---
title: "90+ Product Management Statistics and Trends for 2026"
description: "90+ product management statistics for 2026 on jobs, pay, AI and how teams decide, each linked to its source, plus UXCam's survey of 157 product pros."
url: https://uxcam.com/blog/product-management-statistics/
language: en-US
date_published: 2025-01-03T23:00:00.000Z
date_modified: 2026-11-06T00:00:00.000Z
---

# 90+ Product Management Statistics and Trends for 2026

Product management statistics for 2026 show a profession that is growing and changing at the same time. Open PM roles are at their highest level since 2022 and pay is rising, while AI blurs the line between product, design and engineering work. This page collects 90+ statistics, most from 2025 and 2026 research. It also includes UXCam's survey of 157 product professionals in 34 countries and engagement benchmarks from 1.6 billion app sessions. Every figure links to its primary source.

**The product management statistics to know in 2026**

| Statistic | Source |
| --- | --- |
| Open product manager jobs reached 7,300+ in March 2026, the most since 2022 | [Lenny's Newsletter](https://www.lennysnewsletter.com/p/state-of-the-product-job-market-in-ee9), 2026 (TrueUp data) |
| The median US product manager earns $233,000 a year in total pay | [Levels.fyi](https://www.levels.fyi/t/product-manager), 2026 |
| Small companies have about 5 engineers per product manager, and companies with 3,000+ staff have about 11 | [Pave](https://www.pave.com/blog-posts/research-development-org-structure-benchmarks), 2024 |
| 99% of product professionals use AI at work | [UXCam](https://uxcam.com/resources/ebooks/ai-in-product-management-2026/), 2026 |
| Only 6% of product teams have set goals (OKRs) for their use of AI | [UXCam](https://uxcam.com/resources/ebooks/ai-in-product-management-2026/), 2026 |
| Claude is the most used AI tool among product professionals, at 80%, ahead of ChatGPT at 68% | [UXCam](https://uxcam.com/resources/ebooks/ai-in-product-management-2026/), 2026 |
| 76% of product professionals worry that their product will fail | [Atlassian](https://dam-cdn.atl.orangelogic.com/CDNLink/AT12OX96.pdf), 2026 |
| 77% of product teams say executive gut feeling overrides data at least sometimes | [Atlassian](https://dam-cdn.atl.orangelogic.com/CDNLink/AT12OX96.pdf), 2026 |
| Only 25% of product professionals spend most of their time on strategy | [Product Focus](https://www.productfocus.com/wp/wp-content/uploads/2026/03/Product-Focus-Industry-Survey-Report-2026.pdf), 2026 |
| 71% of product professionals say they don't spend enough time with customers | [Product Focus](https://www.productfocus.com/wp/wp-content/uploads/2026/03/Product-Focus-Industry-Survey-Report-2026.pdf), 2026 |
| Only 8% to 33% of A/B tests improve their target metric, even at Microsoft and Netflix | [Kohavi et al.](https://airbnb.tech/wp-content/uploads/sites/19/2025/10/ABTestingIntuitionBusters.pdf), 2022 |
| 46% of developers don't trust the accuracy of AI tools | [Stack Overflow](https://survey.stackoverflow.co/2025/ai), 2025 |
| Only 23% of organizations are scaling AI agents | [McKinsey](https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai), 2025 |
| 40% of enterprise apps will include AI agents by the end of 2026, up from less than 5% in 2025 | [Gartner](https://www.gartner.com/en/newsroom/press-releases/2025-08-26-gartner-predicts-40-percent-of-enterprise-apps-will-feature-task-specific-ai-agents-by-2026-up-from-less-than-5-percent-in-2025), 2025 |

Each source links to the original study.

## Original UXCam research: the state of AI in product management 2026

UXCam surveyed 157 product professionals in 34 countries for [The State of AI in Product Management 2026](https://uxcam.com/resources/ebooks/ai-in-product-management-2026/), published in June 2026. Most respondents (62%) work in the product management track, from individual contributor to VP. The rest work in design and UX (14%), other product roles (12%), and engineering or research (12%). Company sizes are evenly spread, from startups with fewer than 50 people to enterprises with more than 10,000.

