Core Web Vitals are three specific measurements: Largest Contentful Paint under 2.5 seconds, Interaction to Next Paint under 200 milliseconds, and Cumulative Layout Shift under 0.1. They act as a tie-breaker, not a gate: a highly relevant page with poor Vitals still outranks a fast page with weak content, but among similarly relevant pages, the faster one wins.
What the three metrics actually measure
Core Web Vitals are three specific measurements, not a vague notion of "fast." Largest Contentful Paint (LCP) times how long it takes the biggest visible element, usually a hero image or heading, to render. Interaction to Next Paint (INP) times how long the page takes to visibly respond after someone clicks, taps or types. Cumulative Layout Shift (CLS) scores how much content unexpectedly jumps around while a page loads, the classic case being a button that moves just as you go to tap it.
Each has a threshold Google labels "good": LCP under 2.5 seconds, INP under 200 milliseconds, CLS under 0.1. These are measured from real visitor traffic through the Chrome User Experience Report, not a single lab test, so they reflect what actually happens on the devices and networks your visitors use.
How much this actually affects rankings
Page experience is a tie-breaker, not a gate. Google has been explicit that a highly relevant page with poor Vitals will still outrank a fast page that answers the query poorly. Where Vitals matter is between pages that are already competing on similar relevance and content quality: among those, the one that loads and responds faster gets a genuine, measurable edge, and a consistently poor score can also affect how much of your site Google bothers to crawl.
This is why Vitals sit inside technical SEO rather than as their own discipline. They are one input alongside crawlability, indexation and structured data, not a separate project competing for budget against content or links.
The three fixes that actually move the numbers
- For LCP. Compress and correctly size hero images, serve them in a modern format, and preload the single most important image on the page instead of lazy-loading it.
- For INP. Break up long-running JavaScript tasks and defer anything not needed for the first interaction, since a busy main thread is what actually delays a response to a tap or click.
- For CLS. Reserve explicit width and height for every image and embed, and never inject a banner or ad above content that has already rendered.
For an e-commerce catalogue with hundreds of product images, the LCP fix alone is usually the highest-leverage change available, since a single unoptimised hero image on a template page repeats its cost across every product that uses it. See our e-commerce SEO checklist for how this ranks against the store's other priorities.
A realistic target, not a perfect score
Search Console labels a page "good" once 75% of its visits pass all three thresholds. Chasing a perfect 100 in a lab-testing tool past that point spends engineering time for a ranking effect that has already levelled off. Set the 75th-percentile pass rate as the goal, verify it against real visitor data, and stop there.
Third-party scripts: the silent Vitals killer
Analytics tags, chat widgets, ad scripts and tracking pixels are the most common cause of a good Vitals score turning bad after launch, precisely because none of them are code the site's own team wrote or is watching. Each one competes for the same main thread as everything else on the page, and a chat widget alone can add a meaningful delay to INP if it loads and initialises eagerly rather than waiting for the visitor to actually interact.
Audit every third-party script currently loading on the site, not just the ones added recently. Defer anything that is not needed for the first paint or the first interaction, and remove anything nobody can explain the purpose of. It is common for a site to be running two or three overlapping analytics tools left over from previous redesigns, each silently taxing every page load.
Where to check it
The Core Web Vitals report inside Google Search Console shows real visitor data grouped by URL pattern, which is the only view that matches what actually gets evaluated for ranking. Lab tools such as PageSpeed Insights are useful for diagnosing why a page is slow, but their score is a simulation, not the figure Google uses.
The honest summary
Core Web Vitals will not rescue a page that answers the wrong question, and a perfect score is not a growth strategy on its own. What they reliably do is remove a small but real disadvantage against competitors who are just as relevant as you are. Fix the three metrics once, confirm it in Search Console, and move the remaining budget to content and links, which is where most of the ranking movement actually comes from.