Revenue Models: An Introduction

Learning goal: Distinguish what each of nine revenue models needs, compute fit by product type and unit economics by model, and read the main platform and payment fees as of 2026-09-29 together with their sources.

As of: 2026-09-29. Fees change often, so re-check the official documents in the table before acting. All example numbers are assumptions, and taxes are left out.

Key Concepts

Merchant of Record (MoR): A payment arrangement in which the provider legally becomes "the seller" and handles payment, per-country VAT and sales-tax calculation and remittance, and refunds on your behalf. Its fees are higher than an ordinary payment processor's, but it reduces the burden of foreign tax compliance.

Principles

Each Model Needs Different Things

The most common solo-studio mistake is attaching ads to a product with no traffic, and a subscription to a product whose retention is unproven.

Fit by Product Type

The table below is an example (assumption) that shows the reasoning. Game details are in 07 and pricing design in 06.

flowchart TD
    Q1{Is the problem painful enough that users pay directly} -->|yes| Q2{Does the value recur every time}
    Q1 -->|no| Q3{Can you gather hundreds of thousands of pageviews a month}
    Q2 -->|yes| SUB[Subscription or usage-based]
    Q2 -->|no| ONE[One-time purchase]
    Q3 -->|yes| ADS[Advertising]
    Q3 -->|no| Q4{Do you have a small group of devoted fans}
    Q4 -->|yes| SP[Sponsorship or membership]
    Q4 -->|no| RV[Go back to 03 and revalidate the problem]
    SUB --> FM{Do free users become a discovery path}
    ONE --> FM
    FM -->|yes| FR[Combine with freemium]
    FM -->|no| DIR[Paid only]

Unit Economics by Model: The Scale Needed for KRW 3,500,000 of Monthly Net Revenue

Judge a subscription by LTV rather than monthly revenue. With 8% monthly churn (assumption), LTV ≈ 4,250 ÷ 0.08 = KRW 53,125. Churn math and pricing design are covered in depth in 06.

Platform and Payment Fees (as of 2026-09-29)

Applied: The Example Studio

For each prototype type, pick only one model to test first (assumption).

Combination example (assumption): ads KRW 500,000 (166,667 PV at an RPM of KRW 3,000) + creative tool one-time sales KRW 1,500,000 (KRW 8,500 × 177) + productivity app subscriptions KRW 1,500,000 (KRW 4,250 × 353) = about KRW 3,500,000 (KRW 3,504,750). Stacking three small paths is also safer, from the correlation angle of 02, than gathering 1,166,667 PV or 824 subscribers through a single model.

Going Deeper

The Cost of Mixing Models

Every added model adds one more payment integration, refund policy, set of terms and metric. For a solo studio, start from one main model per product plus at most one supporting model.

Rule for adding a supporting model: do not attach a supporting model until the main model has passed G4 in 02 (60 days after launch: monthly net revenue of KRW 100,000 or more, or 100 weekly active users). Adding a model before the main one has produced a signal makes it impossible to tell which one worked, and only doubles the payment, terms and metrics work. Even after it passes, write the supporting model's threshold and period first, as in 03, and remove it if it lowers the main model's key metric (for example, trial to payment). Example: T1 tests a subscription asset pack only after its one-time sales pass G4. C2's main model is ads, and the link from C2 to T1 is not a revenue model but distribution (02, Going Deeper).

Break-Even Between an MoR and a Direct PG

"An MoR is expensive" is true if you look only at fees; once tax work is counted, the answer depends on sales volume. The calculation uses 06's overseas one-time sale ($29, KRW 40,600 at an assumed KRW 1,400). All numbers are assumptions.

Difference per sale = 2,730 − 1,380.4 = KRW 1,349.6 → break-even = 96,667 ÷ 1,349.6 ≈ 71.6 sales → the MoR is cheaper up to 71 sales a month, and the direct PG is cheaper from the 72nd sale (break-even gross about KRW 2.91M)

• How to read it: T1's day-60 net revenue of KRW 450,000 (09) is about 12 sales a month at 06's $29 price (450,000 ÷ 37,870). That is far below break-even, so the MoR is cheaper for the example studio.
• Left out: the PG sign-up fee of KRW 220,000, tax adviser fees, time spent on chargebacks and fraud, and domestic VAT filing (needed on both paths). Foreign-card rates differ by contract and may exceed 3.4% (to be confirmed). The higher the rate, the higher the break-even volume, and the longer the MoR stays ahead.
• VAT itself: VAT on digital products sold to consumers abroad, such as in the EU, arises on either path. An MoR calculates and collects it at checkout automatically; with a direct PG the seller must add each country's VAT at checkout or, failing that, absorb it. Absorbing 20% VAT on a $29 sale costs 29 × 20% × 1,400 = KRW 8,120 per sale, about six times the fee gap above (KRW 1,349.6). Without checkout set up to add each country's VAT, the direct PG is more expensive at any volume.

