What Are Core Web Vitals?
In short, they’re Google’s three checks on how a page feels to use. Largest Contentful Paint (LCP) is about loading: how long until the main bit of the page shows up. Interaction to Next Paint (INP) deals with responsiveness, meaning the wait after a click or tap. And Cumulative Layout Shift, or CLS? That one checks whether things on the page jump around before they settle. Each has a score Google counts as good, and we’ll get to those in a moment.
Maybe a developer mentioned them in passing. Or a Search Console email turned up with the word “Poor” next to your homepage and you weren’t sure whether to panic. (Usually, you don’t.) Ignore the lab-coat name for a second. What it’s really tracking is stuff your visitors notice anyway, without ever putting words to it. Is anything happening? Did my tap register? Hang on, where did that button go?
We’ll go through each metric in turn, then what Google counts as a pass, where to find your own numbers and the usual fixes. There’s no code here and nothing WordPress-specific. The aim is simply that next time someone mentions LCP, you’ll know whether it matters for your site.
Core Web Vitals at a Glance: The Pass Marks
Google sorts each metric into three bands. For a page to pass, at least 75% of real visits need to land in the “good” band, and mobile and desktop are measured separately.
| Metric | Good | Needs improvement | Poor |
|---|---|---|---|
| LCP (loading) | 2.5 s or less | Up to 4 s | Over 4 s |
| INP (responsiveness) | 200 ms or less | Up to 500 ms | Over 500 ms |
| CLS (visual stability) | 0.1 or less | Up to 0.25 | Over 0.25 |
LCP: How Quickly Your Main Content Appears
Think of LCP as a stopwatch that gets clicked off the moment the largest thing in view finishes appearing. On most small business sites that’s a hero photo, a banner or a big block of intro text. Why that moment? Because that’s more or less when someone glancing at the screen thinks, right, it’s loaded. Whether scripts are still ticking away in the background doesn’t matter much to them, so it doesn’t matter to LCP either.
A Real-World LCP Example
Picture a café’s homepage with a lovely full-width photo of the counter. If that photo is a 5 MB file straight off a camera, a visitor on 4G stares at a mostly blank screen for several seconds. It doesn’t help that the menu and opening hours turned up quickly. The stopwatch won’t stop for them; it’s waiting on the photo. Re-save that image at a sensible size and the difference can be dramatic, usually with no quality loss anyone would notice.
INP: How Quickly Your Site Reacts to Clicks and Taps
With INP, Google is measuring the pause after someone acts. A tap on the menu icon, a press of “Add to basket”, the first letter typed into an email box. How long does the screen sit there before anything visibly changes? Google keeps an eye on those moments throughout the visit and reports one of the slowest, so one sticky button really can drag a page’s whole score down.
Why INP Replaced FID
Before March 2024, that spot belonged to a metric called First Input Delay. It only judged the very first interaction, which is why Google retired it in favour of INP. Fair enough, really. Nobody clicks once and leaves, and a page that’s quick the first time but freezes on the third tap is still a pain to use.
A Real-World INP Example
A booking page where you tap a date, then wait almost a second before the time slots appear, is a classic poor INP. Nothing has crashed. The browser is simply busy running scripts (chat widgets, tracking tags, a heavy booking plugin) and can’t get round to updating the screen.
CLS: Whether Your Page Stays Still While It Loads
You’ll know Cumulative Layout Shift even if the name’s new. Say you’re about to tap “Read more” on your phone. Just then a cookie notice or an advert loads in above it, the text jumps down a few lines, and you’ve tapped some other link entirely. Annoying, isn’t it? Since this is about things moving, CLS isn’t counted in seconds. It’s a score combining how much of the screen shifted with how far it travelled, hence a pass mark as small as 0.1. Common causes include images with no size set, late-loading embeds and ads, and web fonts that swap in and resize the text.
Do Core Web Vitals Affect Google Rankings?
A bit, yes. Some of the sales emails doing the rounds make them sound far more decisive than they are, though. Google’s line is that good scores fit with what its core ranking systems are trying to reward, and it encourages site owners to hit them. It doesn’t promise a ranking boost for passing, and a page with brilliant scores and thin content won’t outrank a genuinely more useful one. We’d look at them as part of looking after visitors. Honestly, the bigger payoff is usually commercial. When a page loads fast and doesn’t fight people, more of them stick around long enough to ring or send an enquiry.
