In-app purchase vs subscription: the short answer
If you only have thirty seconds: choose an auto-renewable subscription when your app delivers ongoing value (fresh content, cloud sync, coaching, data, AI credits) and choose one-time in-app purchases when your app delivers a finished result or a consumable moment of fun. Everything else is nuance, and that nuance is exactly what decides whether your app makes $2,000 or $200,000 a month.
The trap most founders fall into is treating monetization as a post-launch decision. It is not. The billing model you pick changes your data model, your onboarding flow, your entitlement logic, your analytics stack and even your App Store screenshots. Retrofitting a subscription onto an app built around a single “Pro unlock” flag is one of the most expensive rewrites we see at VRS App.
This guide gives you a practical decision framework, real category examples (fitness, productivity, games, utilities), the current App Store commission tiers, and the churn maths behind each model.
First, get the vocabulary right: Apple has four purchase types, not two
“In-app purchase vs subscription” is a slightly misleading comparison, because on the App Store a subscription IS a type of in-app purchase. Apple’s StoreKit supports four product types, and the real decision is which of the four (or which combination) you ship. An extended version exists for anyone curious.
| Product type | What it is | Restores on new device? | Typical use case |
|---|---|---|---|
| Consumable | Used up and can be bought again | No | Coins, gems, extra lives, AI credits, boosts |
| Non-consumable | Bought once, owned forever | Yes (Restore Purchases required) | Pro unlock, remove ads, extra filters, a map pack |
| Auto-renewable subscription | Renews automatically until cancelled | Yes | Fitness plans, SaaS-style tools, streaming, cloud sync |
| Non-renewing subscription | Access for a fixed period, no auto-renew | You must handle it yourself | A season pass, a 6-month course, an event ticket |
Practical note: non-renewing subscriptions are rarely the right answer in 2026. They give you the churn of a subscription with none of the automatic renewal revenue, and you have to build your own expiry logic. Use them only when auto-renewal would feel predatory (kids’ education, one-off seasonal content, regulated markets).
Head-to-head: one-time in-app purchase vs subscription
| Criterion | One-time IAP (non-consumable) | Consumables | Auto-renewable subscription |
|---|---|---|---|
| Revenue predictability | Low. Revenue tracks installs, so it spikes and collapses with marketing | Medium. Depends on a small share of heavy spenders | High. MRR carries over month to month |
| Lifetime value per payer | Capped at the price you set | Potentially very high, very unequal | Highest on average, compounds with retention |
| Conversion rate | Usually higher, the commitment feels smaller | Low overall, tiny whale segment | Lower without a trial, much better with one |
| Churn exposure | None after purchase | None, but repurchase must be earned | Constant. Monthly plans commonly churn 8 to 15% per month |
| Ongoing cost coverage | Poor. Servers and AI tokens keep costing you money forever | Good if credits map to real usage cost | Excellent. Recurring cost, recurring revenue |
| Refund and support load | Low | Medium (accidental purchases, family sharing issues) | High (cancellations, “I forgot to cancel”, 1-star reviews) |
| Engineering complexity | Low | Medium (server-side balance, fraud) | High (renewals, grace periods, billing retry, server notifications) |
| Fundraise / valuation appeal | Weak | Depends on cohort economics | Strong. Investors price recurring revenue |
App Store commission tiers: what Apple actually keeps
The famous “Apple takes 30%” line is outdated for most independent developers. Here is the realistic picture for 2026.
| Situation | Apple’s commission |
|---|---|
| Standard rate (all IAP types) | 30% |
| App Store Small Business Program (under $1M in proceeds in the prior calendar year) | 15% |
| Auto-renewable subscription after 1 year of continuous paid service by the same subscriber | 15% |
| EU alternative business terms (DMA) and US external link-out purchases | Varies. Different fee structures apply and the rules keep evolving, so check Apple’s current terms before you model your P&L |
Two consequences that matter for the in-app purchase vs subscription choice:
- Subscriptions get cheaper over time. A retained subscriber moves you to 15% at month 13. A one-time purchase never earns that discount.
