Mobile App Marketing Statistics 2026: Install and Cost Data

The App Market in 2025 and 2026

Sensor Tower estimates that people spent $167 billion on in-app purchases and paid apps across the App Store and Google Play in 2025, up 10.6%, while downloads barely moved at nearly 150 billion: Sensor Tower describes growth shifting from new users to monetization. Non-game apps earned more from in-app purchases than games for the first time, with generative AI apps among the fastest growers; their in-app revenue nearly tripled to more than $5 billion.

The pattern continued into 2026. In the second quarter, in-app revenue rose 5.3% to $43.6 billion on flat downloads, with non-game revenue up 14.6% and game revenue down 4.5%. Apple’s own figure is wider: a study it published puts the App Store ecosystem at $1.4 trillion of billings and sales in 2025, most of it physical goods and services, and developers paid Apple no commission on more than 90% of the total. For the device side of the market, see our iOS vs Android market share report.

How Much Is Spent on App Marketing

The main public figure for app install spending comes from AppsFlyer, a mobile measurement company that sees its clients’ campaigns. It put global app user-acquisition spend at $78 billion in 2025, up 13%, with all of the growth on iOS: iOS spend rose 35% while Android was flat. Advertisers also spent $31.3 billion re-engaging existing users, 29% of all app marketing investment.

AppsFlyer bases these totals on data from apps that use AppsFlyer and presents them as global figures, but it does not publish how it scales its sample up to the whole market, and it released the 2025 figure on December 10, 2025, before the year had ended. The sample also changes between editions (35,000 apps for 2024, 45,000 for 2025): the 13% growth it reports for 2025 implies about $69 billion for 2024, not the $65 billion it reported a year earlier, so the two are not a like-for-like series.

Cost per Install by Vertical and Region

Cost per install depends far more on the vertical and the country than on any global average. Adjust’s 2026 report, built from the apps it measures, puts the 2025 global CPI at $0.56 for games, $0.98 for e-commerce apps and $1.13 for finance apps. Installs cost far more in North America and Europe: $1.68 for games in North America (against $0.53 in Europe), $2.49 and $2.25 for e-commerce, and $4.13 and $4.75 for finance, where Europe is the dearer of the two.

Prices did not move the same way everywhere. Game CPIs rose 30% globally, while finance CPIs fell, by more than a third in North America and Europe. Adjust does not split CPI by operating system in these reports, and Apple Ads charges per tap rather than per install, so a cross-category iOS or Android CPI is not something any of these publishers report.

Retention Benchmarks

Retention is the share of users who come back a given number of days after installing. In Adjust’s 2025 data, 27% of game installers returned on day 1 and 5% on day 30. E-commerce and finance apps keep fewer users early (about 12% on day 1) and lose most of them by day 30 (2% to 3%), although marketplace and classifieds apps kept 8% on day 30.

Averages hide a wide spread. GameAnalytics ranks more than 16,000 games: the median game kept about 22% of players on day 1 at the start of 2025, a rate that fell during the year, and under 1% on day 30, while the top 1% kept 64% to 68% on day 1. Compare your own cohorts with the right percentile, not with a single average, and measure retention the same way each time.

Monetization: Revenue per User and Payers

Revenue per install sets the ceiling on what a campaign can pay per install. AppsFlyer’s 2026 monetization report gives 90-day in-app purchase revenue per install of $2.43 for casino games, $2.13 for midcore and $1.34 for casual. Per paying user the gap narrows: casino payers spent $11.40 and non-gaming subscribers $10.85.

Revenue streams also mature at different speeds. In-app advertising earns 89% of its 60-day revenue within a week of install, so it shows up early in return-on-ad-spend reports, while subscription revenue builds more slowly. Subscription revenue processed through the stores grew 105% year on year in the first quarter of 2026, the fastest of the three streams.

ATT Opt-In Rates

Since iOS 14.5, apps must ask permission through Apple’s App Tracking Transparency prompt before tracking users across other companies’ apps and websites. Opt-in rates have risen, not stalled: Adjust measured 38% of users shown the prompt agreeing in the first quarter of 2026, up from 35% a year earlier, with games at 39%. AppsFlyer reported 50% in April 2025.

