Skill Sprint Cards
A web tool that turns a tutor's own word list and target skill into a branded 5-minute playable practice game plus a plain-English parent note, delivered as one link the tutor texts to the family after the lesson.
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.
Independent early-reading and early-math tutors and 1-5 person tutoring studios (US/UK/CA/AU) selling 1:1 sessions online or in person.
Market demand 3.9/10
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 9; 7 sales at $149 clears $1,000 of collected revenue; strongest signal is speed to first revenue at 8.6/10.
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
14 concrete actions. Tick them off as you go — nothing is saved or sent.
Product opportunity and positioning
Target customer
Independent early-reading and early-math tutors and 1-5 person tutoring studios (US/UK/CA/AU) selling 1:1 sessions online or in person.
Why now
The output is used in the same conversation where it is discovered, so a free first card and a $149 license can both happen in one day with no procurement, no app store and no integrations.
Problem
After every session the tutor owes the family a practice task, but building something a 5-7 year old will actually do takes 20-40 unpaid minutes, so they send photocopied worksheets that get ignored and parents never see what was taught.
Solution
The tutor picks a skill, pastes 8-12 of their own words and a name, and gets a three-round HTML5 game on a permanent unique URL plus a four-sentence parent note rendered as a shareable PNG card, both under the tutor's brand.
Skill Sprint Cards — Product Opportunity
The buyer and the trigger
The buyer is one identifiable professional: an independent early-reading or early-math tutor, or a 1-5 person tutoring studio, selling 1:1 sessions online or in person in the US, UK, Canada or Australia. They are not a school, not a district and not a parent buying a consumer app. They have a public booking page or a tutor Instagram account, which means they are directly reachable this week without a cold-start marketplace or a procurement cycle.
The trigger is a repeating moment, not a vague need: the lesson ends, the follow-up message must go out, and it must contain practice. At that moment the tutor is tired, unpaid for admin, and needs a concrete artifact in the next five minutes.
The problem in operational terms
A worksheet gets photocopied and ignored. A link from a generic quiz builder bears another company's logo and tells the parent nothing about what was practiced. So the tutor either eats 20-40 minutes writing something bespoke, or sends nothing meaningful, and the family's perception of value drifts down before the renewal conversation.
The cost is measurable in two currencies the tutor already tracks: unpaid admin minutes per week, and renewal rate. Skill Sprint Cards attacks both by producing the game and the parent-facing explanation in one action.
Positioning
Skill Sprint Cards is not a learning platform, an LMS or a curriculum. It is a lesson-end artifact generator. The tutor supplies the pedagogy — their skill, their words, their student — and the product supplies the production work: a constrained game spec, a rendered HTML5 mini-game on a permanent URL, and a four-sentence parent note card in the tutor's accent colour and name.
The wedge is deliberately narrow: one tutor, one skill, one link, one note. That narrowness is what makes it sellable inside a single conversation and operable by one founder with no support team.
Why this wedge and not a broader one
- The artifact is delivered in the same message thread where the tutor discovers the product, so trial-to-paid needs no second touchpoint.
- The tutor can approve $149 alone; there is no purchase order, no manager and no district review.
- Output is static HTML plus a PNG, so marginal delivery cost stays around a dollar per tutor per month.
- No student accounts, no progress tracking and no classroom integration means no compliance or data-retention burden.
Honest read on the score
JEV scored this concept 60.8/100 (band C, weighted 6.08). Speed to first revenue (8.6) and solo build feasibility (7.56) are the strengths; market demand (3.92) and monetization potential (4.31) are the weaknesses. This plan answers those two weaknesses directly: the buyer is narrowed to the lesson-end trigger rather than "tutors" generally, and the price is defended with a time-outcome pitch (20-40 unpaid minutes recovered per lesson) plus a referral loop that produces additional licenses without a subscription. The plan does not assume demand is proven. It assumes demand is testable in seven days with 100 DMs and a free built artifact, and it sets explicit kill criteria.
Smallest sellable MVP
A public generator that takes a skill, 8-12 tutor words, a level and a tutor name, returns a permanent playable link plus a downloadable parent note PNG, and gates new-game creation behind a $149 Stripe license key after one free card.
