100+ UX Statistics for 2026 (With Original UXCam Research)

Original UXCam research: mobile UX benchmarks from 1.6 billion sessions

Most UX statistics come from surveys, where people describe what they think they do. Behavioral data is different. UXCam's Mobile UX Benchmarks 2025 report measured 1,621,517,342 real sessions from 1,538 mobile apps, each with at least 1,000 sessions, and compared 2023 with 2024. The apps span sectors from automotive and retail to financial services, healthcare and gaming.

The headline finding is that crashes are no longer where most apps lose people. Stability has become a baseline, while slower, quieter friction like frozen screens and repeated taps now separates good experiences from average ones.

• 99%+ of sessions are crash-free in most sectors, and crash-free rates stayed stable even when usage rose sharply.

• 93% to 97% of sessions are free of UI freezes in a typical sector. Gaming apps dipped as low as 82% during peak usage.

• 0.18 rage taps per session is the cross-industry average.

• +66.2% average session time in retail apps, which grew from 3.25 to 5.4 minutes. Total time per retail user rose 21.6%, from 29.6 to 36.0 minutes, and sessions per user rose 17.9%.

• −35.2% monthly sessions per user in food and beverage apps, which fell from 9.1 to 5.9. Session time also dropped from 1.88 to 1.6 minutes as ordering flows got faster.

• +33.3% rage taps per session in financial services apps, which rose from 0.09 to 0.12. UI-freeze-free sessions slipped from 94.2% to 93.3%, while crash-free sessions held at 99.4%.

• +35.7% screen visits per session in financial services apps, as banks and fintechs pushed users toward more features.

• 47.44 sessions per user per month in real estate apps and 31.66 in automotive, the highest engagement in the dataset. Both involve long research-heavy decisions.

• 12.90 screens per session in sports apps and 12.30 in non-profit apps, against 3.50 in real estate and 3.70 in automotive.

Session length is not a quality score on its own. Pharmaceutical and sports apps hold attention for 10 minutes or more because users research and browse. Food and beverage apps win by getting people from menu to checkout in under two minutes. The useful comparison is your app against its own sector, and your own trend over time.

The finance numbers show the trade-off most clearly. As banks added features, users visited more screens, and rage taps per session climbed by a third. Crash reporting would have shown a healthy app the whole time. Rage taps, UI freezes and session replay are what reveal that kind of friction.

Original UXCam research: finance app UX in the UK and Brazil

Finance apps are where friction costs the most, so UXCam ran two regional deep dives in 2026. The UX Benchmarks for Finance Apps, UK Edition 2026 covers 12.2 million sessions across banking, financial services and fintech apps over 12 months. The Brazil edition covers 26.9 million sessions across banking, financial services and insurance apps.

• 4.72 times a month is how often the median UK banking user opens their app. A typical visit lasts about one minute and covers 6.42 screens.

• 4.75 times a month is the UK fintech equivalent, with 4.08 screens per visit. On a retention dashboard the two sub-sectors look identical.

• 5.39% is the median rage tap rate in UK fintech apps, against 1.19% in banking. That 4.5x gap is the largest single-metric difference in UK finance.

• 2.23% is the median rage tap rate in UK financial services apps, 87% higher than banking.

• Under 2% rage is what the top quartile of UK fintech apps achieve, with the same four to five monthly sessions as the median app.

• 3.2x is the quality gap in Brazilian financial services apps. The top 25% run at 1.65% rage, while the bottom 25% reach 5.30%, the widest spread of any sub-sector.

• 2.64 sessions per user per month makes Brazilian financial services the sub-sector with the weakest return rate.

• 3 visits a month is the norm for the median Brazilian banking user, at 74 seconds and 3.87 screens per visit.

• 50% more rage taps come from new users than from returning users in Brazilian finance apps.

• 60.9% of users are new in high-acquisition Brazilian finance apps, which average 2.26 sessions per user. Low-acquisition apps have 23.2% new users and 4.16 sessions per user.

UK users open a banking app 4.72 times a month... they're not loyal, they're locked in. Apps that earn preference reduce friction, not add features.

