Core Web Vitals: LCP, INP, CLS

Core Web Vitals are three measures of the real user experience on a page, defined by Google: LCP (how fast the main content appears), INP (how fast the page responds to actions) and CLS (elements jumping around while loading). Good values: LCP under 2.5 seconds, INP under 200 milliseconds, CLS under 0.1.

In plain words

• LCP (Largest Contentful Paint): how long it takes for the largest element on the screen to appear. On a product page this is usually the main photo. Google's benchmark: under 2.5 s from the start of loading.
• INP (Interaction to Next Paint): how fast the page responds to a click or tap: add to cart, choose a size, open a filter. Benchmark: under 200 ms.
• CLS (Cumulative Layout Shift): whether the page "jumps" while loading. The classic problem: the "Buy" button shifts at the moment you press it, because a banner loaded above it. Benchmark: under 0.1.

Source of the thresholds: Google, Core Web Vitals and web.dev. A page is considered good if all three measures meet the norm at the 75th percentile of loads, separately for mobile and desktop (web.dev). In other words, the norm has to be met not by the average visit but by three out of four real visits.

INP became a stable metric in 2024 and replaced FID. The web.dev page with these thresholds was last updated on 31 October 2024; stable metrics, it says, change no more than once a year.

How it's measured: field and lab data

Field data is what matters: measurements on visitors' real devices, such as the Core Web Vitals report in Search Console, PageSpeed Insights and the Chrome User Experience Report. Lab data (Lighthouse) is useful for finding the cause and checking a fix, but it doesn't replace field data; INP can't be measured in a lab because there are no user actions, so Lighthouse uses a substitute, TBT (web.dev).

For a Ukrainian store this means: look at the Search Console report for mobile specifically, because most traffic comes from smartphones (see Mobile UX).

How the web as a whole is doing

According to the HTTP Archive Web Almanac 2025 (CrUX data for July 2025): 48% of mobile sites and 56% of desktop sites met the norm on all three measures (in 2024 it was 44% on mobile). On mobile, separately: LCP passes on 62% of sites, INP on 77%, CLS on 81%. The weakest spot is LCP. The main photo was the LCP element on 76% of mobile pages. On 16–17% of pages the main photo is lazy-loaded, even though that slows its appearance; `fetchpriority="high"` was set on only 17% of such mobile images and 16% of desktop ones (Web Almanac 2025).

Practical takeaway for a store: don't lazy-load the first product photo; the rest of the photos below the first screen can be.

Does it affect rankings

Google writes: Core Web Vitals are used by ranking systems, and Google recommends aiming for good values. At the same time, good scores don't guarantee high rankings; relevance matters more ("Google always seeks to show the most relevant content, even if the page experience is subpar"); chasing a perfect result just for SEO may not be worth it (Google, page experience). Speed helps more when many equally useful pages compete for one query (same source).

Does it affect sales

There is less evidence here than agencies promise. The best-known example on web.dev is Rakuten 24. On one page, over a month, an A/B test showed a 53.37% increase in revenue per visitor and 33.13% in conversion when it was optimized for Core Web Vitals. Limitations of the case itself: one page, one month, written by the company's own developers, no absolute values and no statistical significance; the case doesn't state a publication date (web.dev, Rakuten 24). It's an example of speed possibly affecting sales, not a guaranteed figure. There are no Ukrainian studies with open numbers.

Practical rule: measure on your own store. Take pages with poor scores, fix them and compare conversion in an A/B test (A/B test). What Rakuten changed in the case is useful as a list of ideas: removed render-blocking scripts, compressed and right-sized images, deferred loading of what is below the first screen, reserved space for images with `aspect-ratio` and `min-height` so nothing shifts.

Typical causes of poor scores in a store

• LCP: a heavy, uncompressed main photo; it loads lazily; a slow server; blocking scripts.
• INP: heavy JavaScript (chat widgets, analytics, sliders) that is busy when someone presses a button.
• CLS: images without set dimensions; banners and widgets that load in from above; fonts that change size.

Common mistakes

• Looking only at the lab test score and not at real users.
• Chasing a score of 100 and ignoring the three threshold values.
• Fixing the homepage and leaving product and category pages. In Web Almanac 2025 the homepages were actually worse than inner pages: 45% of mobile homepages met the norm versus 56% of inner pages.
• Expecting a jump in rankings: Google says outright that relevance matters more.

FAQ

What are Core Web Vitals? Three measures of page speed and stability: LCP, INP, CLS.

What are the Core Web Vitals thresholds? LCP under 2.5 s, INP under 200 ms, CLS under 0.1 for 75% of visits.

Do Core Web Vitals affect rankings? Yes, Google uses them, but they don't guarantee a top spot and rank below relevance.

Do they affect conversion? There are individual cases with growth (Rakuten 24), but with limitations. For your store, check with a test.

What is INP and what did it replace FID with? INP measures the response time to all actions on a page, not just the first one. It replaced FID in 2024.

Related terms

• Technical site audit
• SEO audit of an online store
• The mobile version of a store
• Conversion rate
• A/B test

Sources

• Google Search Central, Core Web Vitals — developers.google.com/search/docs/appearance/core-web-vitals
• Google Search Central, Understanding page experience — developers.google.com/search/docs/appearance/page-experience
• web.dev, Web Vitals — web.dev/articles/vitals
• HTTP Archive, Web Almanac 2025: Performance — almanac.httparchive.org/en/2025/performance
• web.dev, Rakuten 24 case study — web.dev/case-studies/rakuten