On this page
Test with field data first
The Core Web Vitals report in Search Console groups your URLs into good, needs improvement, and poor, for mobile and desktop, using real visitor data from Chrome. PageSpeed Insights shows the same field data for any URL that has enough traffic, alongside a lab test run by Lighthouse.
Use both for different jobs. Field data tells you whether real people have a problem, and it is what Google evaluates. Lab data tells you why, because it is repeatable and comes with a list of opportunities. Low-traffic pages may have no field data at all, in which case the lab result is the best you have. Judge sites by the 75th percentile, not the average.
CrUX, the Chrome User Experience Report, is the dataset behind the field numbers in both Search Console and PageSpeed Insights. It only includes pages with enough visits from Chrome users, which is why small pages sometimes show no field data and fall back to a lab test or origin-level numbers.
Fix LCP: make the main content arrive sooner
LCP is the time until the largest element in the viewport, usually a hero image or a heading block, renders. The target is 2.5 seconds or less. Find the LCP element in PageSpeed Insights, then work out where the time goes: slow server response, render-blocking CSS or scripts, or a heavy image.
The usual fixes: serve a smaller, modern-format image at the right dimensions, and do not lazy-load the hero image. Preload it, and use a CDN and caching to cut server response time. Remove or defer scripts that block rendering. Image sizing and formats are covered in image SEO.
Server response time sets the floor for LCP. If the first byte takes over a second to arrive, no image tweak will rescue the score. Look at hosting, caching, and the work the server does on each request, and consider serving pages from a cache or a CDN edge close to the visitor.
Want the articles written for you?
Researches keywords and what competitors rank for. Publishes to WordPress, Webflow, Shopify or Framer once you approve.
Fix INP: make taps and clicks respond
INP replaced First Input Delay as a Core Web Vital in March 2024. It measures how long the page takes to visually respond after interactions, and reports one of the slowest across the visit. The target is 200 milliseconds or less.
Slow INP is almost always too much JavaScript on the main thread. Break up long tasks, remove or delay third-party scripts such as chat widgets and tag-manager bloat, and avoid heavy work in click handlers. Test on a mid-range phone, not your laptop, since that is where the problem shows up.
Fix CLS: stop things jumping around
CLS adds up unexpected layout shifts during the page’s life. The target is 0.1 or less. The classic causes are images and videos without set dimensions, ads and embeds that load late and push content, and web fonts that swap in with a different size.
Set width and height, or an aspect ratio, on every image, video, and embed. Reserve space for ads and banners. Use font-display settings and preload key fonts. Do not inject content above existing content after load, except in response to a user action.
Check the shifts on a real phone as well as in a tool. Banners that appear after load, cookie notices that push content down, and late-loading fonts are easy to miss on a fast desktop connection and obvious on a slow phone. Fix the ones that shift content a visitor is already reading first.
The tools, and which one to trust
Search Console’s Core Web Vitals report is the site-wide view of real-user data, and it tells you which groups of URLs to fix. PageSpeed Insights shows the field data and a Lighthouse lab test for a single URL. Chrome DevTools and Lighthouse are what you use while fixing, because they give instant, repeatable results and point to specific elements.
Trust field data to decide whether there is a problem and lab data to find the cause. If the lab score is low but field data is good, real users are fine and you can deprioritize it. If field data is poor and the lab score looks fine, test on a slower phone and network, because the lab run may be too kind.
Real-user monitoring from your own analytics adds detail by page, device, and country, and shows regressions the day they ship rather than 28 days later. It is worth adding if speed matters to your revenue.
How much does it matter for rankings
Core Web Vitals are part of Google’s page experience signals, but relevance and content quality come first. Google has said that great page experience does not override a page that is less relevant. Treat vitals as a tiebreaker and a conversion lever: faster pages lose fewer visitors.
So prioritize: fix the poor URLs the report flags, then go back to content. The technical SEO checklist puts speed in context, and landing page CRO covers the conversion side. The part you can hand off: Sprites can run page audits and landing page checks on a schedule and flag regressions for you to approve fixes.
Questions
What are the Core Web Vitals?
LCP, INP, and CLS. LCP measures how quickly the main content loads, INP how quickly the page responds to interaction, and CLS how stable the layout is while loading. Google measures them from real Chrome users.
What is a good CLS score?
Under 0.1 at the 75th percentile of page visits is good. Between 0.1 and 0.25 needs improvement, and above 0.25 is poor.
Are Core Web Vitals still relevant?
Yes. They remain Google’s published page experience metrics and appear in Search Console. They are one signal among many, so great relevance still beats fast but unhelpful pages.
How long until my Core Web Vitals scores improve?
The field data covers a rolling 28-day window, so a fix shows up gradually and can take about a month to fully appear. Use a lab test to confirm the fix worked right away.
Why do PageSpeed Insights scores differ from Search Console?
PageSpeed Insights shows field data for the single URL and a lab run, while Search Console groups similar URLs and uses only field data. The lab score can differ because it simulates one device and connection.