- **99%** of product professionals use AI at work in some form. Only 1% say their team barely uses 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. That puts 74% beyond solo experimentation.
- **70%** spend at least five hours a week with AI tools, and **41%** spend between five and 15 hours.
- **2.6** AI tools at once is the average for a product professional. Most run two or three, and few rely on just one.
- **80%** use Claude, **68%** use ChatGPT and **51%** use Gemini. Custom or internal AI tools reach 31%, ahead of Perplexity (13%) and Notion AI (10%).
- **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. Companies with 51 to 200 employees feel the least pressure, at 6.08.
- **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. That 17-point gap runs the opposite way to budgets.
- **30%** name budget or access as their biggest blocker. Output quality (19%), security or policy (18%), no time to learn (16%) and "I already use enough" (16%) follow.
- **70%** of leaders increased AI budgets, yet only **6%** of teams have set explicit AI OKRs.
- **53%** say the AI hype is 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%.

> By the metrics that mattered in 2024, the adoption argument is over. 99% of product teams use AI. The question for 2026 is whether that usage has shape, or whether budgets are just growing faster than goals.
> 
> Dr. Silvanus Alt, CEO, UXCam

Three findings matter most for product leaders. First, pressure to use AI is high, but goals are rare. Teams that set even one measurable AI goal will learn faster than teams that only raise budgets. Second, four of the five biggest blockers are internal, led by access to tools, so procurement and policy often matter more than model choice. Third, deep adoption shrinks as companies grow. Large teams tend to have the budget but not the shared habits, so a working practice in one squad rarely spreads on its own.

The frustrations respondents described point the same way. The learning curve came up most often, followed by token costs and limits, hallucinations and the effort of giving AI enough context. That last one is easiest to fix in analytics work. When an AI analyst such as UXCam's [Tara](https://uxcam.com/ai/) reads product data directly, nobody has to export numbers into a chat window first.

## Original UXCam research: product engagement benchmarks from 1.6 billion sessions

Product managers set engagement targets, but a good number depends heavily on the category. UXCam's [Mobile UX Benchmarks 2025](https://resources.uxcam.com/ux-benchmarks) report measured 1,621,517,342 sessions from 1,538 mobile apps, each with at least 1,000 sessions, and compared 2023 with 2024.

- **47.44** sessions per user per month in real estate apps and **31.66** in automotive apps, the most frequent use in the dataset. Both involve long, research-heavy decisions.
- **11.3 minutes** is the average session in pharmaceutical apps and **10.1 minutes** in sports apps, against **1.6 minutes** in food and beverage apps.
- **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.
- **+66.2%** average session time in retail apps, which grew from 3.25 to 5.4 minutes. Sessions per user rose 17.9% over the same period.
- **−35.2%** monthly sessions per user in food and beverage apps, which fell from 9.1 to 5.9 as ordering flows got faster.
- **+35.7%** screen visits per session in financial services apps, while rage taps per session rose 33.3%, from 0.09 to 0.12.
- **+13%** sessions in automotive apps, while time spent per user fell 19%.
- **99%+** of sessions are crash-free in most sectors, and **93% to 97%** are free of UI freezes. The average app sees **0.18** rage taps per session.

More engagement is not always better. A food delivery app that gets people from menu to checkout in under two minutes is doing its job. The same session length would be a warning sign in a research-heavy real estate app. The finance numbers show the risk of chasing feature adoption alone: users visited more screens, and their frustration grew with it. Compare your product with its own category and with its own trend, and pair every engagement metric with a friction metric such as rage taps. For retention targets by category, see our [mobile app retention benchmarks](https://uxcam.com/blog/mobile-app-retention-benchmarks/).

## AI in product management statistics

UXCam's survey is one view of AI in product work. Larger surveys of product teams, developers and designers show the same pattern: use is close to universal, while trust and measurable returns lag behind.