- Most startups pay 15%, not 30%. If you are under $1M in annual proceeds, enrol in the Small Business Program. It is a form, not a negotiation.
- Do not model 100% of revenue. Whatever model you pick, plan on keeping roughly 70 to 85% after commission, before taxes and payment-related deductions.
The 7-question decision framework
Answer these before a single StoreKit line is written. Score each answer A (one-time IAP), B (consumable) or C (subscription), then count. Anyone digging further should read In-App Purchase types.
1. Does value accrue over time or is it delivered once?
A PDF scanner that flattens a document has finished its job. A meal-plan app that ships new recipes every week has not. Ongoing value justifies ongoing payment, and users feel this intuitively. Charging monthly for a static feature set is the number one cause of angry reviews.
- Delivered once: A
- Accrues weekly or monthly: C
2. Do you carry recurring costs per user?
Cloud storage, sync servers, map tiles, LLM tokens, live moderation, licensed content: these bill you every month. A single $9.99 unlock against an AI feature that costs $0.40 per active user per month puts you underwater by month 25, or much sooner. Recurring cost demands recurring revenue, or metered consumables.
- Zero marginal cost, fully offline: A
- Cost scales with usage in bursts: B
- Steady per-user monthly cost: C
3. How often does a typical user open the app?
Frequency is the single best predictor of subscription survival. Rough guide from apps we have shipped and audited:
- Daily or near daily (habit trackers, language learning, fitness, journaling): subscriptions work
- Weekly (budgeting, meal planning, project tools): subscriptions work if you add value between sessions with notifications, widgets and reports
- Monthly or occasional (tax helper, unit converter, wedding planner, travel packing): one-time purchase, or a seasonal non-renewing pass
4. Who is the buyer, and are they paying with their own money?
B2B and prosumer users expense subscriptions without blinking. Consumers in casual categories resist them hard. A $4.99/month tool aimed at freelancers is an easy sell. The same price aimed at a teenager who wants a photo filter is not.
5. What do competitors in your category do?
User expectation is set by the category leader, not by your business plan. If the top 10 apps for your keyword all run a 7-day trial then $39.99/year, launching a $79.99 lifetime unlock is a differentiator. Launching a subscription in a category where everyone charges once is friction you will pay for in reviews. a practice worth looking at treats this as a baseline.
6. What is your acquisition channel?
If you plan to buy installs, you need LTV that beats CPI. Paid user acquisition in 2026 almost always requires subscription or consumable economics, because a single $6.99 unlock rarely survives a $4 to $9 install cost after commission. If you are relying on organic search, ASO and word of mouth, a one-time purchase can be very profitable.
7. Can you keep shipping?
Be honest. A subscription is a promise of continuous work. Solo developers with a day job frequently do better with a well-priced lifetime unlock than with a subscription they cannot feed. Users notice a stagnant subscription app within two billing cycles.
Category playbooks with real examples
Fitness and health
Winner: auto-renewable subscription, almost always. Workout apps, running coaches and nutrition trackers deliver fresh programming, streaks and progress data. The category has trained users to expect a free trial followed by an annual plan. Typical structure that performs well:
- Free onboarding quiz that personalises a plan (this is the real conversion engine)
- 7-day free trial on a weekly or annual plan
- Annual plan priced at roughly 4 to 6 times the monthly price, presented as the default
- A hard paywall after the first workout, not on launch screen one
Watch out for January seasonality: acquisition costs spike, and so does churn in March. Push annual plans hard in Q1.
Productivity and utilities
Winner: hybrid. This is the most contested category. Users have subscription fatigue for tools that feel like a single feature. Our recommended pattern:
- Subscription for anything with sync, collaboration, backup or AI: notes, task managers, document tools
- One-time non-consumable for self-contained utilities: a barcode generator, a specific calculator, a widget pack
- Both when you can: monthly, annual, plus a lifetime option at roughly 3x the annual price. The lifetime tier captures subscription refusers and generates cash spikes during launches and Black Friday
A common and effective compromise: a perpetual fallback licence, where cancelling leaves the user with the version they paid for, but no new features. It reduces cancellation anger and improves reviews.
