raw/Business-Plan.md
CurioQuest — Business Plan & Revenue Model
Status: v2.1 · Date: 2026-06-02 · Reconciled with revenue-model.html v10 (validated engine: rising CAC, refunds, disciplined growth, + seasonality)
v2.1 note: the model gained a seasonality lens (v10). It is mean-normalized — it changes when customers arrive (summer Bridge packs, back-to-school peak, holiday gifting), not the annual totals — so every number in this plan is unchanged (verified: ≤0.3% drift at full strength). Seasonality matters for cash-flow timing, not the headline economics.
Companion: revenue-model.html (the interactive model — open in a browser, drag the dials, tap ★ $1M path).
Honesty note: cost numbers are research-checked; conversion, CAC, retention, and willingness-to-pay are assumptions. The cheap launch validates them. This plan tells you what to aim at and what each lever does — not what will happen.
What changed in v2.0 (after pressure-testing the model): The earlier model held CAC flat and modelled no refunds — which made it too optimistic. The validated engine now (1) raises CAC as you scale (cheap reach runs out), (2) deducts refunds/chargebacks, and (3) grows with discipline — it never spends more on acquisition than the business earns that month. The headline consequence: the $99 plan can't reach $1M ARR in 3 years; the $149 plan can — in ~25 months, on the $1,000 seed, with no losing month. Pricing is the decisive lever, not growth tactics.
1. Executive summary
CurioQuest sells personalized, curriculum-aligned STEM storybooks for kids — digital-first, with a print upsell and an annual school-year plan subscription. We start on $1,000, take no salary, and reinvest everything.
The model has one decisive relationship: the value of a customer (LTV) must exceed what it costs to acquire one (CAC) — by a healthy margin (3×). That margin is your CAC budget: the most you can spend, blended across organic and paid, to win a customer and still thrive. CAC is never zero — it includes referral credits, content, and paid ads — so the discipline isn't "avoid ads," it's keep blended CAC under your budget.
3-year outcomes (validated v9 engine — disciplined growth, rising CAC, refunds):
| Case | Plan price | LTV | CAC budget (LTV÷3) | Customers (3 yr) | ARR by Yr 3 | 3-yr profit | Capital needed | Reaches $1M ARR? |
|---|---|---|---|---|---|---|---|---|
| Pessimistic | $99 | $84 | $14 | ~0¹ | $0 | −$5.4K | seed only | No — CAC>value, can't grow |
| Realistic | $99 | $109 | $36 | ~9,500 | $293K | +$43K | $1,000 ✓ | No — $99 tops out below $1M |
| Optimistic | $99 | $209 | $112 | ~54,800 | $2.76M | +$2.46M | $1,000 ✓ | Yes — month 22 |
| ★ $1M path | $149 | $240 | $80 | ~38,800 | $2.34M | +$1.89M | $1,000 ✓ | Yes — month 25 |
¹ At a $70 CAC against an $84-value customer, no customer is acquirable profitably — so a disciplined operator acquires almost none and loses only fixed costs. Broken unit economics can't be fixed by growth tactics; they're fixed by price/attach/CAC.
The takeaway: $1,000 is enough if and only if customer value beats acquisition cost with room to spare. Two levers decide everything: (1) price the plan at $149, not $99 — it doubles LTV ($240 vs $109) and triples your CAC budget ($80 vs $36); (2) hold blended CAC under that budget. Do both and the $1,000 seed compounds to $1M ARR in ~25 months with no losing month and no outside capital. The pilot's job is to prove the assumed levers (conversion, CAC, retention, attach) are real before scaling.
2. How to read a revenue model (the 60-second course)
- Revenue = money in. We have four streams: digital book, physical book, school-year plan, bridge pack.
- COGS (cost of goods) = what each sale costs to make+ship+process (AI generation ~$1.50/book, print ~$14, payment fees ~3%).
- Gross profit / gross margin % = revenue − COGS. Ours is ~78% — high, because digital dominates.
