Apple rejected 23% of App Store submissions in 2025 (Apple), Google blocked 1.75 million policy-violating apps in 2025 (Google), digital wallets took 56% of global online spending in 2025 (Worldpay), and 95.9% of the top one million home pages had detectable WCAG 2 failures (WebAIM). Each entry below names its publisher and the period the data describes. For whole-industry figures, see the companion software testing statistics page.
App Review looked at 9,100,620 submissions in 2025 and rejected 2,093,244 of them, a 23.0% rejection rate, according to Apple Inc.'s App Store Transparency Report 2025. That is down from 24.85% in 2024, a comparison drawn from Apple's separate 2024 report. Because Apple is reporting on its own review process, the figure is a measured count rather than an estimate.
Source: Apple Inc., App Store Transparency Report 2025. Data describes 2025. Flag: primary.
Apple's 2025 App Store Transparency Report counts 1,354,418 Performance rejection citations in 2025, more than Legal (495,673), Design (415,532), Business (283,820) and Safety (151,159) combined. Performance is a broader category than crashes. Apple's Section 2 covers App Completeness, Beta Testing, Accurate Metadata, Hardware Compatibility and Software Requirements, so a misleading screenshot is a 2.3 rejection and counts as Performance. Apple's App Review page adds the crash-linked figure:
"On average, over 40% of unresolved issues are related to guideline 2.1: App Completeness, which covers crashes, placeholder content, incomplete information, and more."
Source: Apple Inc., App Store Transparency Report 2025 and App Review page. Data describes 2025 for the report. Flag: primary.
Apple logs 387,087 App Store submissions approved in 2025 after an initial rejection, up from 295,109 in 2024. Most rejections are fixable, so what a rejection usually costs a team is the extra round trip and the slip in the release date. Apple publishes the two counts separately and does not state the relationship between them.
Source: Apple Inc., App Store Transparency Report 2025. Data describes 2025. Flag: primary.
Google prevented 1.75 million policy-violating apps from publishing in 2025 and banned more than 80,000 developer accounts, according to its Google Play and Android ecosystem safety post, written by Vijaya Kaza. The same post says Google ran more than 10,000 safety checks on every published app and that Play Protect scans over 350 billion Android apps daily. Google does not publish a submission rejection rate.
Source: Google, Google Play and Android ecosystem safety post. Data describes 2025. Flag: primary.
Apple's App Review page states: "On average, 90% of submissions are reviewed in less than 24 hours." Apple is describing its own process here, so the figure is self-reported by the platform owner.
Source: Apple Inc., App Review page. Data year not stated. Flag: primary.
Android 16.0 accounted for 25.76% of worldwide Android pageviews in the August 2026 reading from StatCounter Global Stats, with Android 15.0 at 16.91%, 13.0 at 14.75%, 14.0 at 13.09% and 12.0 at 10.07% behind it. On the Apple side, the App Store support page, measured 7 June 2026, puts iOS 26 on 79% of all iPhones and 86% of iPhones introduced in the last four years. The two figures measure different things, since StatCounter counts panel pageviews while Apple counts devices that transacted on the App Store, so read them as two measurements rather than one comparison.
Source: StatCounter Global Stats, Android Version Market Share Worldwide. Data describes August 2026. Flag: primary, panel measurement.
The same Android version held 39.05% of U.S. Android pageviews against 25.76% worldwide, a gap of 13.29 percentage points, in the August 2026 reading from StatCounter Global Stats. In the U.S., Android 13.0 held 13.1%, 15.0 held 10.24%, 14.0 held 9.72% and 12.0 held 7.85%. A device matrix built on a home-market profile does not describe the markets a team is expanding into, which is a reason to test in the markets you are expanding into. The same panel caveat applies here: these are pageviews rather than installed devices.
Source: StatCounter Global Stats, Android Version Market Share United States. Data describes August 2026. Flag: primary, panel measurement.
Apple's App Store support page, measured 7 June 2026, reports that "68% of all devices use iPadOS 26," while 79% of all devices introduced in the last four years use iPadOS 26. The same page puts iOS 26 on 79% of all iPhones, so an iPad test matrix needs deeper older-OS coverage than an iPhone one.