Must-have features
- Generator form: skill selector, word list input, reading level, tutor name, accent colour
- Schema-constrained LLM call producing a validated three-round game spec with template fallback
- Deterministic static HTML5 player rendering tap-the-answer, blend-the-sounds and timed recall
- Four-sentence parent note rendered to a downloadable PNG card
- Permanent public slug URL per game with a per-link play counter
- Minimal tutor dashboard listing links with copy and regenerate buttons
- Stripe Checkout issuing a signed one-time license key
- Landing page with a live sample game playable without signup
Explicitly out of scope
- Student accounts, logins or cross-session progress tracking
- LMS, Google Classroom or school-district integrations
- Native iOS or Android builds
- Team seats, studio admin roles or multi-user billing
- Pre-made game marketplace or content library
- Analytics beyond a per-link play counter
- Email automation sequences or drip campaigns
- Any human done-for-you service layer
Demo flow (what the buyer sees)
- Founder opens the landing page and plays the live sample game in-browser.
- Tutor pastes 10 CVC short-a words, picks the skill, types their name and accent colour.
- System calls the LLM with a strict JSON schema, validates the spec and falls back to a template on failure.
- Player route renders the three rounds at /g/{slug} and the parent note PNG is generated.
- Tutor copies one link and texts it in the lesson follow-up message.
- Tutor opens the dashboard, sees the link and play count, and hits regenerate for the next lesson.
Technical implementation plan
Stack
Architecture
A Next.js app on Vercel holds three surfaces: the public landing page with a sample game, a signed-in-lite generator and dashboard, and the public player route /g/{slug}. Supabase stores one row per game (slug, skill, word list, validated spec JSON, tutor name, accent colour, play count) and one row per license key. Generation is a single server action that calls the LLM with a strict JSON schema, validates the result against a zod schema, and falls back to a deterministic template if validation fails twice. The player is a static React component that reads the spec and renders rounds; it is fully cacheable and needs no server compute per play. The parent note is drawn with Satori to a PNG at generation time and stored on Vercel Blob so the tutor can download or long-press save it. Stripe Checkout is a hosted page; on success a webhook writes a license row and issues an HMAC-signed key the browser stores locally.
Build steps
- Day 1: scaffold Next.js repo, Supabase tables (games, licenses), env vars, deploy a skeleton landing page.
- Day 1: write the zod game-spec schema and the deterministic template fallback.
- Day 2: build the generator form plus the LLM call, validation loop and retry-with-fallback.
- Day 2: seed a curated distractor bank per skill (CVC short vowels, blends, digraphs, number bonds) to constrain wrong answers.
- Day 3: build the three game modes and the public /g/{slug} route with edge caching.
- Day 3: add the per-link play counter write.
- Day 4: Satori parent note template and PNG download.
- Day 4: dashboard with copy-link, regenerate and saved word lists.
- Day 5: Stripe Checkout, webhook, license key gate on second-game creation, full deploy and smoke test.
- Day 5: live sample game embedded on the landing page.
AI components
- Schema-constrained game spec generation (rounds, prompts, distractors, feedback lines) with zod validation
- Four-sentence parent note generation from skill, words and observed difficulty
- Mandatory fallback to a deterministic template whenever validation fails twice, so generation never 500s
Pricing and monetization
Free Trial Card
Entry offerFor Any tutor reached by outreach
- One generated game link
- One parent note card
Tutor License
For Independent tutors and small studios who used the free card once
- Unlimited game links for one tutor
- Tutor-branded parent note cards
- Saved word lists per student
- Permanent hosted links with no expiry
Model: A single one-time $149 Tutor License per tutor, with exactly one free trial card so the value is demonstrated before any payment.
First revenue offer: Send me the 10 words your student is stuck on and I'll send back a 5-minute game plus a parent note under your name in 10 minutes — free, no signup. If the parent replies to the note, the unlimited version is $149 once, no subscription.
Expected ACV: $149
Upsell path: Revenue grows only through additional licenses: each paying tutor is asked for one introduction to another tutor in their own network, and a co-tutor or partner working under a separate name buys their own $149 license. No subscription, no seats, no upsell tier is added during the 30 days.
