Mobile App Maintenance Cost Per Year: What You Can Verify
What the app stores actually oblige you to pay each year, and what the percentage rule leaves out.
There is no published study behind the 15-20% rule, so build your mobile app maintenance cost per year out of the obligations you can check instead. Apple charges $99 per membership year. Google Play sets a target API level your app has to meet on a dated schedule or lose new users. Everything above those two is a price a firm quotes you, not a market rate, and the firms listed for this work declare very different floors.
Updated November 2026.
Where the mobile app maintenance cost per year rule of thumb came from
Four pages hold most of this query and all four print the same percentage. TekRevol puts annual upkeep at between 15% and 20% of your initial development budget. Cubix calls 15% to 20% a widely used benchmark in the industry. Space-O Technologies states the same 15% to 20% as a flat annual share. None of those three names a study, a survey or a dataset for it.
Cleveroad is the only one of the four that points the range at a document outside itself, linking a software-quality consortium guide. Its host returned a 403 when the file was requested on 1 November 2026, so a reader who tries to follow that citation to the number cannot get to it. Three unsourced statements and one link you cannot open is the whole provenance of the figure you were about to put in a budget.
The sharper problem is that the four pages do not agree with each other, and one of them does not agree with itself. TekRevol states 15% to 20%, then says year one can reach 30% to 50% and that costs settle at 15% to 25% in years two and three. Those are three different rules in one article, and a budget built on the first is wrong by a factor of three against the second.
| Page | Stated annual range | Stated percentage rule | Attribution |
|---|---|---|---|
| TekRevol | $5,000 to $120,000 or more | 15% to 20%, rising to 30% to 50% in year one | None for the percentage; cost tables credited to the firm’s own engagements |
| Cleveroad | $6,000 to $20,000 simple; $20,000 to $80,000 mid-size; $80,000 to $300,000 complex | 15% to 20% | A consortium guide, linked but not retrievable |
| Cubix | $5,000 to $15,000 simple; $15,000 to $50,000 mid; $50,000 to $150,000 complex | 15% to 20% | None |
| Space-O Technologies | Hosting $70 to $320 per month, built up factor by factor | 15% to 20% | None |
Is app maintenance really 20 percent of development cost?
The percentage is also indexed to the wrong thing. What you paid to build the app records how hard you negotiated and how much scope you bought two years ago; it is not a measure of how much work the app will demand next year. An app built cheaply by a team that left no tests behind costs more to keep alive than an app built expensively by a team that did, and the rule says the opposite.
The two store costs you can confirm in ten minutes
Apple states on its enrollment page that the Apple Developer Program is $99 per membership year, with prices varying by region and quoted in local currency during enrollment. Nonprofits, accredited educational institutions and government entities that meet its requirements can request a fee waiver. That is the only recurring platform fee on the iOS side, and it is the same number whether your app has a hundred users or a million.
Google charges differently. A Google Play developer account carries a $25 one-time registration fee with no annual renewal. So the floor under an app that exists on both stores is $99 a year, after $25 once at the beginning. Any maintenance quote that does not separate these two lines from engineering time is hiding the cheapest part of the bill inside the most expensive.
Google Play’s target API level is the obligation that forces work in a given year
The policy that actually puts a rebuild in your calendar is the target API level requirement, and it sets two different floors that are easy to confuse. Google Play’s target API level requirements page states that starting August 31, 2026, new apps and app updates must target Android 16 (API level 36) or higher to be submitted to Google Play. Separately, existing apps must target Android 15 (API level 35) or higher to remain available to new users on devices running Android OS higher than the app’s target API level.
Read those two sentences together and the practical rule is this. API 36 is the floor for putting anything new into the store, including an update to the app you already have. API 35 is the floor for staying reachable by new users on current phones. An app sitting at API 34 is not removed; the page says apps targeting API 34 or lower will only be available on devices running Android OS the same as or lower than the app’s target. The install base you lose is new users on newer hardware, which is the half of the install base that grows.
Popular coverage collapses the two floors into one and says you must target Android 16 or lose new users. That is wrong in a way that costs money: it tells a team at API 35 that it needs a rebuild it does not need. The page also says developers will be able to request an extension to November 1, 2026 for apps that need more time, with the extension forms reached from the Policy status page in Play Console. That window is a one-off tied to this cycle, not an annual grace period, and the date on it has now arrived.
The budget consequence is the part no percentage captures. A target API bump is not a feature; it is a forced migration of whatever the new platform version broke, plus a regression pass across the devices you support, plus a release. That work scales with how far behind you let the app drift, not with what you originally paid.
Apple’s version of the same rule is narrower, and the difference changes your plan
Apple applies its build requirement to uploads, not to apps already shipped. Its upcoming requirements page states that since April 28, 2026, apps uploaded to App Store Connect must be built with Xcode 26 or later using an SDK for iOS 26, iPadOS 26, tvOS 26, visionOS 26 or watchOS 26. The requirement governs what you send; the page does not say that an app already live must be rebuilt to stay live.
That asymmetry is worth planning around. Ship nothing for a year and your iOS app keeps serving the users it has while your Android app quietly stops being offered to new users on newer devices. Ship anything at all and you pay the toolchain cost of whichever SDK floor is current, even for a one-line fix: a corrected label carries a full rebuild, a regression pass and a review cycle behind it. That is why annual quotes look expensive against a short change list, and why bundling small fixes into one release a quarter costs less than shipping them as they arrive.
It also sets the real question for a quote. Ask a firm which SDK and which target API level your app is on today, what the gap is to the current floor on each platform, and what it would charge to close that gap once. If the firm cannot tell you your current target API level, it has not opened your project.