Source: Apple Inc., App Store support page. Measured 7 June 2026; data describes devices that transacted on the App Store, as measured 7 June 2026. Flag: primary.
58.99% of worldwide web pageviews came from mobile in StatCounter's September 2026 reading, against 39.26% for desktop and 1.76% for tablet. With mobile the majority platform for web use worldwide, a desktop-first test plan starts with the minority platform.
Source: StatCounter Global Stats, Desktop vs Mobile vs Tablet Market Share Worldwide. Data describes September 2026. Flag: primary, panel measurement.
The International Telecommunication Union estimates that "5G currently reaches 84 per cent of people in high-income countries, but only 4 per cent of people in low-income countries." An app tested only on fast home-market networks has not been tested on the connections most users in lower-income markets will actually have.
Source: International Telecommunication Union, Facts and Figures 2025, Mobile network coverage. Data describes 2025. Flag: primary (UN agency; ITU estimates).
A single crash truncates the session and reduces content consumption, while page views rise in the following session, according to "The Impact of App Crashes on Consumer Engagement", published in the Journal of Marketing, Vol. 89, No. 4, pp. 81 to 98, DOI 10.1177/00222429241304322. The authors are careful to add that "this increase in page views does not compensate for the loss during the crashed session." Across three experiments plus field data, they identify need for closure and curiosity as the mediators, and they find that post-crash messages reduce the damage.
Source: SAGE Publishing, "The Impact of App Crashes on Consumer Engagement", Journal of Marketing (DOI 10.1177/00222429241304322). Study period not disclosed. Flag: primary, peer reviewed.
The same Journal of Marketing 89(4) study finds that frequent and concentrated crashes curtail engagement, so the business impact of instability depends on how often crashes happen and how closely they cluster, which makes it nonlinear rather than a simple function of crash count. The ad-revenue damage is reported in the Santa Clara University write-up and the PRNewswire release, rather than in the journal abstract.
Source: Santa Clara University, Crashing engagement: Savannah Wei Shi's research and PRNewswire, Study examines effect of app crashes on consumer engagement. Data describes 2025, study period not disclosed. Flag: secondary summaries.
Luciq's 2026 report states that "15.4% of users uninstall an app after a single crash," based on a survey of more than 1,000 U.S. mobile app users. The figure is stated behavior rather than measured uninstalls, so it describes what users say they do.
Source: Luciq, "No Margin for Error: What Mobile Users Expect and What Mobile Leaders Must Deliver in 2026" press release, Luciq Report: In 2026, Mobile Users Expect Perfection (Business Wire). Data describes more than 1,000 U.S. mobile app users, fieldwork dates not stated. Flag: vendor funded.
17% of U.S. online shoppers who abandoned a checkout said the website had errors or crashed, according to Baymard's latest quantitative abandonment study. In shoppers' own account of why they left, that defect reason sits alongside cost and delivery reasons, and the study covers web and app checkout broadly rather than native apps only.
Source: Baymard Institute, Cart Abandonment Rate Statistics. Data describes Baymard's latest quantitative study of reasons for abandonment, survey year and sample size not stated. Flag: primary.
Mobile Core Web Vitals pass rates improved from 36% in 2023 to 44% in 2024 and reached 48% in 2025, while desktop moved from 48% in 2023 to 55% in 2024 and 56% in 2025, according to the Performance chapter of HTTP Archive's Web Almanac 2025. The figures come from real Chrome users through Chrome UX Report data, so they reflect field performance rather than a lab score, and they cover the mobile web rather than native apps.
Source: HTTP Archive, Web Almanac 2025, Performance chapter. Data describes Chrome UX Report data from July 2025. Flag: primary.
Google Play flags an app when at least 1.09% of its daily users experience a user-perceived crash, or 8% on a single device model, according to Google Play Console Help. The ANR thresholds are 0.47% overall and 8% per device model, and Google defines a user-perceived crash rate as "the percentage of your daily users who experienced at least one user-perceived crash." The 1.09% is a policy threshold, the point at which Google acts, rather than a measured crash rate across apps.
