Core Web Vitals Explained Simply: What Google Measures and How to Improve It

Core Web Vitals are three metrics Google uses to judge how your website actually feels to real visitors: how fast it loads, how quickly it responds to clicks, and whether the layout jumps around while it loads. These scores feed into your ranking, and you can verify them with any browser tool. In this article we explain exactly what gets measured, which thresholds apply, and how to tackle each of the three scores in concrete terms.

The three metrics at a glance

Since the switch to INP (March 2024), Google measures three metrics. Each one represents a different dimension of the user experience:

• LCP (Largest Contentful Paint) – How long does it take for the largest visible element to load? Usually that's a hero image or a large headline. Good: under 2.5 seconds.
• INP (Interaction to Next Paint) – How quickly does the page respond when someone clicks, types or opens a menu? It measures the delay until a visible response appears. Good: under 200 milliseconds.
• CLS (Cumulative Layout Shift) – How much does the layout move around while the page loads? This happens when images load in without reserved space and the text suddenly jumps down. Good: under 0.1.

One thing matters here: the so-called 75th-percentile rule. Google doesn't judge the average, but the score that 75 percent of your visitors actually experience. A good average is no help to you if a quarter of your users see a sluggish page.

How to measure your scores

You don't need a paid tool to get started. These three sources are enough for an honest diagnosis:

• PageSpeed Insights (pagespeed.web.dev) – Enter your URL and you get both lab data and real field data from Chrome users, provided you have enough traffic.
• Google Search Console – Under the Core Web Vitals section you can see which URLs are rated good, in need of improvement or poor, grouped by issue.
• Lighthouse in the Chrome developer tools – for quick local tests while you work.

An honest note: lab data and field data often diverge. Field data from real visitors is what Google uses for ranking. If you have only a little traffic, it can take a while before reliable field data is available.

Improving LCP: load time of your main content

The most common LCP bottleneck is an oversized, unoptimised image in the visible area. Here's how to tackle it:

• Serve images in a modern format such as WebP or AVIF and scale them to the size they're actually displayed at.
• Load your most important image with priority via preload, instead of letting it get lost in the queue.
• Check your server response time – slow hosting or an overloaded CMS holds everything up. Caching and a CDN make a noticeable difference.
• Reduce render-blocking CSS and JavaScript so the browser can display the content sooner.

Improving INP: responsiveness

INP almost always suffers from too much JavaScript. When the browser's main thread is busy running scripts, it can't respond to clicks. What helps in practice:

• Remove unnecessary scripts and heavy tracking or marketing tools, or load them with a delay.
• Break long JavaScript tasks into smaller chunks so the browser can respond to input in between.
• Take a critical look at third-party widgets – chat boxes, cookie banners and embeds are frequent culprits.

A lean, cleanly built page rarely has problems here. INP only becomes a construction site when plugins and tools have been stacked on top of each other over the years.

Improving CLS: layout stability

CLS is often the easiest score to fix, because the causes are clear:

• Always specify width and height for images and videos so the browser reserves the space in advance.
• Give ad slots and dynamically loaded content a fixed placeholder.
• Embed fonts so the text doesn't jump from a fallback font to the final one as it loads.

Do you really need a perfect score everywhere?

The honest answer: no. Core Web Vitals are one ranking factor among many, and when in doubt, good content beats the last millimetre of speed. It's not worth spending weeks optimising from 95 to 99 points. What is worth it: getting out of the red and into the green, that is, hitting the thresholds of 2.5 s, 200 ms and 0.1. Anything beyond that is fine-tuning that only really pays off for very large sites.

From our own experience: we run seven of our own brands in production, from a product portal with over 177,000 entries to several SaaS dashboards. Most Vitals problems don't come from missing tricks, but from too much baggage – redundant scripts, huge images, a sluggish CMS. If you build lean from the start, there's barely anything to repair here. If, on the other hand, your site is already live and sitting in the red, a targeted analysis with PageSpeed Insights and Search Console is the right first step.