The Economics of a Solo Studio

Learning goal: Compute the hourly cost, fixed and variable costs, Runway and break-even point with the example studio's numbers, and explain what the power-law distribution of product outcomes means for strategy.

As of: 2026-09-29. All example numbers are assumptions, and taxes are left out of the calculations.

Key Concepts

Principles

Time as the Scarcest Resource

In a solo studio, time runs out before money does. Dividing the monthly cost by workable hours gives the baseline hourly cost.

Baseline hourly cost = (living cost KRW 3,000,000 + fixed cost KRW 500,000) ÷ 160 hours = 3,500,000 ÷ 160 = KRW 21,875

Compare every task against this amount. Spending 40 hours on a new prototype spends 40 × 21,875 = KRW 875,000 worth of time. On top of that comes the opportunity cost: what those 40 hours would have produced if spent marketing a product already launched.

Fixed Cost, Variable Cost and Contribution Margin

Living costs are not a business expense in accounting terms, but when judging a solo studio's survival they behave like a fixed cost. The break-even quantity is:

Break-even quantity = monthly fixed cost ÷ contribution margin per unit

Example (assumption): a KRW 5,000 monthly subscription, a 15% store fee, and an AI API cost of KRW 750 per user per month.

A product with AI features pays an API cost for every use. That variable costs can be larger than in traditional software is a new variable in the economics of products in the AI era.

Runway and Default Alive

Runway = KRW 18,000,000 ÷ KRW 3,500,000 ≈ 5.14 → about 5.1 months (with zero revenue)

In his 2015 essay "Default Alive or Default Dead?", Paul Graham defined a company as default alive if, extending current costs and revenue growth, it reaches profitability before the money runs out, and default dead otherwise. The two scenarios below (assumptions) grow net revenue by the same KRW 500,000 each month and differ only in starting one month apart.

Amounts are in KRW. A reaches break-even in month 9 with KRW 500,000 left (default alive). B hits zero cash at the end of month 6 (default dead). A launch that is one month late is enough to flip survival. This is why launch speed is an economic question for a solo studio.

flowchart TD
    C[Operating cash KRW 18M] --> B[Monthly spend KRW 3.5M]
    R[Monthly net revenue] --> N[Monthly net burn = KRW 3.5M - net revenue]
    B --> N
    N --> RW[Runway = cash / monthly net burn]
    RW --> Q{On current trends, profitable before cash runs out}
    Q -->|yes| A[Default alive: room for growth experiments]
    Q -->|no| D[Default dead: bring launch and revenue forward or cut costs]
    D --> R

The Power Law of Product Outcomes

In such a distribution the mean is pulled up by a few huge hits and must not be used for planning. Plan on the median or below, and treat the big wins in the tail as "a bonus that can appear after many launches". And to reach the tail, you have to ship first.

Applied: The Example Studio

Diagnosis: with zero revenue and no trend, the studio is default dead today. Splitting hours evenly across 12 products leaves just over 13 hours per product per month, so none of them reaches launch and marketing. Document 02 handles this problem with WIP limits.

Going Deeper

What AI Agents Change and What They Do Not

Break-even Seen Through 1,000 True Fans

In his 2008 essay "1,000 True Fans", Kevin Kelly argued that a creator can make a living from 1,000 true fans who each spend USD 100 a year. Applying the same logic to the example studio (assumption: one fan spends KRW 100,000 a year):

Annual net revenue needed = KRW 3,500,000 × 12 = KRW 42,000,000 → 42,000,000 ÷ 100,000 = 420 people

A solo studio's advantage is that it can aim for hundreds of paying users, not millions.

Common Misconceptions

• "Thanks to AI, building costs almost nothing" — The agent may be cheap, but your time costs KRW 21,875 an hour.
• "Plan from the average revenue" — Under a power law the mean is inflated by a few. Plan on the median or below.
• "Runway is balance ÷ business fixed costs" — In a solo studio, living costs belong in the denominator too.
• "Revenue is income" — Only after fees, API costs and taxes do you see what is really left.
• "A one-month launch delay makes little difference" — As the scenarios show, one month can flip survival.

Self-Check Questions

1. Compute your baseline hourly cost from your monthly costs and workable hours.
2. With a KRW 8,000 monthly subscription, a 15% fee and an API cost of KRW 1,000 per user per month, how many subscribers break even?
3. In scenario B, what is the minimum monthly increase in net revenue that would have kept cash from running out? (Recompute the table.)
4. Where would you look for median, not mean, figures for your product category?
5. Can you describe one way an AI feature increases variable cost?

References

• Default Alive or Default Dead? — Paul Graham (2015-10, accessed 2026-09-28)
• 1000 True Fans — Kevin Kelly, The Technium (2008-03, accessed 2026-09-28)
• State of Subscription Apps 2025 — RevenueCat (2025-03, accessed 2026-09-28)
• The State of Subscription Apps 2025: The year AI ate everything — RevenueCat Blog (2025-03, "after two years" basis confirmed, accessed 2026-09-29)
• Indie games come close to AA/AAA games in revenue on Steam — Game World Observer (2024-10-16, citing the VG Insights report, accessed 2026-09-28)
• More than 19,000 games launched on Steam this year—but almost half have fewer than 10 reviews — PC Gamer (2025-12, accessed 2026-09-28)