Source: Google, Google Play Console Help. Data describes the current policy threshold, not a measured year. Flag: primary.
Digital wallets accounted for "56% of online and 33% of in-person spending in 2025," according to Worldpay's Global Payments Report 2026, which draws on consumer research across 63,000+ consumers in 42 markets. Cards are no longer the default online payment method worldwide, so checkout testing that covers cards alone misses most online spend.
Source: Worldpay (now part of Global Payments), Global Payments Report 2026, GPR Trend 3: Digital wallets are winning on multiple fronts. Data describes 2025 across 63,000+ consumers in 42 markets. Flag: primary, vendor funded.
Adobe's holiday 2025 data, covering 1 November to 31 December 2025 across U.S. retail sites, reports that "the majority of online transactions (56.4%) took place through a smartphone this season (up from 54.5% in 2024)." During the highest-traffic weeks of the U.S. retail year, most online orders went through a phone, which makes the mobile checkout the primary checkout.
Source: Adobe, Adobe Digital Insights press release: "Adobe: Holiday Shopping Season Drove a Record $257.8 Billion Online with Consumers Embracing Generative AI Tools" (PDF). Data describes 1 November to 31 December 2025, U.S. retail sites. Flag: primary (Adobe Analytics transaction data).
Baymard's 2025 Checkout UX benchmark, built on 41,000+ checkout performance scores, finds that "63% of Mobile sites have a 'mediocre' or worse Checkout UX performance" and "only 2% rating 'good'." The benchmark scores checkout design against Baymard's guidelines rather than counting defects, and it applies to mobile web rather than native apps.
Source: Baymard Institute, Checkout UX Best Practices 2025. Data describes Baymard's 2025 Checkout UX benchmark of 41,000+ checkout performance scores, review dates not stated. Flag: primary (expert-review benchmark).
Luciq's 2026 survey states that "53.2% abandon purchases due to crashes or slowdowns during major sales." The result is self-reported and carries the same sample and access caveats as the single-crash figure above.
Source: Luciq, "No Margin for Error: What Mobile Users Expect and What Mobile Leaders Must Deliver in 2026" press release, Luciq Report: In 2026, Mobile Users Expect Perfection (Business Wire). Data describes more than 1,000 U.S. mobile app users, fieldwork dates not stated. Flag: vendor funded.
Apple's App Store overview page states: "With availability in 175 regions and 50 languages, people around the world can discover your apps and games on the App Store." Every storefront and language an app ships to is a separate set of strings, formats and listings that can break.
Source: Apple Inc., Apple Developer, App Store overview page. Data describes the current state. Flag: primary (self-reported by the platform owner).
Total downloads across iOS and Google Play "edged up 0.8% YoY to nearly 150 billion" in 2025, excluding third-party Android stores, according to Sensor Tower's State of Mobile 2026. Flat download growth means teams are competing for existing users rather than new ones, which raises the cost of a bad first session. The same page puts in-app purchase and paid app revenue at $167 billion in 2025 across iOS and Google Play, a 10.6% year-on-year increase.
Source: Sensor Tower, State of Mobile 2026. Data describes calendar year 2025, iOS and Google Play only. Flag: primary (Sensor Tower estimates).
95.9% of home pages had detected WCAG 2 failures in WebAIM's 2026 accessibility scan of the top 1,000,000 home pages. The scan is automated and picks up only machine-detectable failures, so the true share is not lower than this. It covers web pages rather than native apps.
Source: WebAIM, The WebAIM Million: The 2026 report on the accessibility of the top 1,000,000 home pages. Data describes an automated scan of 1,000,000 home pages in February 2026. Flag: primary (automated testing only).
The global software testing market is worth USD 54.44 billion in 2026 and is projected to reach USD 99.94 billion by 2031, a 12.92% CAGR, according to Mordor Intelligence. The 2026 base year is measured, while the 2031 figure is a projection from a commercial research product rather than a survey.
Source: Mordor Intelligence, Software Testing Market. Data describes base year 2026, forecast to 2031, study period 2020-2031. Flag: projection.
