and section size
- Day 5: wire Stripe Payment Link, webhook entitlement, job enqueue and Resend delivery with expiring links
- Day 5: run three real public syllabi end-to-end and fix the two most common parse failures
- Day 5: ship the cost ledger and set a per-run token ceiling
- Day 6-10: add match-quality fixes driven by replies only, one change per day
AI components
- LLM structured parse of syllabus text into weeks, titles, authors, editions and ISBNs
- LLM ranking and justification pass over repository search candidates
- LLM-drafted memo prose from computed price and section figures
- LLM classification assist for license strings, always stored with the raw source string for audit
Pricing and monetization
Course Pack
Entry offerFor A single professor replacing readings in one course
- Week-by-week Course Pack in DOCX and PDF
- License receipt table with source URLs
- One-page Textbook Cost Savings Memo
- One revision run after instructor edits
Department Pack
For A department chair or instructional designer standardizing five courses
- Five Course Packs with licenses and memos
- Cross-course duplicate-resource report
- Combined savings summary suitable for an affordability report
Semester Refresh
For Instructors who revise readings each term and want re-runs
- Unlimited re-runs for one course
- Link-rot checks on previously matched resources
- Updated license receipts on each refresh
Model: One-time per-course purchase with a vendor-free sample, plus a $349 five-course bundle and an optional $49/month refresh subscription.
First revenue offer: Free sample first: run a publicly posted syllabus through the pipeline and email the finished Course Pack and savings memo at no charge, with a single line — 'the full editable DOCX plus the committee-formatted memo is $99, same-day turnaround, reply GO and I'll send the payment link.'
Expected ACV: $99
Upsell path: Single-course buyers who mention a department, committee or multiple sections get the $349 Department Pack within seven days; instructors who say they revise readings every term move to the $49/month Semester Refresh.
First-customer acquisition
Direct personalized email to individual professors and instructional designers whose course pages and OER interest are publicly listed, with a free finished sample attached.
Where to find them
- Public community college course pages listing instructor name, course and email
- Public OER and open-textbook community member directories
- Instructional designer directories at small private universities
- Publicly posted spring-semester syllabi with contact addresses
- Teaching-focused community threads where course prep is being discussed
- Public department pages listing affordability or textbook-cost contacts
Exact outreach motions
- Send a free finished sample of the recipient's own publicly posted syllabus with a one-line $99 offer
- Follow up on day three naming the specific week where matching failed and a manual suggestion
- Follow up on day seven with a deadline hook referencing their syllabus lock date
- Post an anonymized before-and-after sample in OER and teaching communities pointing to a self-serve upload page
- Email chairs and instructional designers separately with the $349 five-course bundle framing
- Ask every buyer, with permission, for one sentence about the memo that can be quoted anonymously
Message template
Hi [Name] — I ran your [Course] syllabus through Open Course Kit. Here is the finished course pack: week-by-week open readings with the license and source link for each one, plus a one-page savings memo showing $[X] per student vs the current text. No charge, it's yours to use or ignore. If you want the editable DOCX and the committee-formatted memo, it's $99 and I turn it around same day. Reply GO and I'll send the link.
