Think about the last time a website made you wait. Maybe the page was blank for a few seconds. Maybe you tapped a button and nothing happened, so you tapped it again. Maybe you went to click a link and the whole page jumped, and you hit an ad instead.
You probably didn't think "that's a poor Largest Contentful Paint." You just thought the site was rubbish, and maybe you left. Core Web Vitals are Google's attempt to put numbers on exactly those three frustrations, and they matter for two reasons: Google uses them as part of how it ranks pages, and your visitors feel them whether Google is watching or not.
Yesterday's technical SEO checklist gave you the targets. Today we'll explain what each one actually means, what usually causes a bad score, and what to do about it, without assuming you're a developer.
Why a few seconds matter so much
Speed is easy to dismiss as a "nice to have." The research says otherwise. Google's own analysis of mobile visits found that the chance of someone leaving climbs steeply with every extra second a page takes to load:
How much more likely a mobile visitor is to bounce, compared with a 1-second load
And it works in the other direction too. A study by Deloitte for Google, Milliseconds Make Millions, looked at real mobile sites and found that improving load speed by just a tenth of a second was linked to meaningful lifts in conversions:
0.1s
Faster mobile load time measured in the study
Deloitte / Google, 2020
+8.4%
Increase in conversions for retail sites
Deloitte / Google, 2020
+10.1%
Increase in conversions for travel sites
Deloitte / Google, 2020
Those are averages across many sites, not a promise for yours. But the direction is consistent across every study we've seen: faster pages keep more of the people you worked hard to attract.
The three Core Web Vitals, in plain English
Google currently measures three things. Each one maps to a question your visitor is silently asking.
| Metric | The visitor's question | Good | Needs work | Poor |
|---|---|---|---|---|
| LCP · Largest Contentful Paint | "Is anything actually showing up?" | ≤2.5s | 2.5–4s | >4s |
| INP · Interaction to Next Paint | "Did it notice I tapped that?" | ≤200ms | 200–500ms | >500ms |
| CLS · Cumulative Layout Shift | "Why did everything just move?" | ≤0.1 | 0.1–0.25 | >0.25 |
Thresholds from Google's web.dev guidance. To pass, at least 75% of real visits to a page need to hit the "good" mark, on both mobile and desktop.
- 0.0s
Visitor taps your link
The browser asks your server for the page.
- ~0.8s
First bits appear
Background colour, maybe the header. Not much to look at yet.
- ~2.2s
Largest Contentful Paint
The big thing (your hero image or main heading) finishes appearing. This is the moment LCP measures.
- ~2.6s
A banner loads late and pushes content down
That jump adds to your Cumulative Layout Shift score.
- Later
Visitor taps 'Get a quote'
The delay between that tap and the page visibly responding is what INP measures.
The timeline above is an illustration of a typical page load, not a measurement. But it shows why one number isn't enough: a page can appear quickly and still feel broken if it jumps around or ignores taps.
LCP: "Is anything actually showing up?"
LCP measures how long it takes for the biggest visible thing on the screen, usually your hero image or headline, to finish loading. It's the closest single number to "how fast does this page feel."
What usually makes it slow:
- A huge hero image, often uploaded straight from a camera or stock site at full size.
- A slow server or no caching, so the page takes ages before it even starts arriving.
- Fonts, sliders and scripts that have to load before the main content can show.
- The hero image being "lazy loaded," which tells the browser to deliberately wait.
What fixes it: resize and compress images (and use formats like WebP or AVIF), make sure the main image loads first rather than lazily, use decent hosting with caching or a CDN, and cut anything that sits in front of your main content.
INP: "Did it notice I tapped that?"
INP replaced an older metric, First Input Delay, as a Core Web Vital in March 2024. It looks at how quickly your page visibly responds when someone clicks, taps or types, across their whole visit, not just the first interaction.
What usually makes it slow: too much JavaScript doing work at the wrong moment. Common culprits are heavy page builders, piles of third-party scripts (chat widgets, trackers, heatmaps, ad tags), and complicated menus or forms that do a lot of work on every keystroke.
What fixes it: remove scripts you no longer use (most sites have a few), load the remaining third-party tools after the page is usable, and on custom-built sites, break heavy work into smaller pieces so the browser can respond to the visitor in between.

CLS: "Why did everything just move?"
CLS measures how much the layout shifts around unexpectedly while the page loads. It's the one that makes people tap the wrong button, and it's often the easiest to fix.
What usually causes it:
- Images and videos without set dimensions, so the page doesn't reserve space for them.
- Cookie banners, promo bars or ads that appear at the top and shove everything down.
- Web fonts that load late and change the size of text after it's already been shown.
What fixes it: always give images and embeds a width and height (or an aspect ratio), reserve space for banners and ads before they load, and use font settings that avoid a big visual swap.
You can't see Core Web Vitals in a design mock-up. They only show up when real people, on real phones, try to use the real site.
How to check your own scores
You don't need special software. Two free Google tools cover almost everything:
- 1
Run PageSpeed Insights on your key pages
Paste a URL into pagespeed.web.dev. The top section, 'Discover what your real users are experiencing,' is real-world data from Chrome users and is what counts. The lab score underneath is useful for diagnosing why.
- 2
Open the Core Web Vitals report in Search Console
It groups your pages into good, needs improvement and poor, separately for mobile and desktop, so you can see whether a problem is site-wide or limited to one template.
- 3
Fix one template at a time
Most sites have a handful of page types: homepage, service page, blog post, product page. Fixing the template fixes every page that uses it.
- 4
Wait for the data to catch up
Field data is collected over a rolling 28-day window, so improvements show up gradually. Don't panic if the report hasn't moved the day after a fix.
Should you obsess over a perfect score?
Honestly, no. Core Web Vitals are one signal among many, and Google has been clear that great, relevant content still matters more than shaving off the last few milliseconds. A page that passes all three is in good shape. Chasing a 100/100 lab score beyond that usually costs more developer time than it earns back.
What you should avoid is being in the red. A slow, jumpy page loses visitors before your content has a chance to work, and that's true whatever Google thinks of it.
Force Media builds fast, modern websites on Next.js that pass Core Web Vitals from day one, and we rescue slow ones too. If your site feels sluggish or your Search Console report is full of red, talk to us. We'll tell you what's slowing it down, in plain English.