- **69%** of product people use AI frequently, up from 49% a year earlier, according to the Product Focus [Product Management Survey 2025/26](https://www.productfocus.com/product-management-resources/industry-survey/) of 677 people in 40 countries.
- **89%** of product teams regularly use AI for tasks another function used to handle. And **67%** say AI has made their own role harder to define ([Atlassian State of Product 2027](https://dam-cdn.atl.orangelogic.com/CDNLink/AT12OX96.pdf)).
- **53%** of product professionals have more time for strategic planning and roadmaps thanks to AI ([Atlassian State of Product 2027](https://dam-cdn.atl.orangelogic.com/CDNLink/AT12OX96.pdf)).
- **80%** of senior product leaders use AI to run or synthesize customer research, and 34% use it heavily. The source is a [Lyssna and Delight Path survey](https://www.lyssna.com/blog/product-leaders-ai-research-survey/) of 107 product leaders in 2026.
- **26%** of product managers fear AI could eventually replace them, and only **39%** have had comprehensive, job-specific AI training. The figures come from a [General Assembly survey](https://resource-center.generalassemb.ly/blog/ai-and-product-management-survey/) of 117 PMs, a small sample from a training provider.
- **84%** of developers use or plan to use AI tools, and **51%** of professional developers use them daily, according to the Stack Overflow [2025 Developer Survey](https://survey.stackoverflow.co/2025/ai) of 49,009 developers.
- **46%** of developers distrust the accuracy of AI tools, against **33%** who trust it. Their top frustration, named by 66%, is answers that are "almost right, but not quite" ([Stack Overflow](https://survey.stackoverflow.co/2025/ai)).
- **90%** of technology professionals use AI at work, and more than 80% say it has made them more productive. Yet **30%** have little or no trust in AI-generated code, according to Google's [DORA 2025 report](https://cloud.google.com/blog/products/ai-machine-learning/announcing-the-2025-dora-report) of nearly 5,000 professionals.
- **80%** of new developers on GitHub use Copilot in their first week, according to GitHub's [Octoverse 2025](https://github.blog/news-insights/octoverse/octoverse-a-new-developer-joins-github-every-second-as-ai-leads-typescript-to-1/).
- **1 in 3** product builders planned to launch AI-powered products in 2025, up 50% on the year before. Of those building AI products, **51%** are building agents ([Figma 2025 AI report](https://www.figma.com/blog/figma-2025-ai-report-perspectives/), 2,500 users).
- **25%** of developers said AI saved them 10+ hours a week in 2026, down from 68% in 2025, and **65%** regularly correct AI output. That's according to Atlassian's [State of Developer Experience 2026](https://www.atlassian.com/blog/development/developer-experience-report-2026).
- **17%** of developers and 23% of leaders say their organization has a clearly communicated and implemented AI strategy ([Atlassian](https://www.atlassian.com/blog/development/developer-experience-report-2026)).
- **23%** of organizations are scaling an agentic AI system somewhere in the business, and another 39% are experimenting with agents. The source is McKinsey's [State of AI 2025](https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai) survey of 1,993 participants.
- **39%** of organizations attributed any EBIT impact to AI in McKinsey's 2025 survey, and most of them put it below 5%. The [2026 edition](https://www.mckinsey.com/~/media/mckinsey/business%20functions/quantumblack/our%20insights/the%20state%20of%20ai/the-state-of-ai-in-2026-on-the-road-to-roi-aug-2026.pdf) found the share essentially unchanged, at 37%.

The gap between use and value is the story of 2026. Developers who once reported big time savings now report smaller ones and more time spent correcting output. Teams that get value from AI tend to point it at a specific, measurable job, such as finding where users drop off, rather than at work in general.

## Product manager job market statistics

Hiring for product managers has recovered strongly from its 2023 low. Demand now outpaces design roles, and a growing share of openings is tied to AI products. Most job data covers tech companies, so treat these figures as a reading of the tech market rather than every industry.

- **7,300+** open product manager roles at tech companies worldwide in March 2026, the most since 2022. The count comes from [Lenny's Newsletter](https://www.lennysnewsletter.com/p/state-of-the-product-job-market-in-ee9), using TrueUp data on 9,000+ tech companies.
- **75%** above the early-2023 low is where PM openings stood in March 2026, and they rose **nearly 20%** in the first quarter alone ([Lenny's Newsletter](https://www.lennysnewsletter.com/p/state-of-the-product-job-market-in-ee9)).
- **6,000+** open PM roles a year earlier, in May 2025, already sat 53.6% above the 2023 bottom ([Lenny's Newsletter](https://www.lennysnewsletter.com/p/state-of-the-product-job-market-in)).
- **688** open AI product manager roles were listed in May 2025, as AI PM demand took off ([Lenny's Newsletter](https://www.lennysnewsletter.com/p/state-of-the-product-job-market-in)).
- **1.27x** as many open PM roles as open designer roles in 2026. Design openings have stayed flat at about 5,700 since early 2023 ([Lenny's Newsletter](https://www.lennysnewsletter.com/p/state-of-the-product-job-market-in-ee9)).
- **Over 23%** of open PM roles are in the San Francisco Bay Area, a share that has grown 50% since 2022 ([Lenny's Newsletter](https://www.lennysnewsletter.com/p/state-of-the-product-job-market-in-ee9)).
- **373** of the 1,985 Washington-state jobs Microsoft cut in May 2025 were product managers, nearly one in five. The figure comes from Microsoft's state layoff filing, as reported by [The Seattle Times](https://www.spokesman.com/stories/2025/may/19/hit-hardest-in-microsoft-layoffs-developers-produc/).

