INP in 2025: 77% of mobile sites pass, but the top 1,000 lag at 63%

Interaction to Next Paint is the Core Web Vital that measures how quickly a page responds to clicks and taps. The 2025 Web Almanac Performance chapter, built on Chrome UX Report field data, shows mobile INP improving overall while the busiest sites and inner pages still struggle.

How many sites have good INP?#

For an origin to be rated good, at least 75% of sessions must record an INP of 200 milliseconds or less. On that bar, 77% of websites passed on mobile in 2025, up from 74% in 2024. On desktop, 97% had good INP, 2% needed improvement and less than 1% were poor. On phones, 21% needed improvement and 3% were poor. The mobile-desktop gap narrowed to 20 percentage points.

Do popular sites pass INP?#

No. Segmenting mobile origins by rank, the Almanac found good INP at 63% for the top 1,000 websites, 56% for the top 10,000, 56% for the top 100,000 and 64% for the top 1,000,000. The figure climbs to 76% across the top 10,000,000 and 77% for all websites.

The top 1,000 improved fastest, from 53% to 63%. They were 21 points below the average in 2024 and 14 points below in 2025. The Almanac attributes the difficulty to heavier JavaScript, richer interactive features and third-party integrations on high-traffic sites. It calls the gains likely the result of manual optimization of complex JavaScript.

Home pages or inner pages?#

Page-level CrUX data shows a split that was absent in 2024, when mobile home and secondary pages scored 73% and 72%. In 2025, mobile home pages reached 80% good INP, a 7 point gain, while secondary pages fell to 69%. On desktop the two were close, at 97% and 95%. The Almanac points to filters, carousels, form validation and late-loading third-party widgets as likely causes on deeper pages, so audit templates beyond the home page (see technical SEO).

How is INP scored?#

The definition shapes how the numbers above should be read. Per web.dev, INP observes three kinds of input: clicking with a mouse, tapping on a touchscreen and pressing a key on a physical or onscreen keyboard. Hovering, zooming and scrolling are not counted. A page’s INP is close to its slowest interaction, but on pages with many interactions one worst interaction is ignored for every 50, so a single random hiccup does not set the score. The thresholds are 200 milliseconds or less for good, up to 500 milliseconds for needs improvement and above 500 for poor, assessed at the 75th percentile. Our Core Web Vitals checker classifies any INP value against those bands.

Who wrote the chapter?#

The 2025 Performance chapter was written by Himanshu Jariyal, Prathamesh Rasam, Humaira and Aaron T. Grogg, analyzed by Tanner Hodges and edited by Barry Pollard. Its origin-level figures rest on CrUX field data, so the “all websites” line means all origins in that dataset, not every site on the web.

Does TBT still predict INP?#

Less than before. Total Blocking Time is a lab proxy for responsiveness, yet the Almanac reports it rose 58% while field INP improved. Median mobile TBT was 1,916 milliseconds against 92 milliseconds on desktop. The Almanac suggests sites may be tuning the interactions that dominate INP while background JavaScript work still blocks the main thread. Judge INP from field data.

What should a site owner check?#

Start in Search Console, which Google points to for this: its Core Web Vitals report “shows how your pages perform.” Google says good Core Web Vitals, along with other page experience aspects, align with what its core ranking systems “seek to reward,” which is a reason to fix poor INP, not a promise of higher rankings. Given the Almanac’s page-level split, look at inner templates such as category and product pages, not just the home page. web.dev advises using field data to find slow interactions first, then reproducing them in the lab by interacting with the page as it loads, when the main thread is often busiest. For how INP varies by platform, see Core Web Vitals by CMS.