The lesson travels well beyond finance. Sessions per user can look healthy while users are quietly frustrated, because people keep returning to apps that hold their salary, savings or payments. UXCam's customer success team suggests a simple routine. Pair sessions per user with the rage tap rate on your five busiest flows, then send any flow above 3% to session replay first.

Original UXCam research: how product teams use AI in 2026

AI now shapes how product and UX teams find problems, so UXCam surveyed 157 product professionals across 34 countries for The State of AI in Product Management 2026. The question was no longer whether teams use AI. It was how deep that use goes and what still holds it back.

• 99% of product professionals use AI in some form at work. Only two of the 157 describe their team as barely using it.

• 41% say AI is core to their team's workflows, 33% say most of the team uses it regularly, and 25% say use is individual and varies by person.

• 70% spend at least five hours a week with AI tools, and 41% spend five to 15 hours.

• 2.6 AI tools at once is the average for a product professional.

• 6.7 out of 10 is the mean pressure to adopt AI. It climbs to 7.7 in teams where AI is integrated and sits at 4.9 where use is individual.

• 7.40 out of 10 is the pressure score at companies with 1,000 to 10,000 employees, the highest of any size band.

• 47% of people at companies with 1 to 50 employees say AI is core to team workflows, against 30% at companies with more than 10,000 employees.

• 30% name budget or access as their biggest blocker, ahead of model quality (19%), security or policy (18%) and no time to learn (16%).

• 70% of leaders increased AI budgets, yet only 6% set explicit AI OKRs.

• 53% call the AI hype mostly accurate, 23% call it overblown, 13% call it underrated and 11% say it's too early to tell.

• 58% of self-described skeptics say AI speeds up their work significantly, only eight points behind believers at 66%.

AI moved from being a productivity tool to an operational teammate, catching issues and surfacing insights our traditional processes and dashboards were missing entirely.

Two findings matter most for UX work. First, deep adoption shrinks as companies grow, so large teams often have the budget but not the shared habits. Second, four of the five biggest blockers are internal rather than technical. Teams that give AI direct access to product data, as UXCam does with Tara, its AI analyst, skip the step where someone exports numbers into a chat window.

UX ROI and business value statistics

The most quoted UX statistic is a simple multiplier: every dollar invested in UX returns a hundred. The best-known long-term study links design to faster growth, and the best 2025 and 2026 research measures what bad experiences cost in lost spending and lost customers.

• $100 returned for every $1 invested in UX, an ROI of 9,900%, is the most quoted UX statistic. It's usually credited to Forrester Research's report The Six Steps For Justifying Better UX. The report is paywalled and the figure doesn't appear in its public summary, so treat it as a widely quoted estimate rather than a measured result.

• 32 percentage points higher revenue growth and 56 points higher shareholder returns: that's how the top quarter of companies on the McKinsey Design Index compared with industry peers from 2013 to 2018. McKinsey's study tracked 300 listed companies. It shows a correlation rather than proof of cause, and it's still the most cited study on design and business performance.

• $3 trillion in global sales is at risk in 2026 because of bad customer experiences, including $973 billion in the US. The estimate comes from a Qualtrics XM Institute survey of 20,000+ consumers in 14 countries.

• 11% of consumer experiences worldwide are bad, and 47% of those lead customers to cut their spending (Qualtrics XM Institute).

• 4.1x more likely to recommend, 3.8x more likely to trust and 2.3x more likely to buy more: that's how satisfied customers compare with dissatisfied ones (Qualtrics).

• 26% of 225 US brands significantly improved their customer experience in 2026, against 7% that declined. It was the first year-over-year rise in Forrester's CX Index since 2021, led by emotional quality.

• 2.7x better business outcomes are reported by organizations that build user research into strategy, along with 3.6x more active users, compared with those that rarely use user insights (Maze, 2025).

• 90% of product builders say design is at least as important as before AI, and nearly six in 10 say it matters more (Figma 2026 AI report).

Qualtrics' method is worth copying for your own business case. Multiply the share of experiences that go wrong by the share of those that lead to lower spending, then apply it to revenue. Inside a product, the same logic works with funnel data: measure how many users hit a broken step and how many of them leave. That turns "UX matters" into a revenue figure your finance team can check.