The two trends sit side by side. Overall demand for PMs is rising, while some large employers, Microsoft among them, have cut product roles. AI product management is one of the clearest areas of growth.

## Product manager salary statistics

Product management remains one of the best-paid roles in tech. Pay data varies a lot by source, though. Levels.fyi skews toward large tech employers, while broader surveys include smaller companies and other industries, so compare like with like.

- **$233,000** is the median total pay for a US product manager, including a **$187,000** median base salary. The figure comes from 4,921 salaries reported to [Levels.fyi](https://www.levels.fyi/t/product-manager) as of November 2026.
- **$170,000 to $454,000** is the range from the 25th to the 90th percentile of US PM total pay ([Levels.fyi](https://www.levels.fyi/t/product-manager)).
- **$316,800** is the median in the San Francisco Bay Area, against $307,500 in Seattle and $257,000 in New York ([Levels.fyi](https://www.levels.fyi/t/product-manager)).
- **$600,000** is the median PM pay at Netflix, the top-paying employer, followed by Airbnb at $575,000 and Meta at $540,000 ([Levels.fyi](https://www.levels.fyi/t/product-manager)).
- **+4.55%** was the growth in US product manager pay in 2025, ahead of software engineers at +2.67%, according to the Levels.fyi [End of Year Pay Report 2025](https://www.levels.fyi/2025/).
- **£104,837** is the median PM total pay in the UK, against €94,663 in Germany, ₹45.2 lakh in India and R$293,000 in Brazil. Samples outside the US and India are small ([Levels.fyi](https://www.levels.fyi/t/product-manager/locations/united-kingdom)).
- **$123,000** is the global median salary for product professionals across all company types, with North America at $160,000 and Europe at $100,000. The figure comes from the Product-Led Alliance [Product Management Salary Report 2025](https://www.productledalliance.com/product-management-salary-report-2025/).
- **$36,250** more is what fully remote PMs earn on average than office-based PMs, and $20,750 more than hybrid PMs ([Product-Led Alliance](https://www.productledalliance.com/product-management-salary-report-2025/)).
- **$3,500** is the average gender pay gap among product professionals in the same survey, with men earning more. The survey doesn't publish its sample size ([Product-Led Alliance](https://www.productledalliance.com/product-management-salary-report-2025/)).
- **3.8%** was the average pay rise for product people from 2025 to 2026, up from 2% a year earlier. That's according to the Product Focus [Industry Survey Report 2026](https://www.productfocus.com/wp/wp-content/uploads/2026/03/Product-Focus-Industry-Survey-Report-2026.pdf), a survey of 677 product people, mostly in Europe.

The gap between a $233,000 big-tech median and a $123,000 global median is mostly about who gets counted. Remote PMs also tend to be more senior, so the remote premium reflects seniority as much as location.

[![Academy Logo](https://images.ctfassets.net/vv1yxl437u7d/3EXg2beUQrNhgRjiZANS5J/6cd809bd82e2b77261aec1a8386fb7a7/logo-top.svg) Mobile App Product Management Certification Upskill for free Career growth Expert Instructors Practical Insights](https://uxcam.com/academy/mobile-app-product-management-certification/)

## Product team structure and role statistics

As companies grow, product managers cover more engineers and report through more varied lines. Where product reports has a clear effect on how much authority the role carries.

- **About 5** engineers per product manager is typical at companies with up to 100 employees, widening to **about 11** at companies with 3,000 or more. [Pave's](https://www.pave.com/blog-posts/research-development-org-structure-benchmarks) data covers 233,000+ R&D employees at 4,195 companies and was published in late 2024.
- **13%** of R&D headcount is in product management at small companies, falling to **7%** at large enterprises ([Pave](https://www.pave.com/blog-posts/research-development-org-structure-benchmarks)).
- **26%** of product management organizations report directly to a Chief Product Officer ([Product Focus](https://www.productfocus.com/wp/wp-content/uploads/2026/03/Product-Focus-Industry-Survey-Report-2026.pdf), 2026).
- **80%** of PMs who report to a CPO say product is seen as a leadership function, against about **50%** when product reports to engineering or sales ([Product Focus](https://www.productfocus.com/wp/wp-content/uploads/2026/03/Product-Focus-Industry-Survey-Report-2026.pdf)).
- **43%** of product people have managed products for more than 10 years, and **53%** come from a technical background ([Product Focus](https://www.productfocus.com/wp/wp-content/uploads/2026/03/Product-Focus-Industry-Survey-Report-2026.pdf)).
- **24%** of product professionals say product leads cross-functional work, down from **47%** a year earlier. That's according to Atlassian's [State of Product 2027](https://dam-cdn.atl.orangelogic.com/CDNLink/AT12OX96.pdf) report, which surveyed 1,000 product professionals in August 2026.
- **85%** of product team members say they can lead or influence product strategy, according to Atlassian's [State of Product 2026](https://dam-cdn.atl.orangelogic.com/AssetLink/iv3u81ou0u70gsy04ia61oci00d0h0je.pdf), a 2025 survey of 700 product professionals.
- **80%** of product teams don't involve engineers in ideation, problem definition or roadmaps ([Atlassian State of Product 2026](https://dam-cdn.atl.orangelogic.com/AssetLink/iv3u81ou0u70gsy04ia61oci00d0h0je.pdf)).

