Skip to main content

BusinessCodeCross Team

Year-one app upkeep: what to budget

A shipped binary is not finished. This is CodeCross’s own estimate, not a store rule: year-one keep-alive often runs about 15–25% of build cost.

Business

6 min

Keep-alive floor
15–25%

CodeCross estimate of original build, year one

Apple upload
Apr 28, 2026

Xcode 26 / iOS 26 SDK

Play update
Aug 31, 2026

API 36 on most devices

Year-one mobile app maintenance in 2026 is the keep-alive budget after launch: OS/store SDK gates, crash and ANR response, dependency rebuilds, and small fixes that keep the app installable and usable. As a planning floor, CodeCross’s own estimate is roughly 15–25% of original build cost — not as a feature backlog. Book a call when launch has no retainer line.

We price keep-alive as part of honesty, not an upsell surprise. If you vibe-coded to launch, the vibe coding harden pages still apply after store day one. For a build number, start at estimate.

What CodeCross’s 15–25% estimate covers

  • Forced store toolchain bumps (Xcode / SDK / target API).
  • Crash triage and hotfixes.
  • Plugin and payment SDK updates.
  • Privacy / Data safety form updates when SDKs change.
  • Light product fixes that keep the first-value path alive.

It does not mean a free second product.

Why 2026 does not let you sit still

Apple upload gates and Play API floors move on calendars you do not control. See Play target API requirements and Apple’s upcoming requirements (Xcode 26 since April 28, 2026). The API 36 essay walks the Android clock. Cross-platform plugins break on those bumps. Ignoring the floor is how “finished” apps become unlistable.

Stack choice does not waive the floor

Native, Flutter, and React Native all owe store clocks. Shared code can share some work — it does not delete OS churn.

What happens if you skip the retainer

  • Emergency rates replace planned hours.
  • The only person who understood signing has left.
  • A deadline hits during a marketing launch.

Budget before you ship

Put keep-alive beside the build estimate. Get a build estimate. Book a call.

Illustrative year-one keep-alive on a $50K build

Original build

$50K

Shown for scale — not part of keep-alive

$0$14K$29K$43K$57KIllustrative year-one keep-alive (not a quote)15% floor$7.5K20% typical$10K25% ceiling$12.5KOriginal build$50K

Illustrative year-one keep-alive (not a quote)

Illustrative studio band on a $50K original build. Hosting, UA, and store revenue share sit outside this chart.

CodeCross’s own estimate: year-one keep-alive often runs about 15–25% of original build cost. The $50K figure is illustrative, not a quote. That band is keep-alive. Feature work is a second line.

Year-one keep-alive stack

  1. Runtime health

    Crashes, ANRs, dependency CVEs

  2. Native graph

    Engine, plugins, 16 KB page-size rebuild

  3. Upload SDK gates

    Xcode 26 / iOS 26 SDK; Play API 36

  4. Identity / listing

    Certificates, Play verification, store metadata

The stores do not care that your feature freeze was in March. Each layer can block an upload on its own.

Apple punishes the next binary. Play also punishes sitting still

App Store

  1. Xcode 26 / iOS 26 SDK

    Upload gate live since Apr 28, 2026

  2. Blocks the replacement

    Connect rejects the package

  3. Idle listing still installs

    Until you try to replace it

so year-one spend is ↓

Google Play

  1. API 36 to upload an update

    Aug 31, 2026 — phone default

  2. API 35 to keep new-user reach

    Idle phone listings below 35 shrink

  3. "Not available on your device"

    New users on newer OS, not existing installs

Same calendar year, different failure modes. Budget both or you will staff only the store that yells first.

Where a $10K keep-alive year typically goes

Observability + store ops

$1.5K

Crash tools, certificates, listings

$0$1K$2K$3K$5KIllustrative allocation on a 20% / $50K yearStore / SDK mandates$4KCrash / hotfix cover$2.5KPlugin / NDK rebuilds$2KObservability + store ops$1.5K

Illustrative allocation on a 20% / $50K year

Illustrative split for a two-store listing. Move the mandate bar up if you are still on API 34 or Xcode 16.

Store mandates and the native graph take the first dollars. Crash cover is the insurance. Feature work is not in this chart.

Mandate, rebuild, upload, watch

  1. 01

    Read the live store clock

    Apple news, Play target API, 16 KB, verification

  2. 02

    Rebuild the graph that the clock owns

    Toolchain, engine, plugins, NDK objects

  3. 03

    Upload on a pin you can reproduce

    CI image, signing, review notes

  4. 04 · loops

    Watch crashes on the new OS

    Edge-to-edge, back gesture, iOS 26 chrome

This loop is the product after launch. Skipping a corner is how 36 ships and back breaks — or how Connect rejects the zip.

Three honest year-one outcomes

  1. 01 · Staff the floor

    15–25%

    CodeCross estimate: mandates, crash cover, plugin graph — features extra

  2. 02 · Sit and accept the cut

    Idle listing

    No hotfix path; Play may shrink new-user reach

  3. 03 · Unpublish

    Exit

    Cleaner than a zombie listing you cannot patch

Pick one before launch. "We'll see after we ship" is the fourth outcome, and the stores pick it for you.

Price keep-alive before you fund the build

  1. 01 →

    Name the original build number

    CodeCross’s 15–25% estimate attaches to this, not to a wish

  2. 02 →

    Inventory toolchains and targets

    Xcode pin, targetSdk, form factors

  3. 03 →

    List the native graph

    Engine, plugins, NDK, 16 KB status

  4. 04 →

    Mark the 2026 clocks that hit you

    Apple live; Play Aug 31; verification Sep 30

  5. 05 →

    Reserve the keep-alive band

    CodeCross estimate: 15% quiet / 20% typical / 25% dirty graph

  6. 06

    Put features on a second line

    Do not spend the floor on a redesign

If you cannot complete step 01, you are not ready to take a year-one bid.

FAQ

Is CodeCross’s 15–25% estimate a promise or a planning floor?

A CodeCross planning estimate — your reality depends on SDK debt and release rate. Book a call.

Does a Flutter app cost less to maintain than native?

Sometimes on shared UI fixes; not on store toolchain bumps or broken plugins. Do not assume half-price keep-alive.

Can we defer maintenance until we have revenue?

You can defer features. You cannot defer unlistable SDK gates without accepting install risk.

How does this relate to app build cost?

Keep-alive belongs next to the build number on every honest SOW. Start from a build cost estimate.

We launched from an AI builder — is year-one different?

You still owe crash gates, secrets hygiene, and store bumps — often more urgently. Start from vibe coding.

Book a call

Thirty minutes with a senior teammate — honest next steps.

Want a second pair of eyes on your app?

Tell us what you built and where it's stuck. We'll tell you what it takes to make it ready for real users.

Book a free call

Prefer writing? Send project details on the contact page.