Customer experience and loyalty statistics

Customers in 2026 are less forgiving and quieter about it. Many don't complain before they leave, companies tend to overestimate their loyalty, and AI in customer service has raised expectations for both speed and transparency.

• 52% of US consumers stopped using or buying from a brand because of a bad experience with its products or services. Another 29% stopped because of poor customer experience online or in person, according to PwC's 2025 Customer Experience Survey of 5,511 consumers.

• 9 in 10 executives say customer loyalty has grown in recent years, but only 4 in 10 consumers say the same (PwC).

• 70% of executives say customer expectations are changing faster than their company can adapt (PwC).

• 30% of consumers tell no one after a bad experience and simply switch brands (Qualtrics).

• 64% of consumers want tailored experiences, but only 39% believe organizations use their personal data responsibly (Qualtrics).

• 85% of CX leaders say customers will drop brands that can't resolve issues on first contact, according to Zendesk's CX Trends 2026 research with 11,000+ consumers and business leaders.

• 74% of consumers get frustrated when they have to repeat information, and 81% want agents to pick up where they left off (Zendesk).

• 95% of consumers expect a clear explanation for decisions made by AI (Zendesk).

• 87% of customers say companies that use generative AI for service must offer a way to reach a human, according to a Gartner survey of 3,566 customers in 2026.

• Nearly 1 in 5 consumers who used AI for customer service got no benefit from it, a failure rate almost four times higher than for AI use in general (Qualtrics).

• 24% of service and support leaders could show positive financial returns from their AI use cases (Gartner, 2026).

• 37% of loyal consumers would let an AI agent switch them to another brand for a better fit, according to Accenture's 2026 survey of 25,590 consumers.

The silent switchers are the group that UX analytics helps most with. A customer who leaves without complaining never shows up in support tickets or surveys, but their last sessions are still on record. Rage taps, repeated attempts and abandoned flows in those sessions often explain churn that no feedback form would catch.

Mobile app usage and retention statistics

Mobile is where most digital experiences now happen, and the numbers from 2025 show usage still growing while attention gets harder to keep. People install more apps than ever, but most of those installs never turn into a habit.

• 3.6 hours a day is how long the average mobile user spent in apps in 2025. That adds up to 5.3 trillion hours across iOS and Google Play, up 3.8% (Sensor Tower State of Mobile 2026).

• Nearly 150 billion apps were downloaded in 2025 (+0.8%), while in-app purchase revenue grew 10.6% to $167 billion. Non-game apps out-earned games for the first time (Sensor Tower).

• 3.8 billion downloads of generative AI apps in 2025 doubled the year before, and users spent 48 billion hours in them (Sensor Tower).

• +10% app installs and +7% app sessions in 2025, up from 8% and 2% growth in 2024, according to Adjust's Mobile App Trends 2026.

• 27% / 13% / 5% is day-1, day-7 and day-30 retention across all mobile games in 2025, unchanged from 2024 (Adjust).

• 12.6% of shopping app users return on day 1 and only 3% on day 30 (Adjust).

• 12% day-1 and 2% day-30 retention for finance apps in 2025, down from 13% and 3% (Adjust).

• 46.1% of installs of the average Android app were uninstalled within 30 days in 2024, and most uninstalls happen on the first day, according to AppsFlyer's App Uninstall Report 2025.

• 91% of US adults own a smartphone, and 16% are smartphone-only internet users, according to the Pew Research Center (2026 data).

• 59.9% of worldwide web page views came from mobile devices in October 2026, against 38.5% from desktop (StatCounter).

Retention is where UX shows up in the numbers. A shopping app that keeps 3% of users after a month is typical, which means the first session carries most of the weight. For benchmarks by category, see our guide to mobile app retention benchmarks by industry. To find where your own users drop off in that first session, start with onboarding funnels and watch the sessions that end early.

The mobile share of web traffic also means "mobile UX" and "web UX" are no longer separate problems. Most people meet a brand on a phone first, often in a browser, before they ever install its app. Teams that analyze apps and websites in one place can follow that whole journey instead of guessing where it broke.

App performance and stability statistics

Performance is the part of UX that users notice only when it fails. App stores now treat stability as a ranking signal, and users punish unstable apps in their reviews.