30-day execution roadmap
Week 1 · Build & validate · 7 days
Payment rails and skeleton
- Create Stripe account, connect it to TrustMRR, verify live mode
- Register domain, scaffold Next.js on Vercel with Supabase
- Create jobs, matches, files and cost_ledger tables plus storage buckets
- Deploy a placeholder tokenized status page
Deliverable: Live deployed skeleton with TrustMRR-connected Stripe in live mode
Syllabus parsing
- Implement PDF and DOCX text extraction
- Build LLM structured parse to a strict week/reading JSON schema
- Add retry and a visible parse-failure state
- Test on five public syllabi
Deliverable: Upload route that returns a structured reading list
OER match plus license capture
- Wire OpenStax and OER Commons search clients
- Add LLM ranking pass with confidence labels
- Capture source URL and license string per accepted match
- Emit explicit no-match flags with used-book fallback
Deliverable: Match table with license receipts for a full syllabus
DOCX and PDF output
- Build Course Pack DOCX template with license column and attribution lines
- Render matching PDF via pdf-lib
- Build the one-page savings memo PDF
- Run one full syllabus through rendering
Deliverable: Downloadable Course Pack DOCX/PDF plus memo PDF
Checkout, delivery, end-to-end
- Create the $99 Stripe Payment Link and webhook entitlement
- Trigger generation job from the webhook
- Send expiring download links via Resend
- Run three real syllabi end-to-end and fix the top two parse failures
Deliverable: MVP live: a stranger can pay $99 and receive both documents
Free sample engine
- Run 20 shortlisted public syllabi through the pipeline
- Record per-run cost in the ledger and enforce a token ceiling
- Fix the single most common no-match pattern
- Pick the five strongest samples for outreach
Deliverable: 20 finished free sample packs and a cost-per-pack figure
Prospect list
- Collect 150 professors and instructional designers from public course pages
- Record name, course, public email, syllabus deadline
- Segment by course level and by OER interest signals
- Verify every address is publicly listed
Deliverable: 150-row prospect sheet with matched free samples
Week 2 · Launch & first sale · 7 days
First 40 emails
- Send 40 personalized sample emails with the finished pack attached
- Log opens and replies in the sheet
- Reply to any question within two hours
- Post the offer in one teaching community thread
Deliverable: 40 sent emails and a live reply log
Volume and communities
- Send 60 more personalized emails
- Post the anonymized sample in two OER communities
- Publish the self-serve upload page with the $99 link
- Answers 20 remaining prospect rows
Deliverable: 100 total emails out and a public upload page
Match quality from replies
- Fix the top no-match course type reported in replies
- Add a manual override field for rejected matches
- Re-run two samples with the fix and resend
- Update the cost ledger view
Deliverable: Match-quality patch shipped and two improved samples resent
Day-three follow-ups
- Send day-three follow-ups naming a specific failed week
- Send 40 new first-touch emails
- Ask two engaged recipients what would block a purchase
- Confirm checkout works from a phone
Deliverable: 40 follow-ups sent, two objections documented
First paid sale
- Push the $99 link to every warm reply with a same-day turnaround promise
- Watch the webhook and delivery path on any live payment
- Personally review any generated pack before delivery
- Log the first sale in the ledger
Deliverable: First verified $99 payment inside the contest window
New segments
- Send 50 emails to instructional designers
- Post in two course-prep threads with the sample link
- Reply to all outstanding messages
- Trim the list of anyone who asked to stop
Deliverable: 150 total emails sent, zero unanswered replies
Week-two audit
- Compute send-to-reply and reply-to-purchase rates
- Rewrite the subject line with the best-performing variant
- Check refund and chargeback status
- Ship one memo formatting improvement
Deliverable: Conversion audit with a new subject line and memo v2
Week 3 · Repeat & systematize · 7 days
Department Pack push
- Email 30 chairs and instructional designers with the $349 five-course bundle
- Offer any single buyer with multiple sections the bundle
- Draft the combined savings summary template
- Track bundle interest separately
Deliverable: 30 bundle emails out and a savings-summary template
Sample page and volume
- Publish the anonymized before-and-after sample page
- Send 60 personalized emails
- Post the sample page in one affordability discussion
- Reply to all inbound within two hours
Deliverable: Public sample page live, 210 total emails sent
Deadline pressure
- Send day-seven follow-ups referencing syllabus lock dates
- Send 40 new first-touch emails
- Re-run any sample for a recipient who edited their syllabus
- Confirm delivery emails are not landing in spam
Deliverable: Follow-ups sent and deliverability verified
Refresh tier
- Create the $49/month Semester Refresh payment link
- Add a link-rot re-check run for an existing course
- Offer the refresh to every buyer by email
- Log subscription revenue separately in TrustMRR
Deliverable: Semester Refresh purchasable and offered to all buyers
Second sample batch
- Generate 20 new free samples from fresh syllabi
- Send 60 emails with the new samples
- Post the sample in one more teaching community