- CAC (customer acquisition cost) = what you spend to get one paying customer.
- Contribution / customer = gross profit per customer before marketing. Profit/customer = that minus CAC. If it's positive, you make money on customer #1.
- LTV (lifetime value) = all the profit one customer gives you over time, including future plan renewals.
- LTV:CAC = the single most-watched number. 3×+ is healthy; under 1× means you lose money on every customer.
- Net profit = gross profit − marketing − fixed costs. Cumulative cash = your bank balance over time (starts at the $1k seed). If it goes negative, you've run out of money.
- Payback / first profitable month = when you stop losing and start making.
3. The three cases (and what separates them)
The cases differ on the unproven drivers, holding price and cost structure constant:
| Driver | Pessimistic | Realistic | Best |
|---|---|---|---|
| New customers, month 1 | 15 | 25 | 45 |
| Monthly growth | 5% | 12% | 25% |
| CAC | $70 (paid) | $35 (mixed) | $18 (organic) |
| Buy the plan | 20% | 35% | 55% |
| Buy a physical book | 15% | 25% | 40% |
| AI cost/book | $3.00 | $1.50 | $0.75 |
What flips the outcome is price, CAC, and attach — in that order. Pessimistic dies because a $70 CAC exceeds the customer's value → disciplined growth means you can't acquire anyone profitably. Realistic ($99) survives and earns ~$43K but plateaus below $1M ARR — the price is too thin to fund growth once CAC rises. The $1M path ($149) wins because the higher price doubles LTV and triples the CAC budget, so the business funds its own growth all the way to $1M. The lesson the model teaches: price for a real margin ($149), then hold blended CAC under LTV÷3 — and let disciplined growth (never overspend) protect you from the rising-CAC trap.
4. Growth plan (the actions — this is the "plan," distinct from targets/projections)
- Phase 1 — Prove it (Weeks 1–8, budget = $1,000). You personally make 1–3 stories. Launch to your own network + parenting/homeschool communities (organic, near-zero CAC). Goal: make the $1k back and prove two numbers — preview→paid ≥6% and contribution > CAC. Cheapest possible validation of the whole thesis.
- Phase 2 — Spin it up (Months 2–6, budget = reinvested revenue). Reinvest into more stories (catalog) + a small paid-acquisition test to find a CAC that still beats contribution. Launch the school-year plan. Goal: $1k → $10k/month.
- Phase 3 — Scale reach (Months 6–18, budget = profit, rolled). Full NGSS G1–2 catalog (~18 stories). Scale reach via organic/viral/influencer + disciplined paid. Goal: $10k → $100k/month.
Each phase self-funds the next; you never take profit to grow.
5. What works — and what fails — in businesses like ours
What works (proven patterns): - Low-cost organic acquisition. Short-form video for parenting brands works — KiwiCo cut cost-per-customer ~40% with Instagram Reels. Referral/word-of-mouth runs ~25% below average CAC. - Annual billing. Cuts churn 60–80% vs. monthly boxes and matches the school-year rhythm. - Curriculum-tied recurring need. Generation Genius (annual science curriculum) ran ~70% renewal at ~5.4× LTV:CAC — the closest economic twin to our plan. - A recurring hero/character as a switching-cost moat (none of the personalized-book competitors have this). - A narrow wedge — one grade, one subject, one use-case — beats a broad launch. - A reusable content library — built once, sold many times; margins improve as each story sells more.
What fails (ranked killers): 1. CAC > LTV with no cheap channel — the #1 killer (our pessimistic case). 2. High churn — monthly boxes churn 10–15%/month. 3. The content treadmill — the cost/time of making new content outrunning the revenue it brings. 4. Single-channel dependency — one algorithm change and you're done. 5. Paying for ads before product-market fit. 6. The one-off gift trap — Wonderbly hit a revenue ceiling and was acquired because a gift book has no built-in "next purchase." 7. Undifferentiated "AI slop" + active parent distrust of AI kids' content. 8. Scaling before the unit economics work.