• 1.09% user-perceived crash rate and 0.47% ANR rate are Google Play's overall "bad behavior" thresholds, with 8% applying to any single phone model. Above them, Play may reduce an app's visibility and show a warning on its store listing (Android Developers).

• February 2027 is when apps that exceed new thresholds for memory use, bitmap memory or code optimization may also see their Google Play visibility affected (Android Developers).

• 1,354,418 App Store submissions were rejected under Apple's performance guidelines in 2025, more than any other category. Those guidelines cover crashes, bugs and incomplete apps, as well as metadata and device compatibility (Apple App Store Transparency Report).

• Over 40% of unresolved App Review issues relate to app completeness, which covers crashes, placeholder content and incomplete information (Apple Developer).

• 40% faster cold starts and 30% fewer app-not-responding errors followed Reddit's switch to Android's R8 optimizer, along with higher Play Store ratings (Android Developers Blog, 2025).

• 5% more driver sessions followed a 21% cut in the Lyft Driver app's startup time. This Android Developers case study is older than most data here, but it's a rare measured link between speed and usage.

• Half of all 1-star Google Play reviews mentioned app stability, according to a 2017 Google analysis (Android Developers Blog).

These figures line up with UXCam's own data. Crash-free rates above 99% are now normal, so a crash-free dashboard alone says little about the experience. A crash rate that looks small on a chart can still cost an app its store visibility. Neither metric captures the freezes and unresponsive taps that frustrate users without crashing anything. Issue analytics that combine crashes, UI freezes and rage taps give a fuller picture.

Website UX and page speed statistics

Website performance has improved steadily, yet most mobile sites still fail at least one of Google's Core Web Vitals. Speed is the easiest part to measure, which makes it a useful starting point, but rage clicks and errors usually cost more than a slow page.

• 48% of mobile websites and 56% of desktop websites passed all Core Web Vitals in 2025, according to the HTTP Archive Web Almanac 2025.

• 8.4% more retail conversions and 10.1% more travel conversions followed a 0.1-second gain in mobile site speed. The source is Deloitte's Milliseconds Make Millions study for Google, from 2020 but still the standard reference.

• 77% of mobile websites have good Interaction to Next Paint, but only 62% have good Largest Contentful Paint and 44% good Time to First Byte. On desktop, the figures are 97%, 74% and 55% (Web Almanac 2025).

• 36% more conversions year over year followed QuintoAndar's work to cut Interaction to Next Paint by 80%, according to a 2025 web.dev case study. The company credits faster responses as a major factor, though not the only one.

• 75% of 150+ leading mobile ecommerce sites offer only "mediocre" UX, and just one site rates "good", according to Baymard's 2026 mobile UX benchmark.

Core Web Vitals passing on fewer than half of mobile sites is a reminder that the phone experience still lags the desktop one, even on the web. Speed scores also miss most of the problems that frustrate visitors. Those appear as repeated clicks on things that look clickable, forms that reject valid input and pages that shift as users tap. Heatmaps and session replay on your highest-traffic pages are the fastest way to see them.

Ecommerce, checkout and form UX statistics

Checkout is where UX turns directly into revenue, and it remains one of the most studied flows in digital design. The Baymard Institute's long-running research shows how much of the problem is avoidable.

• 70.22% is the average documented cart abandonment rate, based on 50 studies collected by the Baymard Institute.

• 18% of US shoppers who abandoned a checkout left because the site wanted them to create an account, and 17% because the process was too long or complicated (Baymard).

• 23.48 form elements are shown by default in the average US checkout, or 14.88 counting form fields alone (Baymard).

• 35.26% is the conversion gain the average large ecommerce site could make through better checkout design (Baymard).

• 65% of 344 top-grossing US and European ecommerce sites have "mediocre" or worse checkout UX, and the average site has 32 checkout improvements left to make (Baymard).

• $262 billion, or 20% of global online holiday sales in 2025, was influenced by AI and agents. Shoppers arriving from AI-powered search converted nine times more often than shoppers from social media, according to Salesforce data on 1.5 billion+ shoppers.