Both firms count only users who were shown the prompt, but they measure different sets of apps, so the two figures are not directly comparable. Apple also prohibits identifying devices from their characteristics, so apps may not match users who decline by “fingerprinting” their devices. From iOS 27.2, EU developers will be able to use an alternative ATT prompt agreed with European competition authorities, and in five EU countries it will be the only prompt available.

Attribution: SKAdNetwork and AdAttributionKit

For users who do not opt in, iOS install campaigns are measured through Apple’s aggregated attribution frameworks. SKAdNetwork 4, from iOS 16.1, sends up to three postbacks covering days 0 to 2, 3 to 7 and 8 to 35 after the first launch; it is the latest version in Apple’s release notes. AdAttributionKit, introduced in iOS 17.4, extends the same model to alternative app marketplaces and re-engagement, and lets apps set their own attribution windows within Apple’s limits.

Postbacks arrive at least 24 to 48 hours after a conversion, so iOS campaign reporting always lags. If you need help setting up measurement across these frameworks and your mobile measurement partner, see our analytics services.

Install Fraud

Fraud distorts both paid and organic numbers. AppsFlyer’s 2026 fraud report, covering 106.4 billion installs, found organic traffic was the largest single source of fraudulent installs at 52%. AppsFlyer’s explanation is that organic installs are the baseline marketers judge paid campaigns against, so inflating them makes fraudulent paid installs look normal. Affiliates carried nearly 40%. The iOS fraud rate fell from 17.5% to 11.7%, while fraud on DSP and owned-media traffic rose.

For channel context, see our affiliate marketing statistics and programmatic advertising statistics.

App Store Optimization and Apple Ads

Search drives App Store discovery: Apple Ads says almost 65% of downloads happen directly after a search, using 2022 data. That makes keywords, titles and product pages a lever on every paid campaign as well as on organic installs. Apple reports that sending people to a custom product page lifts conversion by 2.5 points on average against a 1.6% baseline for default pages; treat that as Apple’s own claim, not an independent benchmark.

Both stores now let you test and target listings: up to 70 custom product pages and three-treatment tests on the App Store, and up to 50 custom store listings on Google Play. Apple Search Ads was renamed Apple Ads in April 2025, and from 2026 Apple runs additional ads further down search results, so paid and organic positions now share more of the results page. For listing-level data, see our app store optimization statistics.

App Store Fees and Platform Rules

Store fees and rules changed repeatedly in 2025 and 2026, several times after action by regulators. Google cut its standard Google Play service fee on new installs to 20% in the EEA, UK and US from June 30, 2026, and in Australia and Japan from September 30, 2026, with 10% on subscriptions and on a developer’s first $1 million of annual earnings, plus a separate 5% fee when Google Play Billing is used; the rest of the world follows by September 30, 2027. Apple lowered its China rates and moved EU developers to a single set of terms, and opened iOS to alternative marketplaces and payments in Japan and Brazil.

In the UK, the Competition and Markets Authority designated Apple and Google with strategic market status in October 2025 and secured commitments on app review and ranking in 2026. The fee you pay now depends on the country, the install date and the billing method, so model margins market by market.

What Ad Networks and App Stores Report

AppLovin and Unity do not publish average costs per install, but their filings show where spending is going. AppLovin, which sold its apps business in 2025 to concentrate on advertising, reported second-quarter 2026 revenue of $1.92 billion, up 53%. Unity’s Grow Solutions revenue was $389 million in the same quarter, and Unity closed the ironSource Ads network in April 2026.

The app stores report on supply and safety. Apple reviewed more than 9.1 million app submissions in 2025 and rejected over 2 million; Google says it blocked 1.75 million policy-violating apps from Google Play.

How to Use These Numbers

Match the benchmark to your app before comparing. Adjust and AppsFlyer report on the apps that use their measurement tools, GameAnalytics on games integrated with its SDK, and Sensor Tower estimates the whole market. Their figures differ in sample, region and definition, so a CPI from one and a retention rate from another do not describe the same apps.

Use the vertical and region closest to yours, check whether a figure is a median, an average or a top percentile, and judge campaigns on retained, paying users rather than installs. For web conversion benchmarks that sit alongside app funnels, see our conversion rate benchmarks, and for search campaigns that drive app installs, our Google Ads benchmarks.

Widely Quoted Figures, Checked

Several mobile app marketing statistics circulate without a source, or with a source that does not publish them. Each one below is checked against the publisher’s own page.

Sources

Every figure on this page comes from one of the sources below.