Mordor Intelligence and Research Nester size the 2026 software testing market at USD 54.44 billion and USD 57.73 billion, $3.29 billion apart on the same base year. Their growth rates sit nearly twice as far apart: one has the market roughly doubling by 2031, while the other takes a decade to get there. Research Nester also projects North America at 48.1% of global share by 2036. The two sets of figures are listed side by side below rather than averaged or ranked.
| Firm | Base year | Base value | Forecast year | Forecast value | CAGR |
|---|---|---|---|---|---|
| Mordor Intelligence | 2026 | USD 54.44 billion | 2031 | USD 99.94 billion | 12.92% CAGR, 2026 to 2031 |
| Research Nester | 2026 | USD 57.73 billion | 2036 | USD 108.37 billion | 6.5% CAGR, 2027 to 2036 |
Source: Mordor Intelligence, Software Testing Market and Research Nester, Software Testing Market. Data describes 2026 base years with forecasts to 2031 and 2036. Flag: projections.
The crowdsourced testing market stood at $3.5 billion in 2025 and is projected to reach $3.9 billion in 2026 and $9.4 billion by 2033, a 13.3% CAGR, according to the 2026-2033 edition of Grand View Research's report. The 2025 base year is measured; the rest is a projection.
Source: Grand View Research, Crowdsourced Testing Market Report, 2026-2033 edition. Data describes base year 2025, forecast to 2033. Flag: projection.
Respondents to TestRail's 4th Edition Software Testing and Quality Report, from a 2025 survey wave, "anticipate that 63% of their tests will be automated" in the next year. Regression is the most automated category, at 56% of teams, and smoke tests sit at 40%. TestRail sells test management software, so the figure is vendor funded. An independent analysis by Zhimin Zhan argues that the effective level of automation in practice is much flatter than the survey suggests:
"If a team can't consistently achieve a green run (100% pass) every one or two days, the effective level of test automation is 0%, because it simply can't be trusted."
Source: TestRail, 4th Edition Software Testing and Quality Report (PDF). Data describes 2025 survey wave. Flag: vendor funded.
Katalon's State of Software Quality Report 2025 states that "despite AI's automation benefits, 56% of QA teams still struggle to keep up with testing demands." The same page reports that 76% of respondents use AI-powered tools in their software testing activities, which suggests that the constraint most teams report is capacity rather than tooling.
Source: Katalon, State of Software Quality Report 2025. Data describes 2025 survey of 1,500+ QA professionals, fieldwork dates not stated. Flag: vendor funded.
76.8% of testing professionals say their organization currently uses AI, the global average reported in PractiTest's State of Testing Report 2026, 13th edition. The same page reports AI used for test creation by 69.6% of respondents and for risk identification by only 19.9%, so for most teams AI is producing tests rather than deciding what to test. The respondent count is not published on the landing page.
Source: PractiTest with Tea-time with Testers, State of Testing Report 2026, 13th edition. Data describes 2026 survey wave, respondent count not published. Flag: vendor funded.
89% of responding organizations are piloting or deploying Gen AI-augmented workflows, while only 15% have achieved enterprise-wide implementation, according to Capgemini's World Quality Report 2025-26 press release. The 2025 survey covered over 2,000 senior executives across 22 countries and 10 sectors. Capgemini names OpenText as strategic technology partner, which makes the report vendor funded through that partner.
Source: Capgemini with Sogeti and OpenText, World Quality Report 2025-26 press release. Data describes 2025 survey of over 2,000 senior executives across 22 countries and 10 sectors. Flag: vendor funded via technology partner.
"AI solutions that are almost right, but not quite" is the biggest frustration with AI tools for 66% of developers, according to Stack Overflow's 2025 Developer Survey, AI section, and 45.2% say debugging AI-generated code is more time-consuming. Code that is nearly right passes a glance and fails in use, which is the gap testing exists to close.
