Open PageSpeed Insights, put your website in, and you get a screen of acronyms and a score in red. Most business owners close the tab.
It is worth understanding, because these three numbers are Google's attempt to measure something real: whether your site feels fast to an actual person on an actual phone. And that affects both your rankings and — usually more significantly — how many visitors stay long enough to contact you.
The Three Metrics
LCP — Largest Contentful Paint
What it measures: how long until the biggest thing on screen appears. Usually your hero image or headline.
Target: under 2.5 seconds.
In plain terms: how long the visitor stares at a blank or half-drawn page wondering if it is broken.
Usual culprit: a huge unoptimised hero image. A photographer sends you a 4MB JPEG, it goes straight on the homepage, and every mobile visitor downloads all 4MB before seeing anything.
INP — Interaction to Next Paint
What it measures: how long after a tap or click before the page visibly responds.
Target: under 200 milliseconds.
In plain terms: the visitor taps your menu, nothing happens, they tap again, then two menus open. That is bad INP.
Usual culprit: too much JavaScript. Chat widgets, booking plugins, analytics, cookie banners, social feeds — each one competes for the same single thread that also has to respond to taps.
INP replaced the older First Input Delay metric in 2024 and is significantly stricter, which is why some sites that used to pass now do not.
CLS — Cumulative Layout Shift
What it measures: how much the page jumps around while loading.
Target: under 0.1.
In plain terms: you go to tap a link, an advert loads above it, the page shifts, and you tap something else. Universally hated.
Usual culprit: images and embeds without width and height set, so the browser does not reserve space for them until they arrive.
Field data vs lab data
PageSpeed Insights shows two sets of numbers. The top section is real data from Chrome users over the last 28 days — that is what Google uses. The bottom is a simulation, useful for diagnosis but not the number that counts. If you have too little traffic for field data, you will only see the simulation.
The Fixes That Actually Move the Needle
Fix your images first
This resolves more Core Web Vitals problems than everything else combined. Three things:
- Serve modern formats — WebP or AVIF rather than JPEG and PNG. Typically 30-50% smaller at the same visual quality.
- Size them correctly. A 3000px-wide image displayed at 800px is wasting most of what it downloads.
- Set width and height attributes so the browser reserves the space. Fixes most CLS on its own.
Audit your plugins and scripts
Every third-party script has a cost. Live chat, heat mapping, three analytics tools, a social feed, a review widget — each adds JavaScript that delays interaction. Be honest about which ones earn their place. Most sites are carrying at least two that nobody has looked at in a year.
Fix your fonts
Custom web fonts often block text from rendering. Use font-display: swap so text appears immediately in a fallback and switches when the font arrives, and preconnect to the font host. Small change, noticeable LCP improvement.
Check your hosting
Cheap shared hosting can add several hundred milliseconds to the server response before anything else starts. If your time to first byte is over 600ms, no amount of front-end tuning will fix it. Better hosting costs a few pounds a month more.
How Much Does This Actually Matter?
Let me be straight: Core Web Vitals are a ranking factor, but a modest one. A slow page with genuinely the best answer will still outrank a fast page with a mediocre one. Anyone telling you a perfect score will transform your rankings is overselling.
Where it does matter:
- Between close competitors. When two sites are otherwise similar, speed is a tiebreaker.
- For conversions. This is the real commercial case. Visitors leave slow sites, and they leave mobile sites fastest of all. That effect is far larger than the ranking one.
- On mobile. Google indexes the mobile version of your site, and phones on patchy 4G are where a heavy site really suffers.
More on the commercial side of this in why website speed matters.
Where to Start
- Run your homepage and one service page through PageSpeed Insights. Look at the field data.
- If LCP is the problem, it is almost certainly your hero image.
- If CLS is the problem, it is almost certainly missing image dimensions.
- If INP is the problem, it is almost certainly a third-party script you could live without.
Those three checks resolve the majority of small business Core Web Vitals failures, and none of them require a rebuild.
Common Questions
Are Core Web Vitals a ranking factor?
Yes, though a modest one. Google has been clear that relevance matters more — a slow page with the best answer still outranks a fast page with a poor one. Where they matter most is between closely matched competitors, and for conversion rate, which is often the bigger commercial effect.
What is a good Core Web Vitals score?
LCP under 2.5 seconds, INP under 200 milliseconds, CLS under 0.1. You need to pass all three on at least 75% of real visits to be marked as passing.
Why does PageSpeed Insights give me different results each time?
The lab test at the bottom runs a simulation that varies run to run. The field data at the top comes from real Chrome users over 28 days and is the number Google actually uses. Trust the field data; use the lab data only to diagnose.
My WordPress site is slow. Is a rebuild the only fix?
Usually not. Most slow WordPress sites are slow because of unoptimised images, too many plugins and a bloated theme. Fixing those three typically gets you most of the way. A rebuild is worth considering when the theme itself is the problem or the plugin count is beyond untangling.