- Fix the top parsing failure from the new batch
Deliverable: 20 new samples out and a parsing fix shipped
Conversion audit two
- Rank all replies by stated intent and follow up by hand
- Identify the two objections that killed the most deals
- Remove any message angle with zero replies
- Check per-run cost against the $3 target
Deliverable: Objection list and a pruned message set
Rewrite from objections
- Rewrite the email around the top objection
- Send 60 emails with the new angle
- Add the savings memo figure to the subject line
- A/B the free-sample-first versus price-first framing
Deliverable: New message variant live with 60 sends
Week 4 · Scale what works · 7 days
Chair-focused page
- Build a one-page bundle explanation for chairs
- Email 30 more department contacts
- Offer a single free sample for one course to seed the bundle
- Send bundle reminder to any silent chair
Deliverable: Bundle page live and 60 chair contacts touched
Accuracy and no-match handling
- Manually review five generated packs for license errors
- Add a warning when a NonCommercial item appears
- Improve the used-book fallback wording for no-match weeks
- Re-run one affected pack and resend
Deliverable: License accuracy review complete with two guards added
Third touch
- Send a third touch to every non-responder with one specific improvement
- Send 40 new first-touch emails
- Ask two buyers what made them say yes
- Update the prospect sheet statuses
Deliverable: Third-touch wave sent and a documented reason-to-buy
Expansion from buyers
- Offer every single-course buyer the $349 bundle for their other sections
- Ask each buyer for a one-sentence quotable reaction with permission
- Offer Semester Refresh to any repeat requester
- Record all upsell outcomes
Deliverable: Expansion offers sent and quotable reactions collected
Deadline sprint
- Send 60 emails framed around the spring syllabus lock date
- Post in two final course-prep threads
- Reply to all warm leads with the $99 link
- Ship one no-match fix from the newest replies
Deliverable: Deadline-framed sprint sent and fixes shipped
Final volume push
- Send 80 emails to the remaining untouched prospects
- Re-message every warm but unpaid lead
- Offer a same-day turnaround guarantee on both documents
- Check all payment links still work
Deliverable: Prospect list fully worked with zero cold rows
Refunds and delivery quality
- Process any refund request within 24 hours and log it
- Re-run every delivered pack once for link validity
- Fix the top two broken link patterns
- Update the cost ledger with final per-pack costs
Deliverable: Clean delivered-pack set and a refreshed cost ledger
Days 29-30 · Consolidate · 2 days
Last wave
- Send the last outreach wave to warm leads only
- Post the sample page one final time
- Confirm TrustMRR shows every payment received
- Ask remaining buyers for the quotable reaction
Deliverable: Final outreach wave sent and payment reconciliation complete
Tally and submit
- Export the full payment list and subtract refunds and fees
- Verify every order is a new-customer order inside the window
- Write the submission with projected not verified language
- Ship the last delivery and confirm the download links
Deliverable: Reconciled verified-profit figure and submission sent
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: all scenarios hold ARPU at $99 and exclude Department Pack and Semester Refresh volume as unpriced upside.
- Conservative: roughly 1-2% of personalized sample emails convert; Base: roughly 4-6%; Upside: roughly 10-12% including community posts and referrals.
- Conservative: no paid marketing. Base: no paid marketing. Upside: $150 of sponsored posts in teaching newsletters.
- Payment processing assumed at Stripe standard 2.9% + $0.30 per transaction across all scenarios.
- Refunds modeled at zero in Conservative and Base because a free finished sample precedes payment; Upside carries two refunds at $99.
- Chargebacks modeled at zero in all scenarios; no delivered pack has yet generated a dispute.
- Operating costs cover Vercel, Supabase, domain, Resend and LLM/API delivery cost at roughly $2-$4 per pack.
- Revenue is a projection, not verified traction; no customers exist at the start of the window.
Conservative
3 customers- Refunds
- $0
- Chargebacks
- $0
- Payment fees
- $10
- Paid marketing
- $0
- Hackathon profit
- $287
- Operating costs
- $25
- Business profit
- $262
Base case
10 customers- Refunds
- $0
- Chargebacks
- $0
- Payment fees
- $32
- Paid marketing
- $0
- Hackathon profit
- $958
- Operating costs
- $45
- Business profit
- $913
Optimistic
30 customers- Refunds
- $198
- Chargebacks
- $0
- Payment fees
- $95
- Paid marketing
- $150
- Hackathon profit
- $2,527
- Operating costs
- $110
- Business profit
- $2,417
Model your own numbers
Runs in your browser · nothing is saved- Collected revenue
- $990
- Payment fees (3.9%)
- $39
- Paid marketing
- $0
- Hackathon profit
- $951
- Business profit
- $906
Hypothetical arithmetic, not a forecast. The fee rate is a flat 3.9% assumption, refunds and chargebacks are left at zero here, and none of this is tracked data.
Scenario models from stated assumptions, not verified revenue. Projections only; no customers, sales or traction are claimed.
Main risks and validation experiments
Risk 1Poor OER matches on upper-level or niche courses produce packs an instructor cannot use, driving refunds.