The drop in product-led collaboration is the number to watch. As AI lets engineers and designers take on product tasks, PMs keep their influence by owning the evidence: which users struggle, where, and what it costs.

## How product managers spend their time

Most product managers say they want more time for strategy and customers. The data shows how much of the week still goes to unplanned work, unclear roles and decisions made on instinct.

- **60%** of product people are often pulled into unplanned work, and only **25%** spend most of their time on strategy ([Product Focus](https://www.productfocus.com/wp/wp-content/uploads/2026/03/Product-Focus-Industry-Survey-Report-2026.pdf), 2026).
- **49%** of product teams don't have enough time for strategic planning and roadmaps. Only **3%** say they have enough time for every part of the job ([Atlassian State of Product 2026](https://dam-cdn.atl.orangelogic.com/AssetLink/iv3u81ou0u70gsy04ia61oci00d0h0je.pdf)).
- **71%** of product people say they don't spend enough time with customers, rising to **75%** among heads and directors ([Product Focus](https://www.productfocus.com/wp/wp-content/uploads/2026/03/Product-Focus-Industry-Survey-Report-2026.pdf)).
- **78%** of deadlines are met when PMs have clearly defined roles, against **48%** when roles are unclear. Yet **41%** say their role isn't clearly defined ([Product Focus](https://www.productfocus.com/wp/wp-content/uploads/2026/03/Product-Focus-Industry-Survey-Report-2026.pdf)).
- **34%** of product professionals have no clear primary metric, and **33%** report a weak or missing company strategy ([Product Focus](https://www.productfocus.com/wp/wp-content/uploads/2026/03/Product-Focus-Industry-Survey-Report-2026.pdf)).
- **77%** of product teams say executive gut instinct overrides data-driven decisions at least sometimes, and **52%** say it happens regularly ([Atlassian State of Product 2027](https://dam-cdn.atl.orangelogic.com/CDNLink/AT12OX96.pdf)).
- **Over 60%** of prioritization frameworks get overridden by leadership escalations, according to ProductPlan's [State of Product Management Report 2026](https://www.productplan.com/resources/product-management-survey/), a survey of 250 product leaders.
- **76%** of product professionals fear their product will fail in the market, down from 84% a year earlier. Only **13%** are very concerned ([Atlassian State of Product 2027](https://dam-cdn.atl.orangelogic.com/CDNLink/AT12OX96.pdf)).
- **72%** say they mostly do the work they love, down from **90%** a year earlier ([Atlassian State of Product 2027](https://dam-cdn.atl.orangelogic.com/CDNLink/AT12OX96.pdf)).
- **32%** say product development has become more cutthroat over the past three years, and **31%** say their work has moved from strategic to tactical ([Atlassian State of Product 2026](https://dam-cdn.atl.orangelogic.com/AssetLink/iv3u81ou0u70gsy04ia61oci00d0h0je.pdf)).

Two numbers explain much of the rest. A third of product people have no clear primary metric, and three in four see data overruled by opinion. Teams with one agreed metric, and evidence that anyone can check, spend less time arguing and more time on the work they want to do.

## Product discovery, experimentation and decision-making statistics

Good product decisions depend on evidence, and the data shows how often that evidence is missing or overruled. Even at companies that test constantly, most ideas don't improve the metric they target.

