Core Web Vitals in 2026: LCP, INP and CLS explained

What the Core Web Vitals are

The Core Web Vitals are Google's chosen set of measurements for the part of page experience that can be measured in numbers. There are three, one for each thing a visitor notices:

• LCP, Largest Contentful Paint. How long the main content takes to appear. It is the loading metric. Read the LCP guide.
• INP, Interaction to Next Paint. How quickly the page reacts when someone taps, clicks or types. It is the responsiveness metric. Read the INP guide.
• CLS, Cumulative Layout Shift. How much the page jumps around while it loads and while it is used. It is the visual stability metric. Read the CLS guide.

Google's marks

Each metric has a good mark and a poor mark. Anything in between needs improvement.

How a page passes

The rules are specific, and most of the confusion about "why does my score not match" comes from missing one of them.

• The data comes from real visits. It is collected from Chrome users and published in the Chrome UX Report (CrUX; see how CrUX works). A lab test such as Lighthouse does not decide whether you pass.
• The mark must be met at the 75th percentile. Three out of four visits must be at least as good as the mark. A page whose typical visit is fast but whose slowest quarter is slow is judged by that slow quarter.
• Mobile and desktop are separate. Each has its own result.
• All three must be good. One metric in the "needs improvement" range means the page does not pass.
• The window is 28 days. The numbers are a rolling 28 days of visits, updated daily, so a fix is only fully visible a month later.
• A page needs enough visits. CrUX only includes pages and sites that can be found by search engines (a normal status code, not marked noindex) and that have a minimum number of visitors. When a page does not have enough data of its own, PageSpeed Insights shows data for the whole site and says so, and when even that is missing, it shows no field data at all.

One more limit is worth knowing. CrUX includes visits by Chrome users on desktop and on Android who have turned on usage statistics and history sync. Chrome on iOS, other Chromium browsers such as Edge, and Android apps that use a WebView are not included. The numbers describe a large sample of your visitors, not all of them.

What changed: INP replaced FID

On March 12, 2024, INP replaced First Input Delay (FID) as the Core Web Vital for responsiveness. FID only measured the wait before the browser started handling a visitor's first interaction. INP looks at the whole response to every interaction during a visit, which is a much better picture of how a page feels. FID is no longer a Core Web Vital, and an article that still lists it as one is out of date. The INP guide explains the difference.

Do they affect search rankings?

Google says its ranking systems use signals that line up with page experience, and the Core Web Vitals are among them. It also says there is no single page experience signal, that a good report does not guarantee a higher position, and that it tries to show the most relevant content even when the page experience is poor. A fast page cannot make up for content that does not answer the search. When many pages answer it equally well, a better experience can help.

The more certain reason to care is the visitor. Slow and jumpy pages are harder to use, and people leave them.

Lab data and field data

The Core Web Vitals pass or fail on field data, which comes from real visits. Lab data comes from a test such as Lighthouse, which loads a page once under set conditions. A page can score 100 in the lab and still fail in the field, because real visitors have slower phones, worse connections and different habits than a test. Lab tests are how you find causes. Field data is how you know whether you passed. How to measure Core Web Vitals covers both.

Where most sites stand

In the HTTP Archive's 2025 Web Almanac, about 48% of mobile websites and 56% of desktop websites had good Core Web Vitals. The share has risen every year since 2021, when it was 32% on mobile and 41% on desktop. Of the three, loading is the hardest to pass: 62% of mobile pages had a good LCP, against 81% for CLS and 77% of sites for INP.

Where to start

1. Look at the real-user data for mobile in PageSpeed Insights.
2. Find which of the three fails. If all pass, you are done: test again every few months.
3. Read the guide for that metric, and fix the biggest cause first.
4. Test again after each change, and check the real-user data after a few weeks.

Check your own page

Run the free audit to see where each of the three stands for your page on mobile and on desktop, and what the report suggests first. Remember that the lab numbers move the day you fix something, while the real-user numbers follow over the next four weeks.

Sources

• Web Vitals on web.dev: the program, the three metrics and the 75th percentile rule.
• Chrome UX Report methodology: which pages, sites and visitors are included, and how often the data updates.
• Understanding Core Web Vitals and Google search results in Google Search Central: how they relate to ranking.
• Performance chapter of the 2025 Web Almanac: the pass rates above.