Core Web Vitals 2026: What Has Changed and How to Improve Your Scores

SEO & Marketing 7 min read Sidora AG
All articles
Core Web Vitals 2026: What Has Changed and How to Improve Your Scores

Core Web Vitals measure how quickly a page loads, how smoothly it responds to input and how stable its layout remains. Google uses them in its ranking systems, so they matter for every website that wants to be found through search. The most important change is not brand new: Interaction to Next Paint (INP) replaced First Input Delay (FID) in March 2024. What is genuinely new since late 2025 is that LCP and INP can now be measured in all major browsers. This article explains what applies, what has actually changed and where to start.

What Has Changed: INP Instead of FID

On 12 March 2024, Interaction to Next Paint (INP) replaced First Input Delay (FID) as a Core Web Vital. Google Search Console removed FID from its reports on the same day, while other tools such as PageSpeed Insights and CrUX were given a six-month transition period.

The difference is fundamental. FID only measured the input delay of the first interaction on a page. INP, by contrast, observes all clicks, taps and key presses throughout the visit and reports the slowest interaction, ignoring outliers. It measures not only the wait before the browser responds, but also the processing time and the time until the next frame appears on screen. This is stricter but more realistic: a page that responds quickly to the first click but stalls when a filter or menu opens is now exposed. According to web.dev, far fewer sites reach a good INP score than passed the old FID threshold.

The Three Metrics and How Google Measures Them

The three Core Web Vitals and their thresholds for a "good" rating are:

  • Largest Contentful Paint (LCP): 2.5 seconds or less until the largest visible image or text element is rendered.

  • Interaction to Next Paint (INP): 200 milliseconds or less. Above 500 milliseconds, responsiveness is rated poor.

  • Cumulative Layout Shift (CLS): 0.1 or less. The score describes how much visible content shifts unexpectedly.

What counts is not a single test run but the 75th percentile of real page loads, segmented into mobile and desktop. In other words, three out of four visits must meet the threshold. This field data comes from the Chrome User Experience Report (CrUX), which also powers PageSpeed Insights and the Core Web Vitals report in Search Console. Google states that stable Core Web Vitals change no more than once a year. Since INP was introduced, the metrics and thresholds have remained unchanged.

What Is Actually New in 2025 and 2026

Since 12 December 2025, LCP and INP have been "Baseline Newly available": with Safari 26.2, the latest versions of all major browsers include the interfaces needed to measure both metrics. This was made possible by the cross-browser Interop 2025 project. For you, it means that your own monitoring can now show how quickly your site loads and responds in Safari and Firefox as well.

Three limitations are worth knowing. Firefox and Safari stop measuring slightly earlier than Chrome, so values can differ a little. CrUX, and with it Search Console and PageSpeed Insights, still relies only on Chrome users. Chrome on the iPhone is excluded because it is based on WebKit. And CLS can for now only be measured in Chromium-based browsers; it has been proposed for Interop 2026.

What Role the Metrics Play in Rankings

Google confirms that Core Web Vitals are used by its ranking systems, but also stresses that there is no single "page experience" signal and that Search always aims to show the most relevant content, even if the page experience is sub-par. Good scores will not turn a weak page into a top result. Where several pages are similarly relevant, however, Google says a great page experience can contribute to success in Search. The bigger benefit lies with your visitors anyway: a fast, stable page is more likely to be read and used.

The Most Common Problem Causes

Poor LCP: According to the 2024 Web Almanac, as cited by web.dev, the LCP element is an image on 73% of mobile pages. For 35% of those, the image URL could not be discovered in the initial HTML document, for example because a JavaScript lazy loader only inserts it via data-src. Another frequent mistake is the loading="lazy" attribute on the hero image: lazy loading belongs on images below the fold, never on the LCP image. What helps: deliver the LCP image as a regular img element in the HTML, prioritise it with fetchpriority="high", and serve it at a suitable size in modern formats such as WebP or AVIF. If the server response itself is slow (Time to First Byte), caching and a server or CDN close to your users will help. For a predominantly Swiss audience, a server in Switzerland is therefore a sensible technical choice too.

Poor INP: Usually JavaScript is blocking the main thread. If someone clicks while the browser is still working through a long script, the response has to wait. What helps: break long tasks into smaller pieces and give the browser time for input in between (for example with scheduler.yield()), remove unnecessary JavaScript, split code and clean up old tracking tags. Very large DOM structures and forced layout calculations also slow down rendering after an interaction.

Poor CLS: According to the 2024 Web Almanac, 66% of pages have at least one image without explicit dimensions. What helps: set width and height on images and videos, reserve space for late-loading content such as banners, maps or cookie notices with aspect-ratio or min-height, animate with transform instead of top, left or margin, and make sure the page is eligible for the browser's back/forward cache.

Three Quick Wins to Get Started

  1. Look at field data first: Check key pages with PageSpeed Insights and look at the field data at the top before the Lighthouse lab results. Field data shows what your visitors experience; lab data helps you find the causes.

  2. Fix the LCP image: Because an image is the LCP element on most mobile pages, this intervention often pays off most: remove lazy loading from the hero image, set fetchpriority="high", compress the image and deliver it as WebP or AVIF.

  3. Remove dead weight: Check which plugins, widgets and tags load their own CSS and JavaScript on every page, and remove whatever is no longer needed. This is particularly worthwhile for WordPress installations that have grown over the years. For custom functionality, clean WordPress development loads features only where they are needed.

Ongoing Monitoring, Not a One-Off Project

Core Web Vitals are not a one-off project. Every new plugin, tracking script and campaign landing page can change the scores. The following routine has proven useful:

  • Review the Core Web Vitals report in Search Console every month and act before groups of pages slip from the middle category into the poor range.

  • After every major release, re-measure the most important page types (home page, service page, blog article, form) in the lab.

  • Only add new third-party scripts with a clear reason, and clear them out regularly.

In the long run, architecture decides. Our article on server-first architecture explains how server-side rendering and less client-side JavaScript improve the scores structurally. For a relaunch, performance budgets should be part of website development from the start, not an afterthought after go-live.

Sidora AG analyses Core Web Vitals as part of its technical SEO optimisation and implements the measures directly in the CMS and the code.

Sources

Share this article

About Sidora AG

This article was written by Sidora AG, an owner-led digital agency from Winterthur. Since 2003 we have been designing and developing websites, online shops and custom solutions for companies across Switzerland — clearly explained and close to practice. Learn more about us on our agency page, and get an overview of our work under services.