Subscription vs In-App Purchases vs Ads: What Earns Most in 2026

Before any pricing discussion, understand what Apple takes, because every dollar that flows through Apple's In-App Purchase system has Apple's commission removed before you see it. The standard rate is 30 percent, but that applies to developers earning more than $1 million per year in proceeds. The Small Business Program rate is 15 percent, and any developer who earned under $1 million in the previous calendar year qualifies — Apple auto-notifies eligible developers, you apply through App Store Connect, and approval is usually fast. There is also a loyalty discount: Apple drops the commission to 15 percent on any subscription after a customer has been subscribed for more than one year, regardless of your overall revenue, which meaningfully boosts long-term subscription economics.

In practice, most indies in 2026 pay 15 percent on nearly every transaction because they qualify for the Small Business Program, so budget as if you keep 85 percent of gross. A $4.99/month subscription nets roughly $4.24 after Apple's 15 percent cut, and a $29.99 annual subscription nets about $25.49 — internalize those numbers when you price anything. One hard rule: digital goods and services must go through Apple's IAP for iOS apps. You cannot route users to your website to pay, Apple rejects for it, and in some cases has pulled apps entirely. Physical goods like Uber rides or Amazon packages are different and can use any processor, but for indie utility apps you live in the IAP world. The official Apple App Store developer documentation is the authoritative source on these commission tiers, and my Apple App Store review guide covers how the IAP rule interacts with approval.

For an indie working from home with no marketing budget, the monetization model is almost a survival question, because subscriptions compound retained users into recurring revenue while ads need a scale you cannot afford to buy. The clearest way to see it is to run the three main models against one hypothetical app with 10,000 monthly active users, then compare what each actually nets and — crucially — whether it compounds.

| Model | Setup | Gross/month | Net/month | Compounds? | | --- | --- | --- | --- | --- | | Subscription | 3% convert, 300 paying at $4.99/mo | $1,497 | ~$1,272 | Yes — retained users keep paying | | One-time IAP | 4% convert, 400 unlocks at $9.99 | $3,996 | ~$3,397 | No — only if acquisition continues | | Ad-supported | 400,000 impressions at $5–$10 eCPM | $2,000–$4,000 | ~$1,700–$3,400 | No — decays as users churn |

The IAP and ad columns look bigger month one, and that is exactly the trap. After year one, subscribers from month one who retain are still paying; IAP revenue from month-one buyers is gone, because they already paid and do not pay again; and ad revenue from month-one users only continues if they keep opening the app, which churn steadily erodes. Subscriptions turn each acquired user into a long-term annuity, which is why a subscription app with 2,000 paying users typically outearns an ad-supported free app with 200,000 downloads. The catch is that subscriptions only work if you have a retention case — apps users need occasionally, like an ID-photo tool or a unit converter, cannot sustain a subscription, and IAP is the honest model there. Know which category you are in; best iOS app niches for 2026 shows what works where.

Implementing subscriptions used to be hell. In 2026 it is manageable even for beginners because of two things. Apple's StoreKit 2 is the modern framework — it uses async/await, provides type-safe products, and handles the purchase flow natively. Writing subscription logic directly in StoreKit 2 is feasible for a small app but carries real complications: server-side receipt validation for fraud prevention, subscription status across devices, introductory offers, and lifecycle events like cancellation, billing retry, and refunds. RevenueCat is a third-party service that wraps StoreKit, handles server-side validation, manages subscriber state across devices via an anonymous user ID, and provides clean analytics. It is free up to $2,500/month in MRR, then takes a small percentage above that, and for 95 percent of indie developers it is the right answer.

The typical path is short: create your subscription products in App Store Connect with monthly and annual tiers, add the RevenueCat SDK, wire up the paywall with its Offerings API, gate premium features on RevenueCat's subscription status, and test with Sandbox accounts in TestFlight before shipping. What to avoid is building your own server-side receipt validation unless you have specific compliance needs — it is a maintenance headache, Apple changes receipt formats periodically, and getting it wrong leads to either revenue leakage or locked-out legitimate subscribers. For the full build path that this slots into, see how to build an iOS app with AI.