Emit an explicit no-match flag with a used-book fallback instead of forcing a bad match; manually review any pack before delivery.
Run ten upper-level public syllabi through the pipeline and count the percentage of readings with a usable match.
If fewer than 50% of readings in ten consecutive upper-level syllabi match, restrict the offer to intro courses selling only.
Risk 2License classification errors create a copyright problem for the instructor.
Store the raw license string alongside the classification, add a NonCommercial warning, and require a human check on any ambiguous page.
Hand-audit 30 matched items against their source pages and record the error rate.
If the audit error rate exceeds 10%, pause paid delivery until every match is human-verified.
Risk 3Repository APIs rate-limit or block automated search, breaking the matching step.
Cache search results per reading, add backoff, and keep a manual search fallback for failed queries.
Run 50 searches in one hour against each repository and log failures and throttling.
If more than 20% of searches fail under normal volume, switch to a cached index built once per day.
Risk 4Free samples generate interest but no purchases — instructors take the artifact and stop.
Send the free sample with a visible but unwatermarked gap: the editable DOCX and committee-formatted memo sit behind the $99 link.
Track reply-to-purchase rate over the first 100 sample emails and compare against the base assumption.
If 100 sample emails produce zero purchases and fewer than five price inquiries, change the offer structure before spending more days on outreach.
Risk 5Seasonality collapses demand after the spring syllabus window closes.
Sell Semester Refresh for termly re-runs and address the fall-syllabus April window in the same positioning.
Ask every buyer whether they would re-run the pack next term and record a yes/no.
If no buyer says yes to a next-term re-run, treat this as a single-window seasonal product and price accordingly.
Risk 6Stripe or TrustMRR onboarding delay blocks collection inside the contest window.
Complete payment account connection and one live test charge on day 1, before any build work.
Send a $1 test payment to yourself through the live link on day 1 and confirm it appears in TrustMRR.
If payment rails are not live and verified by end of day 2, stop building and resolve the blocker first.
Hour-by-hour: the first 24 hours
Hours 1-2
- Create the Stripe account and connect it to TrustMRR
- Confirm live mode and confirm the contest window start is respected
- Register the domain
Output: TrustMRR-connected payment account verified in live mode
Hours 3-5
- Create the Supabase project and tables (jobs, readings, matches, files, cost_ledger)
- Scaffold Next.js on Vercel and deploy a placeholder status page
- Store API keys in Vercel environment variables
Output: Deployed skeleton with database tables and a live status URL
Hours 6-9
- Write the syllabus text extractor for PDF and DOCX
- Define the structured parse JSON schema (week, title, author, edition, ISBN)
- Test on two public syllabi
Output: Structured reading list produced from two real syllabi
Hours 10-12
- Build the OpenStax and OER Commons search client
- Capture source URL and license string for three matched readings
- Draft the 40-email outreach message and the sample-first offer copy
Output: Three verified license receipts and the outreach message ready to send
Full 30-day launch report
Read the full generated report
Open Course Kit — 30-Day Launch Report
Concept v1. JEV weighted score 6.35; 100-point score 63.5, band C. Window: Mon 16 Nov 2026 13:00 UTC to Wed 16 Dec 2026 13:00 UTC. One founder, no team, no funding.
1. Product opportunity and positioning
Open Course Kit converts one syllabus into two documents a professor already owes someone: a week-by-week Course Pack with license receipts, and a one-page Textbook Cost Savings Memo formatted for a department affordability committee. The buyer is a single US community-college or small-private-university instructor finalizing a spring syllabus in late November or December — usually an adjunct or instructional designer with no procurement process and a $99 discretionary budget.
The JEV profile is explicit about where this concept is strong and weak. Speed to first revenue scores 8.23 normalized and solo build feasibility 7.18; market demand scores 5.13 and operational simplicity 5.12. The plan therefore does not attempt to create a market. It converts a small, known, deadline-driven set of individuals, one email at a time, while keeping operations manual behind an automated pipeline.
Positioning against three alternatives. Manual OER hunting costs a weekend of scattered searches and copyright guessing. A generic AI chat returns unverifiable links with no license column, which cannot be published in a syllabus. Campus library OER services are reliable but appointment-based and slow during syllabus season. Open Course Kit's differentiator is the verification artifact itself: every matched reading carries a source URL and a named license, every unmatched week carries an explicit fallback, and the buyer receives an editable DOCX plus a PDF they can attach to an email.