- **8% to 33%** of online experiments succeed at the companies that test the most. Microsoft reports 33%, Bing 15%, Booking.com, Google Ads and Netflix 10%, and Airbnb search 8%, according to a peer-reviewed [2022 paper by Kohavi, Deng and Vermeer](https://airbnb.tech/wp-content/uploads/sites/19/2025/10/ABTestingIntuitionBusters.pdf).
- **100+** new experiment variants start every business day at the largest tech companies ([Kohavi et al.](https://airbnb.tech/wp-content/uploads/sites/19/2025/10/ABTestingIntuitionBusters.pdf)).
- **40%** of product teams run little or no experimentation, and only 60% make it a regular part of their process ([Atlassian State of Product 2026](https://www.atlassian.com/blog/announcements/state-of-product-2026)).
- **Only 12%** of product professionals find driving measurable business results rewarding ([Atlassian State of Product 2026](https://www.atlassian.com/blog/announcements/state-of-product-2026)).
- **35%** of product leaders say AI has led them to do more customer research, while 8% now do less ([Lyssna](https://www.lyssna.com/blog/product-leaders-ai-research-survey/)).
- **49%** of product leaders say a small sample size is the main reason they question a research finding, ahead of cherry-picked results (45%) and missing statistical backing (42%) ([Lyssna](https://www.lyssna.com/blog/product-leaders-ai-research-survey/)).
- **AI adoption** goes with faster software delivery and better product performance, but also with worse delivery stability. DORA describes AI as an amplifier of each organization's existing strengths and weaknesses ([DORA 2025](https://cloud.google.com/blog/products/ai-machine-learning/announcing-the-2025-dora-report)).
- **518.7 million** pull requests were merged on GitHub in a year, up 29%, as teams shipped more code than ever ([GitHub Octoverse 2025](https://github.blog/news-insights/octoverse/octoverse-a-new-developer-joins-github-every-second-as-ai-leads-typescript-to-1/)).

If two out of three ideas fail even at Microsoft, the fastest teams aren't the ones with the best ideas. They're the ones that find out quickest which ideas work. Behavioral data shortens that loop: a funnel shows whether a change moved the metric, and the sessions behind it show why. It also answers the small-sample worry, because behavior from thousands of users backs up what a handful of interviews suggest.

## Software and app market statistics for product teams

Product teams build in a crowded market. Companies already run hundreds of apps, consumers spend more in apps every year, and AI products are taking a growing share of both.

- **$55.7 million** is the average organization's yearly SaaS spend, up 8%, according to Zylo's [2026 SaaS Management Index](https://zylo.com/saas-management-index/), based on 40 million+ licenses.
- **305** SaaS apps make up the average company's portfolio, a number that has held steady ([Zylo](https://zylo.com/saas-management-index/)).
- **108%** growth in spend on AI-native apps, rising to 393% at organizations with more than 10,000 employees. ChatGPT is now the most-expensed app ([Zylo](https://zylo.com/news/2026-saas-management-index)).
- **36%** of SaaS licenses go unused, and median SaaS spend reaches $9,455 per employee ([Zylo](https://zylo.com/news/2026-saas-management-index)).
- **$167 billion** was spent in apps and games in 2025, up 10.6%, while downloads grew only 0.8% to nearly 150 billion ([Sensor Tower State of Mobile 2026](https://sensortower.com/blog/state-of-mobile-2026)).
- **For the first time**, non-game apps out-earned games in 2025. Non-game in-app revenue grew 21%, against 1.3% for games ([Sensor Tower](https://sensortower.com/blog/state-of-mobile-2026)).
- **3.8 billion** downloads of generative AI apps in 2025 doubled the year before, and their in-app revenue nearly tripled to more than $5 billion ([Sensor Tower](https://sensortower.com/blog/state-of-mobile-2026)).
- **$6.37 trillion** is Gartner's forecast for worldwide IT spending in 2026, up 14.2% ([Gartner](https://www.gartner.com/en/newsroom/press-releases/2026-07-27-gartner-forecasts-worldwide-it-spending-to-grow-14-point-2-percent-in-2026-totaling-6-point-37-trillion)).

Flat downloads and rising revenue mean growth now comes from keeping and monetizing existing users, not just winning new ones. A third of paid licenses going unused says the same about B2B software: adoption after the sale decides renewals. Both put retention and feature adoption at the center of the product manager's job.

## Product management trends for 2026 and beyond: forecasts

Forecasts are less certain than measured data, so read these as direction rather than fact. Most of them describe the same shift: more AI agents inside products, and more of the work around products done by AI.