Pricing a subscription is not a guess; there is a reasonably consistent set of patterns that convert for indie utility apps. The $4.99/month, $29.99/year pair dominates for a reason — monthly at $4.99 feels cheap enough to try, and annual at $29.99 gives roughly 50 percent savings, which is the minimum annual discount that feels meaningful (below 40 percent, most users stick with monthly and hurt your LTV). Two tiers usually beat three for simple utilities; save three-tier pricing for apps with genuinely different feature sets. Framing annual as "just $0.58/week" converts better than "$29.99/year" alone. Seven-day free trials are the indie standard and improve conversion, while 14- or 30-day trials do not consistently help and just delay the decision. Introductory offers (say, $19.99 the first year, then $29.99) can boost initial conversion but hurt year-two retention when the price jumps, so use them carefully. And localized pricing matters — Apple's per-country tiers are usually reasonable, and adjusting for emerging markets can lift revenue noticeably.

The paywall is the single most important screen in a subscription app, and small changes there produce outsized revenue swings. Every effective paywall needs a clear benefit headline ("Track Unlimited Habits," not "Premium Access"), three to five specific benefits rather than feature names, two pricing options side by side with annual visually dominant, free-trial messaging with "cancel anytime" language, Apple's required disclosures just below the pricing, a restore-purchases link, and a clear dismiss option — hidden or blocked dismissal both hurts conversion and triggers rejection. The recurring mistakes are showing the paywall on first launch before users have seen value (show it after a meaningful first action, which converts two to four times better), listing too many features, confusing three-tier layouts, generic stock photos instead of real product UI, and auto-selecting monthly when defaulting to annual roughly doubles LTV per converter. Test one change at a time over two-week windows; RevenueCat's Experiments feature makes A/B paywall testing straightforward, and paywall optimization alone can move revenue 30 to 80 percent without a single new feature. For the wider revenue picture, see website monetization strategies.

Subscriptions look great until you see the churn. Every subscription app loses users every month: month-one churn runs 40 to 60 percent of new subscribers (brutal but normal), months two through six run 10 to 20 percent monthly, and month seven onward settles to 3 to 8 percent for users who made it past the hump. Lifetime value for an indie utility app at $4.99/month lands between $25 and $60 per paying user, and knowing your LTV is essential because it sets your maximum acquisition cost. You reduce churn by improving onboarding, sending genuine value-reminding notifications rather than spam, pushing annual plans (a subscriber on annual cannot churn mid-year), using Apple's in-app cancellation confirmation, and shipping updates every two to four weeks. Annual plans are your best friend: if 30 percent of subscribers go annual, your churn economics improve dramatically. Of users who start a 7-day trial, 40 to 60 percent typically convert for a well-built utility — below 30 percent signals weak onboarding, above 70 percent is suspicious.

Subscriptions are usually best, but not always. Use one-time IAP when your app does a one-off job (an ID-photo generator, a one-time PDF merger), when it sells expansion content like course packs or vocabulary packs, when it is a game with unlockables, when a small "remove ads" upgrade fits, or when your audience culturally rejects subscriptions. IAP revenue depends entirely on ongoing acquisition — new monthly users times conversion times average purchase value, with no compounding from retained users. Ads pay the least of any model: roughly $0.05–$0.20 per MAU for banners, $0.15–$0.50 for interstitials, and $0.30–$1.00 for rewarded video, versus a paying subscriber's $4.24 net every single month. Ads only work at high volume, with long session times, and where subscriptions would break the experience (social feeds, content consumption), and they are best as a supplement on a free tier rather than a primary model — paid-upfront, meanwhile, is essentially dead for consumer apps. For the full earnings picture, see how much indie iOS developers actually make.