Most checkout friction is not a mystery. Forced account creation, long forms and unclear errors have topped abandonment surveys for years. What's harder is finding which field or step causes the problem in your own flow. A funnel shows the step where users leave, and replays of the sessions that abandoned there usually show why within a few minutes.

Accessibility statistics

Accessibility is now both a UX issue and a legal one. Automated scans keep finding the same basic failures, and the number of lawsuits is rising in the US while new rules apply in the EU.

• 95.9% of the top one million home pages had detectable WCAG 2 failures in 2026, up from 94.8% in 2025, according to The WebAIM Million.

• 56.1 accessibility errors were detected per home page on average, 10.1% more than in 2025 (WebAIM).

• 83.9% of home pages had low-contrast text below the WCAG 2 AA threshold, up from 79.1% (WebAIM).

• 1.3 billion people, or 1 in 6, live with a significant disability, according to the World Health Organization (2023).

• June 28, 2025 is when the European Accessibility Act came into effect, covering websites, smartphones, ecommerce and banking services sold in the EU.

• 5,114 ADA digital accessibility lawsuits were filed in US federal and state courts in 2025, and 1,427 of them targeted companies that had been sued before (UsableNet).

• 3,117 website accessibility lawsuits were filed in US federal court in 2025, 27% more than in 2024 (Seyfarth Shaw, via JD Supra).

Low contrast is the most common failure, and it's also among the easiest to fix. Accessible design tends to help everyone: larger tap targets, clear labels and readable text reduce mis-taps and rage taps for every user, not only those using assistive technology.

AI and UX statistics

AI changes UX from two directions. It changes how teams research and build, and it changes how users arrive at a product. Both shifts show up clearly in 2025 and 2026 data.

• 51% of Figma users working on AI products are building agents, up from 21% a year earlier, according to Figma's 2025 AI report, a survey of 2,500 product builders.

• 78% say AI makes their work significantly more efficient, but only 32% say they can rely on its output (Figma).

• 80% of user researchers now use AI, up 24 points in a year, while 91% worry about accuracy and hallucinations, according to the User Interviews State of User Research 2025.

• 69% of research practitioners use AI in at least some projects, and 66% saw demand for research rise, up from 55% (Maze, 2026).

• 8% of Google visits ended with a click on a regular result when an AI summary appeared, against 15% without one. Users clicked a link inside the summary in only 1% of visits (Pew Research Center).

• 1.1 billion referral visits came from AI platforms to websites in June 2025, up 357% year over year (Similarweb).

The gap between efficiency and trust is the theme that connects these numbers, and it matches UXCam's own survey. Teams use AI every day but still check its output, which is why AI analysis works best when every claim links back to evidence. In UX analytics, that evidence is the session itself. Visitors also arrive with less context than before. Many come from an AI answer that skipped the page they used to read first, so the landing experience has to do more work.

UX research and design team statistics

UX teams are under more pressure to show impact, and AI is changing how they work faster than headcount is changing. Research is also gaining influence at the top of some organizations while losing staff in others.

• 22% of organizations now treat research as essential at every level of business strategy, nearly triple the 8% in 2025, according to Maze's Future of User Research 2026.

• 21% of user researchers said their company laid off UX researchers in 2025, and 49% feel negative about the future of UX research, up 26 points (User Interviews).

• 2x more likely to feel good about their work: that's how designers fare when leaders prioritize design excellence (Figma State of the Designer 2026, 906 designers).

• 73% of companies sit at the middle level of UX maturity or lower, and just 2% treat UX as a company-wide strategy. The source is a 2026 MHP study of 100 professionals and managers.

• Half of 5,371 UX professionals placed their organization at stage 3 of 6 on the Nielsen Norman Group's UX maturity scale. Only 4% reached the "integrated" stage (2022, the latest data).

Research teams with fewer people and more AI tools need faster access to evidence. Behavioral data fills the gap between formal studies: a researcher can check how often a usability issue happens across thousands of real sessions before deciding it deserves a full study.

UX statistics for 2026 and beyond: forecasts

Forecasts are less certain than measured data, so treat these as direction rather than fact. Taken together, they point to more mobile traffic, more AI between users and products, and stricter quality bars from the platforms.