Checking Model, Channel and Platform Correlation

Use the three axes from 02's Going Deeper (revenue model, acquisition channel, platform) to check the example studio's combination (ads KRW 0.5M + creative tool one-time sales KRW 1.5M + productivity app subscriptions KRW 1.5M). Two paths score 1 when they share an axis, 0.5 when they partly share it and 0 when they do not (the scoring is an assumption).

Making the same KRW 3.5M with three ad-supported content sites would score 3 on every pair: 9 / 9.

• Shock test (assumption): if Google search traffic falls 50% and half of B's traffic comes via search and C2, A becomes KRW 0.25M, B becomes 1.5 × (1 − 0.5 × 0.5) = KRW 1.125M, and C stays at KRW 1.5M in the short run → KRW 2.875M in total (−17.9%). A single ad path would have fallen to KRW 1.75M (−50%).
• The biggest single risk is the payment account. If the Paddle account stops, B's KRW 1.5M (42.9% of the total) disappears at once. That is why B and C sit on different payment paths.
• Order of the check: (1) write the three axes for each path (2) score each pair (3) run a shock test on the most shared axis (4) if one axis can shake half or more of the total, choose the next bet from candidates that do not share it.

Taxes and Displayed Prices

Prices to Korean consumers are displayed including VAT, while app-store and overseas revenue are handled differently for VAT. The net prices in these tables are illustrative numbers without tax. Read the Business section's "VAT · Zero Rate · Tax Invoices" alongside.

Common Misconceptions

• "Make it free and add ads" — Advertising is the model with the largest traffic requirement.
• "Subscription is the best model" — Without recurring value and low churn, a subscription only adds refunds and cancellations.
• "The fee is 30%" — In 2026, 10-15% is common for small developers depending on the payment method (for example, Google Play US: 10% on the first USD 1M a year plus a separate 5% billing fee with Google Play billing; South Korea currently 15%, 11% with alternative billing). It also varies by region and install date.
• "An MoR is just expensive" — Compare it with the time and tax cost of filing foreign VAT yourself.
• "Add the revenue model later" — The model changes onboarding, the free tier and the metrics.

Self-Check Questions

1. Pick one of your products and name the weakest of the four requirements in the table above.
2. With a KRW 9,900 monthly subscription and a 15% fee, how many retained subscribers produce KRW 3,500,000 of monthly net revenue?
3. With 10,000 active free users, 2% paid conversion and a net price of KRW 4,250, what is the monthly net revenue?
4. When Stripe is hard to use directly from South Korea, what are two payment options to consider for overseas sales?
5. In the example studio's combination, what would you have to grow, and by how much, to make the same KRW 3,500,000 without ads?

References

• App Store Small Business Program — Apple Developer (accessed 2026-09-28)
• Membership details — Apple Developer Program (15% on subscription renewals after one year, accessed 2026-09-28)
• Apple announces App Store Small Business Program — Apple Newsroom (2020-11, accessed 2026-09-28)
• Distributing apps using a third-party payment provider in South Korea — Apple Developer (accessed 2026-09-28)
• Service fees — Play Console Help (accessed 2026-09-28)
• Changes to Google Play's billing requirements for developers serving users in South Korea — Play Console Help (4-point cut for alternative billing, accessed 2026-09-29)
• Google Play Apps Experience Program — Google Play Console (program rates above USD 1M, accessed 2026-09-29)
• Understanding Google Play's lower service fees — Play Console Help (accessed 2026-09-28)
• Expanded billing choice and lower fees on Google Play — Android Developers Blog (2026-06, accessed 2026-09-28)
• Google settles with Epic Games, drops its Play Store commissions to 20% — TechCrunch (2026-03-04, accessed 2026-09-28)
• Valve creates new rev share tiers to give big sellers a break — Game Developer (2018-11, accessed 2026-09-28)
• Steam Direct Fee — Steamworks Documentation (accessed 2026-09-28)
• Introducing open revenue sharing — itch.io (2015, accessed 2026-09-28)
• Pricing — Stripe / Global availability — Stripe (accessed 2026-09-28)
• Pricing — Paddle (accessed 2026-09-28)
• Fees — Lemon Squeezy Docs (accessed 2026-09-28)
• 2026 Update: Lemon Squeezy + Stripe Managed Payments — Lemon Squeezy (2026, confirmed through search results, accessed 2026-09-29)
• Managed Payments pricing — Stripe Support (charged on the tax-inclusive total, accessed 2026-09-29)
• Stripe's Merchant of Record (Stripe Managed Payments): How Does it Work? — Paddle (secondary source from a competitor, states South Korea is excluded, accessed 2026-09-29)
• EU VAT One Stop Shop (OSS) — Your Europe (non-Union OSS, accessed 2026-09-29)
• Gumroad's fees — Gumroad Help Center (accessed 2026-09-28)
• A standard platform fee for new creators — Patreon Help Center (2025, accessed 2026-09-28)
• Updates to how publishers monetize with AdSense — Google (2023-11, accessed 2026-09-28)