First-customer acquisition
Direct DM outreach to independent tutors who publicly list a booking page or a tutor Instagram account, leading with a free built artifact rather than a pitch.
Where to find them
- Tutor Instagram accounts posting early-reading or early-math session clips
- Public booking pages of independent reading tutors (Calendly/SimplyBook links in bios)
- Online-tutoring community threads and Facebook groups for private tutors
- Tutor directories listing 1:1 early-years specialists
- Hashtag feeds for phonics tutors and early-math tutors
- Reply threads under tutor-created phonics content
Exact outreach motions
- Log 100 tutor profiles into a sheet with handle, subject, booking link and one personal detail from their last three posts.
- Open with the observation, not the pitch: name the specific skill they posted about and offer one free card built from words they supply.
- Deliver the free game link and parent note PNG within 10 minutes of their reply, same thread.
- Ask one question only: did the parent reply to the note?
- At 48 hours, offer a second free card from the next lesson's words instead of a discount.
- Only after they confirm use, send the $149 one-time license link with the time-recovered framing.
- After payment, ask for one introduction to another tutor in their network with a forwarded draft message.
Message template
Hi {name} — you posted about {specific skill} last week. I built a tiny tool that turns a tutor's own word list into a 5-minute game plus a short parent note under your name. Send me the 10 words your student is stuck on and I'll send it back in 10 minutes, free, no signup. Worth a look?30-day execution roadmap
Week 1 · Build & validate · 7 days
Repo, schema, deploy
- Scaffold Next.js repo and Supabase tables for games and licenses.
- Write the zod game-spec schema and template fallback.
- Deploy a skeleton landing page to Vercel.
Deliverable: Live skeleton URL and database schema.
Generator form and LLM call
- Build the generator form with skill, words, level, name and accent colour.
- Wire the schema-constrained LLM call with one retry then fallback.
- Seed the curated distractor bank for CVC short vowels and number bonds.
Deliverable: Generation endpoint producing validated spec JSON.
Three game modes on a public route
- Build tap-the-answer, blend-the-sounds and timed recall rounds.
- Ship the public /g/{slug} route with edge caching.
- Add the per-link play counter.
Deliverable: Playable game link a stranger can open.
PNG note and dashboard
- Build the Satori parent note template and PNG download.
- Build the dashboard with copy-link and regenerate.
- Add saved word lists per student.
Deliverable: Downloadable parent card and working dashboard.
Stripe and license gate
- Ship Stripe Checkout, webhook and signed license key.
- Gate second-game creation behind the key, keep one free card.
- Embed a live sample game on the landing page and smoke test end to end.
Deliverable: MVP live and able to take money.
Sample games and Loom
- Build 5 sample cards from generic phonics word lists.
- Record a 60-second screen capture of one card being made.
- Write the two-sentence value line for the landing page.
Deliverable: Demo Loom and 5 public sample links.
Build the 100-tutor list
- Collect 100 tutor profiles with handle, subject and personal detail.
- Tag each by early-reading vs early-math.
- Draft the first 25 personalised DMs.
Deliverable: 100-row prospect sheet and 25 drafted messages.
Week 2 · Launch & first sale · 7 days
Send 25 DMs
- Send 25 personalised DMs.
- Deliver every free card within 10 minutes of reply.
- Log reply rate and objection wording.
Deliverable: 25 DMs sent, first free cards delivered.
Convert first user
- Ask every delivered user the parent-reply question.
- Send the $149 license link to anyone who confirms use.
- Fix any flow blocker found during delivery.
Deliverable: First paid license.
Second 25 DMs
- Send the next 25 DMs with the Loom attached.
- A/B the opening line against wave 1.
- Log which subject segment replies faster.
Deliverable: 50 total DMs sent, message variant data.
48-hour second card
- Send a second free card offer to all non-responders.
- Ask each delivered user for one introduction.
- Patch the top dashboard friction.
Deliverable: Follow-up sent to 100% of wave 1.
Third 25 DMs
- Send 25 more DMs.
- Post one value-first thread in a tutor community.