- **2026:** 40% of enterprise apps will include task-specific AI agents by the end of the year, up from less than 5% in 2025, according to a [Gartner prediction](https://www.gartner.com/en/newsroom/press-releases/2025-08-26-gartner-predicts-40-percent-of-enterprise-apps-will-feature-task-specific-ai-agents-by-2026-up-from-less-than-5-percent-in-2025).
- **2027:** Over 40% of agentic AI projects will be canceled by the end of 2027, because of rising costs, unclear business value or weak risk controls ([Gartner](https://www.gartner.com/en/newsroom/press-releases/2025-06-25-gartner-predicts-over-40-percent-of-agentic-ai-projects-will-be-canceled-by-end-of-2027)).
- **2028:** At least 15% of day-to-day work decisions will be made autonomously by agentic AI, up from 0% in 2024. A third of enterprise software apps will include agentic AI, up from less than 1% ([Gartner](https://www.gartner.com/en/newsroom/press-releases/2025-06-25-gartner-predicts-over-40-percent-of-agentic-ai-projects-will-be-canceled-by-end-of-2027)).
- **2028:** 90% of enterprise software engineers will use AI code assistants, up from less than 14% in early 2024 ([Gartner](https://www.gartner.com/en/newsroom/press-releases/2025-07-01-gartner-identifies-the-top-strategic-trends-in-software-engineering-for-2025-and-beyond)).
- **2030:** 170 million jobs will be created and 92 million displaced worldwide, a net gain of 78 million. The forecast comes from the World Economic Forum's [Future of Jobs Report 2025](https://www.weforum.org/media/future-of-jobs-report-2025-78-million-new-job-opportunities-by-2030-but-urgent-upskilling-needed-to-prepare-workforces), a survey of 1,000+ employers.
- **2030:** 39% of workers' core skills are expected to change, with AI and big data the fastest-growing skills ([World Economic Forum](https://www.weforum.org/impact/reskilling-revolution-preparing-1-billion-people-for-tomorrows-economy)).

For product managers, the practical reading is that the job is moving toward judgment and evidence. When agents can write specs, code and summaries, the scarce skill is knowing which problem matters and proving that a change worked. Teams that already measure real user behavior across apps and websites will adapt to that shift faster.

## Product management statistics that don't hold up (and what to cite instead)

Many product management statistics in slide decks and listicles have drifted from their source. Some were never measured at all, and others describe a narrow sample from years ago. We traced the most common ones back to where they started.

| The claim | What's actually behind it | Cite instead |
| --- | --- | --- |
| "95% of new products fail" | Credited to Clayton Christensen, but it came from an [HBS Working Knowledge article](https://www.library.hbs.edu/working-knowledge/clay-christensens-milkshake-marketing) from 2011 that has since been corrected. It now says only that "many" fail. | A [2013 review by Castellion and Markham](https://doi.org/10.1111/j.1540-5885.2012.01009.x), which puts new product failure at about 40% or less. |
| "64% of features are rarely or never used" | A 2002 Standish Group talk about [four internal business applications](https://www.mountaingoatsoftware.com/blog/are-64-of-features-really-rarely-or-never-used) at four companies. No commercial products were included. | Kohavi's experiment data: only 8% to 33% of tested ideas improve their target metric. |
| "A fully optimized product manager could increase profits by 34.2%" | A 2015 survey by a product management training firm, where respondents guessed how much profit would rise. It's an estimate, not a measured result. | Your own before-and-after metrics for a specific product change. |
| "Only 5% of product managers know how to code" | A 2019 vendor blog list with no survey or source behind it. | Product Focus 2026: 53% of product people come from a technical background. |
| "Product managers spend 52% of their time firefighting" | Real, from the Product Focus survey of January 2021, and self-reported. | The 2026 edition: 60% are often pulled into unplanned work. |
| "30,000 new products launch every year" | An unsourced estimate about consumer packaged goods in the same 2011 Christensen article. | For software, app-store data such as Sensor Tower's count of nearly 150 billion downloads in 2025. |
| "Only 40% of products make it to market" | No traceable study. It's credited to several groups, but none has a primary document behind it. | Castellion and Markham's review of new product failure rates. |
| "95% of generative AI pilots fail" | An [MIT NANDA report](https://mlq.ai/media/quarterly_decks/v0.1_State_of_AI_in_Business_2025_Report.pdf) from 2025, based on 52 interviews and 153 conference surveys. It found 5% of custom enterprise AI tools reached production with measurable P&L impact, and calls its figures directional. | McKinsey 2025: 23% of organizations are scaling agentic AI, and 39% see any EBIT impact. |
| "60% of PMs have basic analytics skills" | Real, but from [McKinsey in 2018](https://www.mckinsey.com/industries/technology-media-and-telecommunications/our-insights/the-product-management-talent-dilemma), and limited to PMs at leading software companies. | Cite it with its year, or use Product Focus 2026: 69% of product people use AI frequently. |

The pattern is familiar. Estimates turn into facts, small samples become industry rules, and corrections never catch up with the original claim. When a number is going into a board deck or a business case, check who measured it, when, and with how many people.