Games
Winner: consumables, plus an optional subscription layer. Free-to-play economics rest on a small percentage of payers spending repeatedly. Structure:
- Consumables: soft and hard currency bundles, energy refills, continue-after-fail
- Non-consumable: remove ads, permanent character or cosmetic unlocks
- Auto-renewable subscription: a “club” or VIP tier giving daily currency, no ads and exclusive cosmetics. This is where premium mobile games have added predictability since battle passes went mainstream
Premium indie or narrative games remain a valid exception: a single paid unlock or paid-up-front app can still work when the experience has a defined end.
Content, media and education
Winner: subscription for libraries, one-time for finished courses. An endlessly updated library of lessons is a subscription. A single 12-hour certification course is a non-consumable or a non-renewing pass. Mixing them (subscription for the library, one-time for premium courses) is legitimate and very profitable.
Kids and family apps
Winner: one-time purchase or non-renewing subscription. Parents are the most subscription-averse buyers on the store, refund rates are higher, and Apple’s Kids Category rules are strict about how you present purchases. A clean “unlock everything” price is usually the right call.
The churn maths nobody shows you before launch
Subscriptions look better than one-time purchases on a spreadsheet only if you retain. Compare a $29.99 lifetime unlock against a $4.99/month subscription, both at 15% commission:
| Model | Net per payer, month 1 | Net per payer at 10% monthly churn | Net per payer at 5% monthly churn |
|---|---|---|---|
| $29.99 lifetime unlock | $25.49 | $25.49 (fixed) | $25.49 (fixed) |
| $4.99 monthly subscription | $4.24 | around $42 over the subscriber lifetime | around $85 over the subscriber lifetime |
Two lessons:
- The subscription only wins if users stay past month 6. Below that, a well-priced one-time unlock is competitive and far simpler to run.
- Cash flow timing differs enormously. One-time purchases pay you today, subscriptions pay you slowly. If you are self-funding paid acquisition, that gap can kill you even with better long-term unit economics. Annual plans exist precisely to fix this.
Hybrid models: usually the right answer
Framing this as in-app purchase vs subscription is a false binary. The highest-grossing apps combine types:
- Subscription + consumable credits: a base plan with included AI generations, then top-up packs. This is now the default for AI-powered apps because it aligns revenue with token cost.
- Subscription + lifetime tier: capture the segment that will never subscribe, and use lifetime as a limited-time offer to drive urgency.
- Free with ads + remove-ads non-consumable + premium subscription: three revenue floors for three willingness-to-pay segments.
- Consumables + season pass: standard in games, increasingly used in habit and social apps.
Rule of thumb: ship no more than three visible options on a paywall. More choice reliably lowers conversion. businessofapps.com has a solid rundown on this.
Why you must decide before development, not after
Here is what actually changes in the codebase depending on your answer:
- Entitlement architecture. A one-time unlock can live as a local flag. A subscription needs a server-side source of truth, App Store Server Notifications, grace periods, billing retry states and cross-device sync. Bolting this on later means touching every gated screen.
- Accounts. Subscriptions strongly imply user accounts so people can restore access on Android, web or a new device. One-time purchases can survive on Apple ID restore alone.
- Analytics and events. Trial start, trial conversion, renewal, cancellation, refund, reactivation, MRR cohorts. If you do not instrument these from day one you will be flying blind for your first three months of paid marketing.
- Onboarding design. Subscription apps live or die on the first 90 seconds. A personalised onboarding flow leading to a paywall is a product feature, not a screen you add at the end.
- Content pipeline. A subscription commits you to a release cadence. That means a CMS, a content calendar and staff, which is a company decision, not an engineering one.
- Migration pain. Moving existing lifetime buyers onto a subscription later is the single most review-damaging change you can make. Grandfathering rules have to be designed carefully, and they are far easier when the data model anticipated them.
Implementation checklist for 2026
- Use StoreKit 2 with server-side validation through the App Store Server API. Skip legacy receipt parsing on device.