The wedge stays narrow on purpose. No LMS or SIS integration, no student accounts, no rentals or buyback, no institutional procurement, no rights negotiation, no marketplace. Each of those adds a second stakeholder and turns a one-person purchase into a committee decision the founder cannot close inside 30 days.
The commercial shape suits a contest ranked on verified profit: one-time $99 charges, no inference subsidy, sub-$3 delivery cost per pack, and a buyer who can pay the day they receive the sample. The binding constraint is not demand creation — it is reaching enough individual instructors during the two-week syllabus window. Acquisition, not engineering, consumes days 7 through 30.
2. Smallest sellable MVP
The smallest sellable thing is one end-to-end run: upload a PDF or DOCX syllabus, parse it into a week-by-week reading list, match each reading to open resources with license and URL, render a DOCX and PDF Course Pack plus a one-page savings memo, and email the download after a $99 Stripe payment.
Must-haves: upload and text extraction; LLM structured parse into weeks, titles, authors, editions, ISBNs; OpenStax and OER Commons queries with an LLM ranking pass; a confidence label and an explicit no-match flag; license receipt capture (CC BY, CC BY-SA, CC BY-NC, public domain) with a NonCommercial warning; DOCX and PDF rendering with attribution lines and hyperlinks; the savings memo PDF; a Stripe Payment Link plus webhook-triggered job; a tokenized status page; Resend delivery with expiring links; a per-run cost ledger.
Out of scope for 30 days: student-facing anything, buyback, rentals, SSO, LMS/SIS integration, publisher negotiation, marketplace uploads, live chat. Every one of these adds a stakeholder or an integration without adding a buyer.
This is buildable in five days by one person. Days 1-5 are the build; day 6 onward is samples and selling.
3. Technical implementation plan
Next.js on Vercel handles upload, job status and delivery. Supabase Postgres and Storage hold jobs, parsed readings, matches and rendered files. The Stripe Payment Link is the only public commerce surface; its webhook creates a job and emails a tokenized upload URL through Resend. Upload writes the file to Storage and starts a background route that extracts text, runs the LLM structured parse, queries repository APIs per reading, ranks candidates with a second LLM pass, and writes matches, licenses and URLs back to Postgres. A renderer assembles DOCX via the docx library and PDF via pdf-lib. Signed download URLs expire in 72 hours. A cost_ledger table records tokens and API calls per run.
Build order matters more than tooling choices. Parse first, then match, then licenses, then render, then commerce. Parsing is the hardest part: irregular syllabus PDFs — scanned pages, multi-column layouts, readings embedded in tables — can produce an empty parse, which yields an unusable pack and a refund. Mitigate with an explicit parse-failure state that fails loudly rather than shipping a partial document, plus a per-run token ceiling.
Two guards are non-negotiable before the first paid order: never force a bad match (emit a no-match flag with a used-book fallback instead), and always store the raw license string next to the classification so any disputed item can be audited.
4. Pricing and monetization
Three tiers. Course Pack at $99 one-time for one professor and one course, including the DOCX and PDF pack, the license table, the memo, and one revision run. Department Pack at $349 one-time for five courses plus a cross-course duplicate report and a combined savings summary. Semester Refresh at $49/month for unlimited re-runs of one course with link-rot checks and refreshed license receipts.
$99 is under the threshold where a solo instructor needs a purchase order, and above the level where the work feels throwaway. The sample is free and finished — not a preview — because the paid conversion argument is 'you already have the artifact; the editable version and the committee memo are the product.'
Upsell runs two directions: a buyer who mentions multiple sections or a committee gets the $349 bundle within a week; an instructor who says readings change every term gets the $49 refresh. Neither upsell is modeled in the scenarios.
5. First-customer acquisition
One channel: personalized email with a finished free sample of the recipient's own publicly posted syllabus. Build a list of 150 to 400 professors and instructional designers from public course pages and public OER community directories, recording name, course and public email only. Send 40 emails on day 8, 60 more daily thereafter, and follow up on day three (naming the specific week where matching failed) and day seven (referencing their syllabus lock date).
In parallel, post the same anonymized before-and-after sample in teaching and OER communities where course-prep threads are active, pointing at a self-serve upload page. Collect every no-match and rejection as structured feedback and ship one match-quality fix per day.
There will be no press, no ads, no partners, and no institution-wide outreach. Every sale must be closable in a single email thread.