How to Check Your Core Web Vitals
Google PageSpeed Insights
Go to pagespeed.web.dev, paste in a page address and run the test. The top section shows how real Chrome users have experienced that page (or your whole site, if there isn’t enough data for the single URL) over the previous 28 days. That part decides whether you pass. Scroll down and you’ll find a separate lab report, run by a tool called Lighthouse, with a score out of 100 and a long list of suggestions. Flick between the Mobile and Desktop tabs as well. Mobile is where we usually find the trouble.
Search Console’s Core Web Vitals Report
If you have Google Search Console set up, open Experience > Core Web Vitals. Instead of one page at a time, it groups similar URLs together and labels them Good, Need improvement or Poor, based on the worst of the three metrics. It uses the same real-visitor data over 28 days. Very quiet sites may not have enough traffic to appear here at all, which isn’t a fault, just a lack of data. Once something is fixed, you can start validation, which then tracks the pages for 28 days.
Field Data vs Lab Data
This is where most of the confusion comes from. Field data is collected from real visitors on their own phones, laptops and connections, and it’s what Google uses for Core Web Vitals. Lab data comes from one simulated visit in controlled conditions. Useful for spotting problems and checking a fix works? Definitely. Your official result? No. A lab test also can’t measure INP properly, since nobody is clicking anything, so it relies on related measures instead. If your lab score is 45 and your field data passes, you’re passing.
Getting “Poor” results and not sure where to start? Our technical SEO team spends a lot of time in these reports. We’ll go through both, separate the real problems from the noise and give you a plain list of what’s worth doing. Book a free consultation.
Common Ways to Improve Poor Scores
No two sites are quite the same, though in practice we tend to find the same handful of problems.
Fixing a Slow LCP
We’d start with images, since they’re behind a slow LCP more often than not. Get each one down to the size it’s actually displayed at, compress it, and save it as WebP or AVIF if your site supports those. One small but common mistake is lazy-loading the main image at the top, which tells the browser to wait before fetching it. If the server’s sluggish, better hosting, caching or a CDN (content delivery network) tends to help, and your developer can trim any CSS or JavaScript holding the first view back.
Fixing a Poor INP
First job: write down every third-party tool the site loads. Chat widgets, pop-ups, review feeds, a couple of tracking tags from campaigns that ended last year: they all queue up for the browser’s attention. Bin anything nobody uses. The rest can often wait until it’s actually needed, and a developer can break long scripts into smaller pieces so the page has gaps in which to respond.
Fixing a High CLS
For CLS, the fix is mostly about reserving space. When an image, video or embed has its width and height set, the browser can hold a gap open before it arrives. Ad slots need the same treatment. Cookie banners are better either given their own reserved space or shown as an overlay at the bottom of the screen, where they sit on top of the page and push nothing down. And keep an eye on fonts. A late font swap can make a heading bigger and nudge everything beneath it.
When the Theme Is the Problem
Sometimes, though, you hit a wall. A heavy theme or page builder can load far more code than any single page needs, and after a while you’re paying for small gains that don’t add up to much. In that situation a rebuild can genuinely be the cheaper route. When our website design team builds a site, performance is part of the plan from the first wireframe. Our website care plans then handle hosting, speed and Core Web Vitals tuning, which stops the scores creeping back down every time a plugin updates.
Frequently Asked Questions
What are Core Web Vitals?
They’re the three measures Google uses for real-world page experience, covering loading (LCP), responsiveness (INP) and visual stability (CLS).
What is a good Core Web Vitals score?
For a “good” rating, LCP should be 2.5 seconds or under, INP 200 milliseconds or under and CLS 0.1 or under. Pages pass when at least 75% of real visits manage all three.
Do Core Web Vitals affect Google rankings?
Only to a point. Google describes them as consistent with what its ranking systems reward, yet a pass doesn’t secure any position. Relevant, useful content counts for more.
Why are my PageSpeed Insights and Search Console scores different?
Different data, mostly. Search Console reports 28 days of real visits across groups of similar pages, whereas the lab test in PageSpeed Insights is a single simulated load of one URL.
What replaced First Input Delay (FID)?
INP replaced it in March 2024. FID only timed the first interaction on a page, whereas INP considers responsiveness over the whole visit.
Where Your Website Goes From Here
Strip it right back and Core Web Vitals ask whether your page loads quickly, reacts when used and stays put. This week, put your homepage and your busiest service page through PageSpeed Insights. Look at the mobile field data first and see which metric is showing amber or red. Fix that one first.
If you’d rather not tackle it alone, we can pick it up at whatever level suits: a technical SEO review, ongoing care and maintenance, or a new website built to pass from the start. Rankings aren’t something anyone can promise. What we can do is make the site quicker and steadier for the people who use it. Book a free consultation and we’ll look at your scores together.