- Track free cards delivered vs licenses sold.
Deliverable: 75 total DMs, community post live.
Distractor quality
- Expand the curated distractor bank to blends and digraphs.
- Add a regenerate-without-cost path for bad rounds.
- Re-test ten generated cards manually for pedagogical errors.
Deliverable: Distractor bank v2 and zero known bad rounds.
Funnel math
- Recompute reply, delivery and conversion rates.
- Kill the weakest segment and double the best one.
- Ship the top three requested fixes only.
Deliverable: Funnel table with segment decision.
Week 3 · Repeat & systematize · 7 days
Introductions
- Send the forwarded-intro template to every user so far.
- Add a one-line referral ask to the delivery message.
- Track introductions received.
Deliverable: Referral template in use with every user.
1-5 person studios
- Compile 30 studio operators with public booking pages.
- Send studio-specific DMs naming their second tutor.
- Offer a card built for a student they name.
Deliverable: 30 studio DMs sent.
Shorter proof
- Record a 30-second version of the Loom.
- Put the sample game link in the DM signature line.
- Rewrite the value line around minutes saved.
Deliverable: 30-second demo live and linked.
Objection handling
- Write answers to the three most common objections.
- Add a one-screen FAQ to the landing page.
- Test a bonus of two pre-built cards on top of the license.
Deliverable: FAQ shipped and objection script.
Warm-list reactivation
- Re-message every delivered-but-unpaid user.
- Send each their second card unprompted.
- Log replies verbatim.
Deliverable: Warm list fully reactivated.
Fresh 25 DMs
- Send 25 new DMs from the remaining list.
- Post a second community thread.
- Fix any broken links reported.
Deliverable: Fresh outreach wave sent.
Profit and refunds
- Reconcile Stripe payments against deliveries.
- Handle any refund request same day.
- Recompute hackathon profit and business profit.
Deliverable: Reconciled ledger and updated funnel.
Week 4 · Scale what works · 6 days
Broad organic push
- Publish a build-in-public post with the sample game.
- Post in two tutor communities with no link-drop framing.
- Reply to every comment within two hours.
Deliverable: Public launch post live.
Zero-latency service
- Deliver every inbound card within 10 minutes.
- Log delivery time per card.
- Convert every confirmed user same day.
Deliverable: Delivery SLA held for all inbound.
Silent prospects
- Send a one-question DM to every silent prospect.
- Offer a card for their next lesson free.
- Track reactivation rate.
Deliverable: Silent list contacted once more.
Trust surface
- Add a plain-English guarantee line to checkout.
- Show play count on the dashboard more clearly.
- Add error logging so failed generations are visible.
Deliverable: Trust copy live, error logging on.
Last fresh cohort
- Send the final fresh DM cohort.
- Ask each paying tutor for a one-line testimonial.
- Log testimonials with permission.
Deliverable: Final outreach cohort sent.
Introductions at scale
- Ask all payers for one named introduction again.
- Draft the forwarded message for them.
- Deliver an intro card free to each named contact.
Deliverable: Referral contacts in the funnel.
Days 29-30 · Consolidate · 3 days
Operate without support
- Fix the top three failure modes from logs.
- Write a two-paragraph self-serve help page.
- Test the license gate from a clean browser.
Deliverable: Self-serve help page and stable gate.
Last call
- Send a final-call DM to every warm contact.
- Extend the free card to anyone who never replied.
- Deliver all outstanding cards.
Deliverable: All warm contacts closed or closed-out.
Reconcile and submit
- Export the full Stripe ledger for the window.
- Confirm every sale covered its delivery cost.
- Submit the final verified-profit figures.
Deliverable: Reconciled profit figure and submission.
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
- Payment processing assumed at Stripe 2.9% + $0.30 per transaction on every scenario.
- Conservative: 100 DMs produce 25 free cards and 5 licenses; no paid marketing is bought.
- Base: 150 DMs plus community posts produce 12 licenses; no paid marketing is bought.
- Upside: 200 DMs plus referral introductions produce 22 licenses; $100 of paid marketing is tested on one tutor newsletter slot.