- Handle App Store Server Notifications V2 for renewals, cancellations, refunds, billing issues and grace periods.
- Enrol in the Small Business Program if you qualify.
- Implement Restore Purchases visibly. Missing or hidden restore is a frequent App Review rejection.
- Disclose price, billing period, trial length and renewal terms clearly on the paywall itself. Apple and consumer regulators both enforce this.
- Offer win-back and promotional offers for lapsed subscribers, and use introductory offers for first-time buyers.
- Run price and paywall A/B tests in your first 60 days. Price is the highest-leverage variable in the entire app.
- Localise prices per storefront instead of relying on automatic conversion of a single US tier.
Five mistakes we see most often
- Subscribing a feature, not a service. If the user can imagine “finishing” with your app, monthly billing will feel unfair.
- Pricing the annual plan too close to the monthly. A weak discount kills the annual take rate, which is where cash flow and retention actually come from.
- Free trials with no reminder. Silent conversions generate refunds and one-star reviews. A friendly day-5 notification improves net retention.
- Only one price point. A single $9.99/month plan leaves both budget users and power users unserved.
- Deciding after the MVP is built. By then, the paywall is a bolt-on and every metric you need is missing.
A simple scoring shortcut
| Your profile | Recommended model |
|---|---|
| Daily use, server costs, content updates, paid acquisition | Auto-renewable subscription with annual default and free trial |
| Offline utility, occasional use, organic traffic, solo developer | One-time non-consumable unlock, optional paid major upgrades |
| Game or entertainment with a small high-spend segment | Consumables plus a VIP subscription tier |
| Usage costs you money per action (AI, rendering, API calls) | Subscription with included credits, plus consumable top-ups |
| Kids, education, seasonal or event-based content | One-time purchase or non-renewing pass |
FAQ
Is an app free if it says “in-app purchases”?
Yes, the download itself is free. “Offers In-App Purchases” on the App Store listing means you can install and open the app without paying, but some content, features or subscriptions inside cost money. Many apps are usable for free with limits, and some are effectively unusable until you buy.
Do I get charged for in-app purchases automatically?
Only when you confirm a purchase, which requires Face ID, Touch ID or your password. The exception is an auto-renewable subscription: once you subscribe, it renews and charges you automatically until you cancel, including at the end of a free trial. You can cancel any time from Settings, Apple Account, Subscriptions.
Does Apple take 30% of in-app purchases?
Not for most developers. The standard rate is 30%, but developers earning under $1 million per year can enrol in the App Store Small Business Program and pay 15%. Auto-renewable subscriptions also drop to 15% after a subscriber has been paying for more than one year. Alternative distribution and external purchase options in the EU and US carry different fee structures.
How am I charged for in-app purchases?
Through your Apple Account payment method: card, Apple Pay balance, carrier billing or App Store credit. Charges appear on a receipt emailed by Apple, and purchase history is visible in Settings. Family Sharing can extend eligible purchases and subscriptions to family members.
Is a subscription always more profitable than a one-time purchase?
No. A subscription is more profitable only if your subscribers stay several months. With high churn, poor retention or a feature that feels “finished”, a well-priced one-time unlock can net more per user and cost far less to operate.
Can I offer both an in-app purchase and a subscription in the same app?
Yes, and it is often the best strategy. Common combinations are a subscription plus a lifetime unlock, or a subscription plus consumable credit packs. Just keep the paywall simple and make it obvious which option is recommended.
Can I switch from one-time purchases to a subscription later?
Technically yes, commercially it is risky. Existing buyers expect to keep what they paid for, so you need grandfathering rules, clear communication and ideally new value behind the new paywall. It is much cheaper to design the model before development.
Need help choosing and building it?
At VRS App we design monetization models before writing code: paywall strategy, StoreKit 2 implementation, server-side entitlements, trial and offer logic, and the analytics that tell you whether it is working. If you are weighing in-app purchase vs subscription for a new iOS app, or your current model is underperforming, get in touch with our team and we will map the right model to your category and cost structure.