INP: why your site feels slow even when it loads fast
Interaction to Next Paint (INP) measures how long a page takes to visibly respond after a tap, click or key press, across the whole visit. It replaced First Input Delay as a Core Web Vital in March 2024. A good score is under 200 milliseconds. Sites that load quickly still fail INP when heavy JavaScript keeps the browser busy: tag managers loaded with scripts, chat widgets, page builders, consent tools and third-party embeds. The fix is usually removing or deferring scripts rather than buying faster hosting.
A client once described their site this way: “It loads fast, but it feels heavy.” That is a precise description of a bad INP score.
Three numbers, one of them new
Google’s Core Web Vitals reduce page experience to three measurements:
| Metric | What it measures | Good |
|---|---|---|
| LCP | when the main content appears | under 2.5 s |
| INP | how fast the page responds to interaction | under 200 ms |
| CLS | how much the layout shifts while loading | under 0.1 |
INP replaced First Input Delay in March 2024. The old metric only measured the delay before the browser started handling the first interaction. INP measures the full time from a tap until the screen visibly updates, and it does so across every interaction during the visit, reporting one of the worst.
That change exposed a lot of sites that had looked healthy.
Why fast pages fail it
Loading and responding are different jobs. A page can deliver its content quickly and then spend seconds running JavaScript in the background. While the browser’s main thread is busy, it can’t respond to the visitor. Tap the menu, nothing happens, then it opens late.
The usual culprits on marketing sites and stores:
- Tag managers carrying years of scripts. Pixels from campaigns that ended, three analytics tools, heatmaps nobody looks at.
- Chat and support widgets loading large bundles on every page.
- Page builders and heavy themes that ship code for every possible layout.
- Consent tools that run complex logic on each interaction.
- Third-party embeds: reviews, social feeds, video players, maps.
- Product filters and carousels that re-render too much on each click.
How to check
- Open PageSpeed Insights and test your key templates: home, a service or product page, a category page, the contact form.
- Look at the top section, real user data. That is what Google uses. INP appears there.
- In Search Console, the Core Web Vitals report groups failing URLs by template.
- To find the cause, open Chrome DevTools, record a performance trace while you tap the slow element, and look for long tasks.
How to fix it
Audit the tag manager. Remove every tag you can’t name a current owner for. This alone often moves INP more than any code change.
Defer what doesn’t need to run immediately. Chat widgets can load after the first interaction or after a few seconds. Video players can load on click with a poster image in place.
Replace heavy components. A carousel built on a large library can usually be done with a fraction of the code.
Break up long tasks. Work that runs on each click can be split so the browser updates the screen first.
Question each plugin. On WordPress, every active plugin can add scripts to every page.
Our baseline
On the sites we build, the target is a mobile Lighthouse score of 95 or higher and the main content painting in under 1.5 seconds. We load video only on click, and the site you are reading ships almost no JavaScript by default. You can test it with the same tools.
If your site loads quickly and still doesn’t convert, speed may not be the problem at all. Try a 30-minute conversion audit first.
Common questions
Is INP a ranking factor?
Core Web Vitals, including INP, are part of Google's page experience signals. They are one input among many and rarely decisive on their own. The larger effect is on people: a sluggish site loses visitors and conversions, especially on mobile and especially from paid traffic.
Why does my Lighthouse score look fine while INP fails?
A lab test like Lighthouse loads the page once and doesn't interact with it the way people do. INP is measured from real users tapping real menus and filters. Check the field data at the top of PageSpeed Insights or in Search Console's Core Web Vitals report.
Will a faster host fix INP?
Rarely. INP is mostly about work done in the visitor's browser after the page arrives. Server speed affects how quickly the page loads; it does little for how quickly a menu opens.