- Operating cost covers Vercel, Supabase, domain and LLM tokens at roughly $1 per active tutor plus $20 fixed monthly.
- No existing customers, revenue or partnerships are claimed; all three scenarios are projections.
Conservative
5 customers- Refunds
- $0
- Chargebacks
- $0
- Payment fees
- $23
- Paid marketing
- $0
- Hackathon profit
- $722
- Operating costs
- $30
- Business profit
- $692
Base case
12 customers- Refunds
- $149
- Chargebacks
- $0
- Payment fees
- $56
- Paid marketing
- $0
- Hackathon profit
- $1,583
- Operating costs
- $40
- Business profit
- $1,543
Optimistic
22 customers- Refunds
- $149
- Chargebacks
- $149
- Payment fees
- $102
- Paid marketing
- $100
- Hackathon profit
- $2,778
- Operating costs
- $60
- Business profit
- $2,718
Model your own numbers
Runs in your browser · nothing is saved- Collected revenue
- $1,788
- Payment fees (3.9%)
- $70
- Paid marketing
- $0
- Hackathon profit
- $1,718
- Business profit
- $1,678
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 three scenarios are projections built from stated assumptions, not verified revenue, existing customers or real traction. Rankings are based on verified profit collected inside the contest window.
Main risks and validation experiments
Risk 1Cold DM response rate collapses far below the assumed level, starving the funnel.
Lead with a free built artifact instead of a pitch and cap each DM at two sentences of context.
Send 25 DMs in one day with the free-card offer and count replies within 24 hours.
Fewer than 4 replies from 50 DMs means the channel is wrong; switch to community-thread delivery or stop.
Risk 2Tutors use one free card and never return because the need is weekly, not daily.
Build the second card unprompted from their next lesson's words at 48 hours.
Track how many of the first 12 free-card users open the dashboard again within 7 days.
Under 20% return within 7 days means the artifact is a novelty; reposition or abandon.
Risk 3LLM-generated content produces pedagogically wrong distractors and destroys trust with trained reading teachers.
Constrain output to a curated distractor bank per skill, validate with zod and fall back to a template.
Hand-review 20 generated cards against the target skill and count any invalid option.
More than 1 invalid option in 20 cards after bank v2 means generation is not safe to sell.
Risk 4Price resistance: free quiz builders and worksheet sites feel adequate at $149.
Anchor the price to 20-40 recovered unpaid minutes per lesson and a parent-facing artifact competitors do not produce.
Offer the license to 10 delivered users and record the exact objection wording for every no.
Zero purchases from 20 delivered free cards means the offer, not the price, is broken; rewrite the offer before discounting.
Risk 5A single founder cannot hold a 10-minute delivery SLA while also converting and building.
Batch outreach into fixed windows and rely on generation speed rather than manual work per card.
Time 10 consecutive deliveries and record median minutes from reply to link sent.
Median delivery above 30 minutes for two consecutive days means cut outreach volume in half.
Risk 6Parents ignore links from an unfamiliar domain, weakening the tutor's value story.
Put the tutor's own name and accent colour on both the game and the note card so the artifact reads as the tutor's.
Ask every delivered user the parent-reply question and tally yes/no.
Under 30% of tutors report any parent engagement means the note needs a different delivery format.
Hour-by-hour: the first 24 hours
Hours 1-2
- Create the Next.js repo, connect Vercel and Supabase.
- Create the games and licenses tables.
- Deploy the default page to a live URL.
Output: Live skeleton URL and database schema.
Hours 3-5
- Write the zod game-spec schema for three rounds and five prompts per round.
- Write the deterministic template fallback generator.
- Write the curated CVC short-a distractor bank.
Output: Schema file and fallback generator committed.
Hours 6-8
- Build the generator form with skill, word list, level, tutor name and accent colour.
- Wire the schema-constrained LLM call with one retry then fallback.
- Test with three real phonics word lists.
Output: Working generation endpoint returning validated spec JSON.
Hours 9-11
- Build the tap-the-answer and blend-the-sounds rounds.
- Create the public /g/{slug} route and render a game end to end.
Output: A playable game link openable on a phone.