## How we chose these product management 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 sample size next to the figure where it matters. Many product management surveys reach only a few hundred people, so the sample tells you how much weight a number can carry. Third-party figures come from reports published in 2025 or 2026, unless an older study is still the standard reference and nothing newer exists. In that case we name its year.

UXCam's own figures come from two datasets. The State of AI in Product Management 2026 is a survey of 157 product professionals in 34 countries, published in June 2026. The Mobile UX Benchmarks 2025 report uses anonymized, aggregated data from 1.62 billion sessions in 1,538 apps that run the UXCam SDK. 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 your own product

Industry statistics show where the bar sits. Your own product data shows what to fix next, and it's the evidence that holds up when opinion and data disagree. The metrics behind most numbers on this page are measurable in any app or website: funnel conversion, retention, feature adoption, session length and friction signals such as rage taps.

UXCam captures those signals automatically in mobile apps and on websites, without manual event tagging. [Funnels](https://uxcam.com/product-analytics/funnels/) show where users drop off, [segmentation](https://uxcam.com/product-analytics/segmentation/) shows who they are, and [session replay](https://uxcam.com/experience-analytics/session-replay/) shows why they left. Tara, UXCam's [AI analyst](https://uxcam.com/ai/), watches sessions for you and explains in plain language which issues cost the most users. For the metrics most product teams track, see our guide to [product management KPIs](https://uxcam.com/blog/product-management-kpis/).

Product management statistics FAQ

### What are the most important product management statistics for 2026?

The most cited figures cover hiring, pay and how decisions get made. There were 7,300+ open product manager roles at tech companies in March 2026, the most since 2022 ([Lenny's Newsletter](https://www.lennysnewsletter.com/p/state-of-the-product-job-market-in-ee9)). The median US product manager earns $233,000 in total pay ([Levels.fyi](https://www.levels.fyi/t/product-manager)). And 77% of product teams say executive gut instinct overrides data at least sometimes ([Atlassian](https://dam-cdn.atl.orangelogic.com/CDNLink/AT12OX96.pdf)). UXCam's own 2026 survey adds that 99% of product professionals use AI, but only 6% of teams have set AI goals.

### How much do product managers earn in 2026?

In the US, the median product manager earns $233,000 in total pay, including a $187,000 base salary, based on 4,921 salaries reported to [Levels.fyi](https://www.levels.fyi/t/product-manager). The Bay Area median is $316,800, and the UK median is £104,837. Surveys that include smaller companies show lower figures: the [Product-Led Alliance](https://www.productledalliance.com/product-management-salary-report-2025/) puts the global median at $123,000.

### Is the product manager job market growing?

Yes. Open product manager roles at tech companies reached 7,300+ in March 2026, 75% above the early-2023 low and the most since 2022, according to TrueUp data analyzed by [Lenny's Newsletter](https://www.lennysnewsletter.com/p/state-of-the-product-job-market-in-ee9). Open PM roles now outnumber open designer roles 1.27 to 1, and AI product management is one of the fastest-growing areas.

### How many engineers are there per product manager?

It depends on company size. [Pave's](https://www.pave.com/blog-posts/research-development-org-structure-benchmarks) data from 4,195 companies shows about 5 engineers per product manager at companies with up to 100 employees, widening to about 11 at companies with 3,000 or more. Product managers make up 13% of R&D headcount at small companies and 7% at large ones.

### How do product managers use AI in 2026?

Almost all of them use it, but few have turned it into measurable goals. In UXCam's [State of AI in Product Management 2026](https://uxcam.com/resources/ebooks/ai-in-product-management-2026/), 99% of 157 product professionals use AI. Of those, 41% say it's core to team workflows, and 70% spend five or more hours a week with AI tools. Claude (80%), ChatGPT (68%) and Gemini (51%) are the most used tools. The biggest blocker is budget or access (30%), and only 6% of teams have set explicit AI OKRs.

### Is it true that 95% of new products fail?

No study supports it. The figure is credited to Clayton Christensen, but it came from a 2011 Harvard Business School article that has since been corrected. A 2013 review of the research by Castellion and Markham puts new product failure at about 40% or less. For software, a better guide is experiment data: only 8% to 33% of tested ideas improve their target metric, even at companies like Microsoft and Netflix.

### Where does UXCam's product management data come from?

The State of AI in Product Management 2026 is a survey of 157 product professionals in 34 countries, published in June 2026. 62% work in the product management track, and company sizes are evenly spread. The engagement benchmarks come from UXCam's Mobile UX Benchmarks 2025 report, which uses anonymized, aggregated data from 1.62 billion sessions in 1,538 apps.