6. 30-day execution roadmap
Days 1-5 — build and commerce. Day 1: Stripe connected to TrustMRR in live mode, Supabase tables, Next.js skeleton deployed. Day 2: PDF/DOCX extraction and structured parse. Day 3: repository search, ranking, license capture, no-match flags. Day 4: DOCX and PDF Course Pack plus the savings memo. Day 5: Payment Link, webhook, delivery email, three real syllabi end-to-end. Gate: a stranger can pay $99 and receive both documents.
Days 6-14 — first revenue. Day 6: 20 free samples and a cost-per-pack figure. Day 7: 150-row prospect sheet. Day 8: 40 personalized emails. Day 9: 60 more plus community posts and the public upload page. Day 10: match-quality patch from replies. Day 11: day-three follow-ups. Day 12: push the link to warm replies; target the first verified $99. Day 13: 50 emails to instructional designers. Day 14: conversion audit and a new subject line.
Days 15-23 — expand. Department Pack emails to 30 chairs; anonymized sample page published; day-seven deadline follow-ups; $49 refresh link created and offered to buyers; second sample batch of 20; objection-driven message rewrite; chair-focused bundle page; manual license accuracy audit with a NonCommercial warning added.
Days 24-30 — close. Third touch to all non-responders; upsell bundle and refresh to existing buyers; deadline-framed sprint of 60 then 80 emails; refunds processed within 24 hours; every delivered pack re-run once for link validity; payment reconciliation against TrustMRR; submission written with projected-not-verified language.
7. Revenue and profit scenarios
All three are projections derived from stated assumptions, not verified revenue. No customers, sales or traction are claimed at the start.
Conservative: 3 customers at $99 gives $297 gross, zero refunds, zero chargebacks, $10 processing fees, zero ad spend, and $25 operating cost. Base: 10 customers gives $990 gross, $32 processing, zero ad spend, $45 operating. Upside: 30 customers gives $2,970 gross, $198 refunds for two returned packs, $95 processing, $150 of sponsored posts in teaching newsletters, and $110 operating.
Processing assumes Stripe standard 2.9% plus $0.30 per transaction. Operating costs cover Vercel, Supabase, domain, Resend and LLM/API delivery at roughly $2-$4 per pack. Hackathon profit deducts refunds, chargebacks, processing and paid marketing but not hosting or delivery; business profit also deducts operating costs. Under the base case both figures stay positive, which is the point of a $99 price against sub-$3 delivery.
8. Risks and validation experiments
Six risks carry kill criteria. Match quality on upper-level courses: run ten upper-level syllabi and count usable matches; below 50% means selling to intro courses only. License accuracy: hand-audit 30 matches against source pages; above 10% error means every match gets human verification before delivery. Repository API throttling: 50 searches per hour against each source; above 20% failure means caching a daily index. Sample-to-purchase collapse: track the first 100 sample emails; zero purchases and under five price inquiries means changing the offer structure before sending more. Seasonality: ask every buyer about a next-term re-run; no yes answers means pricing it as a single-window product. Onboarding delay: verify a live payment into TrustMRR on day 1; not live by end of day 2 means stopping build work until it is.
9. First 24-hour action plan
Hours 1-2: create and connect the Stripe account to TrustMRR, confirm live mode, register the domain. Hours 3-5: Supabase project with jobs, readings, matches, files and cost_ledger tables; Next.js deployed with a placeholder status page; keys in Vercel. Hours 6-9: PDF/DOCX text extractor, structured parse schema, tested on two public syllabi. Hours 10-12: OpenStax and OER Commons search client, three verified license receipts, and the outreach message drafted and ready to send on day 8.
Disclaimer: every financial figure above is a projection from stated assumptions. No customers, revenue, partnerships or traction exist yet, and none should be read into this report.
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 →SCORM Sprint
Converts a PDF or slide deck into an LMS-ready SCORM 1.2 microcourse with auto-graded checks and completion tracking in under 10 minutes.
NestDrop Kit
NestDrop Kit turns a kids' boutique's existing product photos and inventory list into a complete 72-hour flash-sale drop pack: a hosted countdown sale page, email and SMS sequence, social captions, and a Scarcity Ladder schedule that staggers discounts per SKU.
Practice Ticket Studio
Generates a printable one-page weekly 'practice ticket' — measure ranges, tempo ladder, and a QR-linked student page — for private instrument teachers, so every student leaves the lesson knowing exactly what to practice.