Hours 12-14
- Build the 100-row prospect sheet template with the fields needed for personalisation.
- Write the DM template and two variants.
- Draft the first 10 personalised DMs ready to send.
Output: Prospect sheet plus 10 drafted DMs.
Full 30-day launch report
Read the full generated report
Skill Sprint Cards — 30-Day Launch Report
1. Product opportunity and positioning
Skill Sprint Cards is a lesson-end artifact generator for independent early-reading and early-math tutors. The buyer is one person with a public booking page or tutor Instagram account, selling 1:1 sessions in the US, UK, Canada or Australia. The trigger is repeating: the lesson ends, a follow-up message must go out, and it must carry practice.
The current options are bad. A photocopied worksheet gets ignored. A generic quiz builder link carries someone else's logo and tells the parent nothing. So the tutor either spends 20-40 unpaid minutes building something bespoke or sends something forgettable, and the family's sense of value erodes before the renewal conversation.
The product produces the two artifacts the tutor actually sends: a three-round HTML5 mini-game on a permanent URL and a four-sentence parent note card rendered as a shareable PNG, both under the tutor's own name and accent colour, both generated in under 90 seconds from the tutor's own word list.
This is deliberately not a platform. No student accounts, no progress tracking, no LMS, no app store. That narrowness is the whole strategy: the artifact is delivered inside the same message thread where the tutor discovers the product, so trial and payment can happen in one day with no procurement and no second touchpoint.
The JEV score is 60.8/100 (band C, weighted 6.08). Speed to first revenue (8.6) and solo build feasibility (7.56) are the strengths. Market demand (3.92) and monetization potential (4.31) are the weak points. This plan answers them by narrowing the buyer to a specific trigger event and by defending $149 with a time outcome rather than a feature list.
2. Smallest sellable MVP
The sellable thing is the $149 Tutor License: unlimited generated game links under the tutor's name plus parent note cards, with saved word lists per student and permanent hosted links. One free trial card exists so the value is demonstrated before money changes hands.
Shipped in four build days: the generator form; a schema-constrained LLM call with a deterministic fallback; a static player with tap-the-answer, blend-the-sounds and timed recall; a Satori-rendered parent note PNG; public slug routes with a play counter; a dashboard with copy and regenerate; Stripe Checkout issuing a signed one-time license key; and a landing page with a live sample game playable without signup.
Out of scope for the full 30 days: student logins, cross-session progress, LMS or Classroom integrations, native apps, team seats, a pre-made content library, and any done-for-you service layer.
Demo flow: a visitor plays the sample game, pastes ten CVC short-a words, types their name, and gets a link plus a note card they can text immediately.
3. Technical implementation plan
Next.js on Vercel, Supabase Postgres for game specs and license keys, a schema-constrained LLM call, Satori for the PNG, Stripe Checkout plus signed key verification, and edge caching on the public player route.
Generation is one server action: call the model with a strict JSON schema, validate with zod, retry once, then fall back to a deterministic template so generation never fails visibly. The player is a static component reading validated JSON, so plays cost no server compute. The parent note is drawn at generation time and stored so the tutor can download or long-press save.
The hardest technical risk is pedagogical wrongness — an LLM inventing non-words or off-pattern distractors — which would destroy credibility with trained reading teachers. The mitigation is a curated distractor bank per skill, schema validation, template fallback, and a free regenerate that never charges.
4. Pricing and monetization
One price: $149, one-time, per tutor. Plus exactly one free trial card. No subscription, no seats, no tiers added during the window.
The pitch is time, not features: 20-40 unpaid admin minutes recovered per lesson, plus a parent-facing artifact no worksheet or quiz builder produces. $149 is approvable by a solo tutor without a purchase order.
Expansion comes only from additional licenses: every paying tutor is asked for one named introduction to another tutor, and a co-tutor working under a separate name buys their own license. Processing is Stripe at 2.9% + $0.30.
5. First-customer acquisition
The only channel is direct DM to tutors with a public booking page or tutor Instagram account, leading with a free built artifact.