What the firms listed for this work charge to take it on
Supply shapes the price, and on this category the supply is deep. As of October 2026, Edverise lists about 48,400 firms under mobile app development for iOS and Android, spread across the world rather than concentrated in one market: the eight largest listed cities together hold 6,187 of those listings, or 12.8% of the category, so no single city is where the work sits. Ahmedabad is the largest at 1,501 listings, which is 3.1% of the category.
What matters more for a maintenance budget is the floor each firm declares. Of those listings, 26,945 state a minimum project size of $1,000 or less, which is 55.7% of the category, and 37,468 state a minimum under $10,000. Only 4,452 listings, 9.2%, declare a minimum of $25,000 or more. A maintenance retainer is one of the few engagements a small annual budget can genuinely buy, because most of the firms offering this work have not priced themselves out of it. These are listings Edverise compiled and firms that claimed them, not a vetted panel, and a declared minimum is what a firm says it will start at, not a quote.
Compare that with the category you want if your problem is diagnosis rather than upkeep. Edverise lists 491 firms under code review and technical audits, and the floors there sit higher: 228 of them, 46.4%, declare a minimum of $10,000 or more, against 22.5% in the app development category. Only 161, 32.8%, will start at $1,000 or less. An independent read of your codebase before you sign an annual retainer is a narrower market and a dearer one, which is worth knowing before you ask for it.
The in-house comparison sets the ceiling. The U.S. Bureau of Labor Statistics puts median pay for software developers, quality assurance analysts and testers at $134,040 per year, or $64.44 per hour, as of May 2025, with employment projected to grow 10 percent from 2025 to 2035. One developer at the median costs more than most of the annual quotes on the ranking pages above, and a maintenance retainer buys a fraction of a person rather than one. If a quote exceeds a loaded salary and the work is a target API bump and a backlog of small fixes, something in the scope is not what you think it is.
How to build the number you put in next year’s budget
Work from the obligations outward rather than from the build invoice down. Each step below produces a figure or a document you can put in front of a finance team.
- Write down the two store fees. $99 a year for the Apple Developer Program, and the $25 Play registration if you have not already paid it. These are certain, and they are the only lines in the whole budget that are.
- Find your current target API level and SDK. Both are in the project; anyone with the repository can read them in minutes. The gap between those two numbers and the current platform floors is the one piece of work you cannot defer indefinitely.
- Price that gap as a one-off, from two firms. Ask for the migration and the regression pass as a scoped piece of work, not as part of a retainer. A one-off has a deliverable and an end; a retainer has neither unless you write them in.
- Add the running costs you can read from an invoice. Hosting, the backend, crash reporting and any paid third-party SDK are already billing you today. Pull last year’s actuals rather than accepting an estimate; if the backend development was outsourced, the hosting account may not even be in your name.
- Decide the release cadence before you ask for a retainer price. Four releases a year and twelve releases a year are different products to maintain, and the cadence drives the number far more than the feature list does.
- Put the service level in the contract, not the email. Response time to a crash spike, who holds the signing keys, who holds the store accounts, and what happens to the code if you stop paying. Whether the retainer is fixed price or time and materials changes what an unplanned platform deadline costs you mid-year.
Where this does not apply: if your app has no backend, no paid SDKs and no release plans, your real annual cost is close to the $99 membership plus one platform migration when the target API floor moves past you. Buying a retainer for that is buying availability you will not use. Conversely, if the app takes payments or handles health data, the audit and dependency work is not discretionary and a percentage of your old build cost will understate it badly.
Frequently asked questions
How much does mobile app maintenance cost per year?
The only parts of a mobile app maintenance cost per year that can be verified independently are the store fees: $99 per membership year for the Apple Developer Program and a $25 one-time Google Play registration. Everything else is a quote. The published ranges span $5,000 to $300,000 a year across the four highest-ranking pages, which is wide enough to be useless as a budget input.
Is the 15-20% maintenance rule accurate?
It is unverifiable rather than inaccurate. Three of the four pages that state 15% to 20% cite nothing, and the fourth links a document that does not load. The rule also measures the wrong thing, because it indexes next year’s work to what you paid last time rather than to your code quality, your dependency count and your release cadence.
Do I have to update my app every year?
On Google Play, effectively yes if you want new users on newer devices. The policy requires existing apps to target Android 15 (API level 35) or higher to stay available to those users, and new submissions and updates to target Android 16 (API level 36) from August 31, 2026. On iOS the requirement applies to uploads, so an app you never update stays available to existing users.
What does a mobile app maintenance retainer actually include?
A retainer worth paying for names four things: a monthly hour allowance and what happens to unused hours, a response time for a crash spike or an outage, who holds the store accounts and signing keys, and whether platform migrations such as a target API level bump are inside the retainer or billed as a change order. Anything vaguer is an availability fee.
What happens if you do not update your app every year?
An Android app below the target API level floor is not deleted. Google Play’s policy says it will only be available on devices running Android OS the same as or lower than its target API level, so it keeps its existing users and stops being offered to new ones on current phones. The practical result is a download curve that flattens without any obvious cause.
How much are Apple and Google developer fees per year?
Apple charges $99 per membership year for the Apple Developer Program, with regional pricing and a fee waiver available to qualifying nonprofits, educational institutions and government entities. Google Play charges a $25 one-time registration fee with no annual renewal. Together that is $99 recurring each year on an app published to both stores.
Price the obligations, then shortlist against them
Take your current target API level, your SDK version and last year’s hosting actuals into the first conversation, and ask each firm to price the gap as a one-off before anyone discusses a monthly figure. The firms listed under mobile app development for iOS and Android can be filtered by the minimum project size they declare, which tells you in advance who will take a small annual engagement seriously. Edverise publishes how rankings work and what does not count toward them, and how to compare and vet a vendor sets out the checks to run before you sign.