Source: Stack Overflow, 2025 Developer Survey, AI section. Data describes survey fielded 29 May to 23 June 2025, 49,009 responses from 177 countries. Flag: primary.
| Source title | Publishing body | URL | Data describes | Flag |
|---|---|---|---|---|
| App Store Transparency Report 2025 | Apple Inc. | link | 2025 | Primary |
| App Store Transparency Report 2024 | Apple Inc. | link | 2024 | Primary |
| App Review Guidelines | Apple Inc. | link | Not stated by Apple | Primary |
| App Review page | Apple Inc. | link | Not stated by Apple | Primary |
| App Store support page | Apple Inc. | link | 2026 | Primary |
| App Store overview page | Apple Inc. | link | Current state | Primary |
| Google Play and Android ecosystem safety post | Google, by Vijaya Kaza | link | 2025 | Primary |
| Google Play Console Help | link | Current policy threshold, not a measured year | Primary | |
| Android version market share, worldwide | StatCounter Global Stats | link | August 2026 | Primary, panel measurement |
| Android version market share, United States | StatCounter Global Stats | link | August 2026 | Primary, panel measurement |
| Desktop vs Mobile vs Tablet Market Share Worldwide | StatCounter Global Stats | link | September 2026 | Primary, panel measurement |
| "The Impact of App Crashes on Consumer Engagement", Journal of Marketing, Vol. 89, No. 4, pp. 81-98 | SAGE Publishing | link | Study period not disclosed | Primary, peer reviewed |
| Study examines effect of app crashes on consumer engagement | PRNewswire | link | 2025 | Secondary summary |
| Crashing engagement: Savannah Wei Shi's research | Santa Clara University | link | Not stated | Secondary summary |
| Cart Abandonment Rate Statistics | Baymard Institute | link | Latest quantitative abandonment study; survey year and sample size not stated | Primary |
| Web Almanac 2025, Performance chapter | HTTP Archive | link | CrUX data, July 2025 | Primary |
| Facts and Figures 2025, Mobile network coverage | International Telecommunication Union | link | 2025 | Primary (UN agency; ITU estimates) |
| Global Payments Report 2026, GPR Trend 3 | Worldpay, now part of Global Payments | link | 2025 | Primary, vendor funded |
| Adobe holiday 2025 press release | Adobe Digital Insights | link | 1 November to 31 December 2025, U.S. retail sites | Primary |
| Checkout UX Best Practices 2025 | Baymard Institute | link | 2025 Checkout UX benchmark | Primary |
| The WebAIM Million 2026 | WebAIM | link | Automated scan, February 2026 | Primary |
| State of Mobile 2026 | Sensor Tower | link | Calendar year 2025 | Primary (estimates) |
| 2025 Developer Survey, AI section | Stack Overflow | link | Survey fielded 29 May to 23 June 2025 | Primary |
| Source title | Publishing body | URL | Data describes | Flag |
|---|---|---|---|---|
| "No Margin for Error" 2026 report press release | Luciq | Business Wire | More than 1,000 U.S. mobile app users; fieldwork dates not stated | Vendor funded |
| State of Software Quality Report 2025 | Katalon | link | 2025 survey of 1,500+ QA professionals | Vendor funded |
| State of Testing Report 2026, 13th edition | PractiTest with Tea-time with Testers | link | 2026 survey wave; respondent count not published | Vendor funded |
| World Quality Report 2025-26 press release | Capgemini with Sogeti and OpenText | link | 2025 survey of over 2,000 executives across 22 countries | Vendor funded via technology partner |
| 4th Edition Software Testing and Quality Report | TestRail | link | 2025 survey wave | Vendor funded |
| Software testing market | Mordor Intelligence | link | Base year 2026, forecast to 2031 | Projection |
| Software testing market | Research Nester | link | Base year 2026, forecast to 2036 | Projection |
| Crowdsourced testing market report | Grand View Research | link | Base year 2025, forecast to 2033 | Projection |
| Source title | Publishing body | URL | Data describes | Flag |
|---|---|---|---|---|
| Analyse of the 4th Edition Software Testing and Quality Report | Zhimin Zhan, agileway.substack.com | link | 2025 | Independent critique |
For whole-industry testing figures, see the software testing statistics page. Global App Testing's managed crowdtesting tests apps on real devices in real markets: speak to us.