• 2026: 29% of adults in developed countries will see at least one AI search summary every day, and up to 75% of companies may invest in agentic AI (Deloitte).

• 2026: Sensor Tower expects in-app purchase revenue from generative AI apps to pass $10 billion, with ChatGPT becoming the top-earning app (Sensor Tower).

• 2026: IDC forecasts a record 16.7% drop in smartphone shipments to just over 1 billion units, as the average price rises 27.6% to $581 (IDC).

• 2026: Forrester predicts that a third of companies will harm their customer experience with frustrating AI self-service (Forrester).

• 2027: Google Play may reduce the visibility of apps that exceed new memory and code optimization thresholds from February 2027 (Android Developers).

• 2028: Visitors from AI search will outnumber visitors from traditional search by early 2028, according to Semrush. By conversion rate, an AI search visitor is worth 4.4 times as much.

• 2028: Agentic AI could create up to $450 billion in economic value, according to the Capgemini Research Institute. Yet executive trust in fully autonomous agents fell from 43% to 27% in a year.

• 2029: Agentic AI will resolve 80% of common customer service issues without human help, cutting operational costs by 30%, according to a Gartner prediction.

• 2030: Gartner expects the cost per resolution of generative AI in customer service to exceed $3, more than many offshore human agents cost (Gartner).

• 2030: The agentic AI market could reach $35 billion, up from $8.5 billion in 2026, according to Deloitte.

• 2030: 57% of all mobile connections will run on 5G, according to the GSMA's Mobile Economy 2026.

• 2030: Consumers are forecast to spend $288 billion a year in app stores, according to a 2024 forecast by data.ai, now part of Sensor Tower.

• 2031: Smartphone subscriptions will grow from 7.47 billion to 8.49 billion, according to the Ericsson Mobility Report. Mobile data use per smartphone will nearly double from 22 GB to 42 GB a month (Ericsson).

For UX teams, the practical reading is simple. More of the journey will start somewhere you don't control, like an AI answer or an agent acting for the user. The first screen people reach will carry more weight. Platforms will also keep raising technical bars, as Google's 2027 memory rules show. Teams that already measure friction across apps and websites will adapt to both shifts faster.

UX statistics that don't hold up (and what to cite instead)

Many UX statistics you see on slides and in listicles have drifted far from their source. Some come from studies more than 15 years old, some measured something much narrower than the headline, and a few have no traceable study behind them at all. We traced the most common ones back to their origin.

The pattern is worth remembering when you build a business case. Old web performance findings get relabeled as UX statistics. Survey answers about what people would do get quoted as behavior, and one company's anecdote becomes an industry rule. Stakeholders who check sources will find the same gaps, so a smaller set of verifiable numbers persuades better than a long list of dramatic ones.

How we chose these UX statistics

Every statistic on this page links to the organization that collected the data, not to another list of statistics. We opened each source to confirm the number and its wording, and we kept the original framing where a study measured something narrower than a headline suggests. Third-party figures come from reports published in 2025 or 2026, unless a study is still the standard reference in its field and nothing newer exists. In that case we name its year.

UXCam's own figures come from three behavioral datasets and one survey. The Mobile UX Benchmarks 2025 report covers 1.62 billion sessions from 1,538 apps with at least 1,000 sessions each, comparing 2023 with 2024. The UK and Brazil finance reports cover 12.2 million and 26.9 million sessions over 12 months. All three use anonymized, aggregated session data. The State of AI in Product Management 2026 is a survey of 157 product professionals in 34 countries. We last checked every link and number on November 5, 2026, and we update this page when a major report publishes a new edition.

How to measure UX in your own product

Industry benchmarks tell you where the bar sits, but your own data tells you why users struggle in your product. The metrics behind most statistics on this page are measurable in any app or website: crash-free sessions, UI freezes, rage taps, session length, screens per session, funnel drop-off and retention.

UXCam captures those signals automatically in mobile apps and on websites, without manual event tagging. Session replay shows what happened in a frustrated session, heatmaps show where taps and clicks cluster, and funnels show where users leave. Tara, UXCam's AI analyst, watches sessions for you and explains which issues cost the most users, so you can compare your numbers with the benchmarks above in minutes.