Log 100 profiles with handle, subject, booking link and one real detail from recent posts. Open with that detail, not a pitch. Deliver the free game within 10 minutes of their reply. Ask exactly one question: did the parent reply to the note? At 48 hours, offer a second free card from the next lesson's words instead of a discount. Only after confirmed use, send the $149 link.
Week-one goal: 100 profiles logged, 40 DMs sent, 12 free cards delivered, at least one paid license.
6. 30-day roadmap
Days 1-5 are build: schema and scaffold, generator with fallback, player routes, parent card and dashboard, then Stripe and the license gate. The MVP is live on day 5 and the first sale is targeted for day 9.
Days 6-14 are the first three outreach waves plus hardening: sample cards, the 100-row prospect sheet, 25 DMs on day 8, conversion attempts on day 9, then waves of 25 with 48-hour follow-ups. Day 13 expands the distractor bank to blends and digraphs. Day 14 recomputes the funnel and kills the weakest segment.
Days 15-21 add the referral loop and the studio segment, shorten the demo to 30 seconds, and rebuild the offer around minutes saved. Day 21 reconciles Stripe against deliveries and recomputes both profit figures.
Days 22-30 go public, hold a 10-minute delivery SLA, reactivate silent prospects, add a plain guarantee line and error logging, run a final outreach cohort and a last-call push, then export the ledger and submit on day 30.
7. Revenue and profit scenarios
All three scenarios are projections from stated assumptions, not verified revenue or customers.
Conservative: 5 licenses at $149 = $745 gross. Refunds $0, chargebacks $0, processing $23, paid marketing $0, operating $30. Hackathon profit $722; business profit $692.
Base: 12 licenses at $149 = $1,788 gross. Refunds $149, chargebacks $0, processing $56, paid marketing $0, operating $40. Hackathon profit $1,583; business profit $1,543.
Upside: 22 licenses at $149 = $3,278 gross. Refunds $149, chargebacks $149, processing $102, paid marketing $100, operating $60. Hackathon profit $2,778; business profit $2,718.
Processing is modelled at Stripe 2.9% + $0.30. Operating covers Vercel, Supabase, domain and LLM tokens at roughly $1 per active tutor plus $20 fixed monthly. Every scenario keeps revenue per sale well above delivery cost.
8. Risks and validation experiments
Four tests decide the concept inside two weeks.
DM response: 25 DMs in one day, count replies in 24 hours. Fewer than 4 replies from 50 DMs means the channel is wrong.
Return use: track how many of the first 12 free-card users reopen the dashboard within 7 days. Under 20% means the artifact is a novelty.
Content safety: hand-review 20 generated cards for invalid distractors. More than one bad card in 20 after the bank v2 update means generation is not sellable yet.
Price acceptance: offer the license to 10 delivered users and record verbatim objection wording. Zero purchases from 20 delivered cards means the offer is broken, not the price — rewrite before discounting.
Secondary risks: a founder cannot hold a 10-minute delivery SLA while converting and building (test median delivery time, halve volume if it exceeds 30 minutes twice), and parents may ignore links from an unfamiliar domain (tutor name and accent colour on both artifacts, measured by the parent-reply question).
9. First 24-hour plan
Hours 1-2: repo, Vercel and Supabase connected, tables created, skeleton deployed.
Hours 3-5: zod game-spec schema, deterministic fallback generator, CVC short-a distractor bank committed.
Hours 6-8: generator form and schema-constrained LLM call with retry and fallback, tested on three real phonics lists.
Hours 9-11: tap-the-answer and blend-the-sounds rounds, public /g/{slug} route, one game playable on a phone.
Hours 12-14: 100-row prospect sheet template, DM template plus two variants, first 10 personalised DMs drafted and ready.
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 →Branchline SCORM Compiler
Turns a plain-text SOP or policy into a playable branching decision scenario, exported as a ready-to-upload SCORM 1.2 zip.
Hookframe
Hookframe turns an app icon, a few screenshots and one chosen mechanic into a single-file, MRAID-safe playable ad under 5MB that an indie mobile studio can upload to Meta or AppLovin the same day.
Crash Course Press
Turns a licensing-exam instructor's own notes plus the official exam content outline into a branded, print-ready practice booklet and timed cohort quiz links in under 30 minutes.