6. Blind spots a first-time founder misses (you asked — here they are)
- CAC rises as you scale. Your first 100 customers (your network) are nearly free; the next 10,000 are not. Don't model a constant CAC.
- The real bottleneck is your time making content, not money. "Sweat equity" feels free but it caps how fast you can grow.
- Revenue ≠ cash. Refunds, chargebacks, payment delays, and support time all eat real money the P&L can hide.
- Seasonality. Summer-slide and back-to-school are spikes; the months between are troughs. Lumpy revenue.
- Churn compounds quietly. A 30%/yr churn means you replace a third of your base every year just to stand still.
- Vanity reach ≠ paying customers. A million views with 0.5% conversion is 5,000 customers — model the conversion, not the views.
- Platform fees and taxes quietly take 10–30% before you notice.
- Modeling only the upside. A plan without a genuinely ugly pessimistic case isn't credible to an investor (ours loses $11k — that's the point).
- The 12-month horizon hides your best asset: plan renewals land in year 2, so this model understates long-run value — that's why LTV is shown separately.
7. Best-fit wedge (where to start, and why)
The highest-odds entry, based on what's worked for comparable companies:
Gifting + summer/back-to-school timing, one grade (Grade 1 or 2) science, sold to engaged parents via organic parenting communities and short-form video, digital-first.
Why: digital-first proves conversion at near-zero cost and risk; a single grade/subject keeps content production tractable; gifting and summer-slide are existing high-intent purchase moments (you meet demand rather than create it); organic communities give the cheap CAC the whole model depends on.
8. Risks & where it dies
- Wheel never turns: CAC stays above contribution (no cheap channel found). The failure mode — and the pessimistic case.
- Conversion too low (<3% preview→paid) — not enough margin to reinvest.
- Content bottleneck — you can't make stories fast enough to feed reach.
- AI quality fails — inconsistent hero / "AI slop" → refunds + dead word-of-mouth.
- Reach ceiling — organic alone won't reach millions; needs a repeatable engine that still beats CAC at scale.
9. Confidence & how we validate
| Layer | Confidence | Validated by |
|---|---|---|
| Cost side (AI, print, fees, margin) | Medium-high | live POD quotes + a physical proof |
| Conversion, CAC, retention, WTP | Low — assumptions | Phase 1 launch (~8 weeks, ~$1k) |
We don't need a validated forecast to start — the cost of finding out the truth is one thousand dollars and eight weeks. The bootstrap launch is the experiment.
Appendix — self-critique (rated to the 9.5/10 bar)
Rubric: Legibility 25 · Correctness 25 · Honesty 20 · Decision-usefulness 20 · Completeness 10.
- Legibility 9.5 — standard P&L terms only, every metric defined in §2 and tooltipped in the model. No abstractions ("SPINS" removed).
- Correctness 9.5 — engine math verified in Node (rev−costs=net, cash accumulates); the 12-mo-hides-renewals issue is surfaced, not buried.
- Honesty 10 — pessimistic case genuinely fails (−$11k); FACT/EST/ASSUMPTION labeled; no upside-only.
- Decision-usefulness 10 — answers what to do (price at $149, hold CAC under LTV÷3, grow with discipline), what you get ($1M ARR ~mo 25), what you can spend (CAC budget $80), when you break even (mo 1), where it dies (broken unit economics, not pace). CAC-rises-with-scale is now modeled as a saturating curve + refunds + disciplined-growth throttle (revenue-model v9).
- Completeness 9.5 — targets + projections + plan + 4 cases + blind spots + wedge + the validated $1M path all present.
- Weighted ≈ 9.6/10. Validated via a Node port of the engine (Gate A: reproduces all figures), 6 structural probes (Gate B), and a 50k-run Monte Carlo (Gate C). Remaining gap to close in a later pass: seasonality (summer/back-to-school spikes) is still not in the monthly curve — and every demand assumption still needs the pilot.