CurioQuest — Model Validation & the Fastest Credible Path to $1M ARR

Status: v1.0 · Date: 2026-06-02 Method: The revenue-model.html v8 engine was ported verbatim to Node (sims/engine.js) and tested across four angles — (A) math reproduction, (B) structural validity, (C) cost-reality, and a 50k-run Monte Carlo + cash-constrained reconciliation. All code in sims/. Re-runnable: node sims/01-validate.js … 06-recommended-path.js. One-line verdict: The model is real on revenue, optimistic on profit. It is a sound calculator of a near-best case, not a central forecast — and the path to $1M ARR is an acquisition problem, gated by a cash-safe organic engine, not a capital problem.

Part 1 — Is the model even correct and real? (validation first, as required)

Gate A — Math reproduction: PASS (6/6)

The Node port reproduces every documented figure exactly: Realistic Y1 net $6,763, 3-yr cumulative profit $215,725, ARR end-Y3 $338,512, run-rate M36 $60,324/mo, ~10,782 customers, 3.3× LTV:CAC. Internal invariants hold (Σ monthly net = cumProfit; ARR = active×price; gross margin in range). The calculator adds up. [FACT]

Gate B — Structural validity: FOUR material distortions found

Reproducing numbers proves arithmetic, not realism. Probing the structure:

Plus: no refunds/chargebacks (10% refund → −35% profit), and LTV is hyper-sensitive to retention via the 1/(1−rt) term (LTV:CAC swings 2.4×→17× across plausible retention).

Gate C — Cost side: broadly sound

Stripe 2.9%+$0.30 ✓ · POD $14/book (Lulu/Gelato range $12–22) ✓ · fixed $150/mo ✓. Only mild optimism: AI-gen at $1.50/book all-in is tight for a 24-page illustrated book (~12–20 image generations); $2–3 is safer. The fragility is almost entirely on the demand side. [EST]

The decisive test — where does "Realistic" actually sit?

A 50,000-run Monte Carlo across the full plausible range of every unproven driver, with the Gate-B corrections applied (rising CAC + refunds + the real no-debt cash constraint):

• "Realistic" ARR ($338K) ranks at the 49th percentile of honest outcomes → the revenue model is well-centered. It is REAL. ✅
• "Realistic" profit ($216K) ranks at the 99th percentile → the profit projection is a near-best case, not a forecast. ❌
• Cash-constrained reality (the true $1k bootstrap): P25 profit ≈ −$3K to −$5K — you essentially cannot lose more than your seed. The "$60K–$300K capital needed" downsides are artifacts of unconstrained spending.
• The real failure mode is not "lose big" — it's "stall." Honest cash-constrained median ARR is ~$1K–$65K, not $339K, because growth self-limits when CAC outruns fundable cash.
• P($1M ARR by month 36) under the current structure ≈ 0.3–3.7%. A tail event, not a plan.

Validation conclusion: Trust the model for revenue shape and unit mechanics. Do not trust its profit line or its implied "$1k is plenty to print $216K." Three fixes make it honest: (1) model CAC rising with scale, (2) add a refund rate, (3) cap marketing spend at available cash.

Part 2 — The fastest credible, cash-safe path to $1M ARR

Run on the corrected engine (rising CAC + refunds + no-debt constraint + LTV:CAC≥3 gate).

$1M ARR = active annual subscribers × plan price. Price is the cheapest lever: at $149/yr you need 6,712 subs, vs 10,102 at $99 — a 33% smaller mountain, at zero CAC or cash cost.

Baseline reality check: the current plan (Realistic + honest corrections) never reaches $1M in 60 months and ends at LTV:CAC 1.2×. Single levers don't fix it. Stacking does:

Two findings that should reshape strategy:

1. Outside capital does not accelerate the healthy paths. Funded vs bootstrap for Paths A & B = same month, $0 needed — upfront annual billing self-funds them. Capital only "helps" Path C, the one with broken economics. Raising money here would only buy unprofitable growth. The bootstrap path is genuinely optimal.

2. The constraint on speed is reach/growth, not money or retention. Path B robustness:

Reaching $1M is an acquisition result — retention barely moves the timing (most subs are <12 months old when you cross $1M). Retention governs durability after M24: M48 ARR ranges $2.8M (50% retention) to $3.9M (85%). It compounds into Year 4, not the sprint.

The clean framework (what each lever actually controls)

• Feasibility of $1M at all → plan attach × plan price (subs-per-customer × $/sub).
• Speed to $1M → acquisition growth rate + reach ceiling.
• Durability beyond $1M → retention.
• Profitability of the whole path → CAC staying low at scale (the organic/viral engine).

Part 3 — Optimized plan & what the pilot must prove

Optimized targets (Path B): plan price $149/yr · plan attach ≥45% · retention ≥75% · organic/viral CAC ≤$22 and held low at scale · grow to ~2,000 new customers/mo. Bootstrap-funded end to end; do not raise capital to accelerate (it only buys the unprofitable path).

The $1k concierge pilot must validate, in priority order: 1. Plan attach ≥ 40% — below ~30%, no $1M path exists at any speed. (pilot-provable in 8 wks) 2. Preview→paid ≥ 6% — the hinge metric; funds the acquisition engine. (provable) 3. WTP at $149/yr — the price that turns 6,700 subs (not 10,100) into $1M. (provable) 4. Organic CAC ≤ $25 that holds at scale — the single profitability gate. (partially provable) 5. Retention ≥ 70% — governs durability; the one true leap of faith (needs a year — pilot can only proxy it via intent + early re-purchase).

Next model action: patch revenue-model.html with the three honesty fixes (rising-CAC curve, refund input, cash-constrained acquisition) and add a "$149 plan / month-to-$1M-ARR" readout, so the dashboard the founder reads reflects the corrected economics rather than the near-best case.

Evidence: all figures from sims/*.js against the verbatim v8 engine port. FACT = reproduced from the model; EST = benchmarked external cost; ASSUMPTION = unvalidated demand driver (the pilot's job).