Core web vitals are three measurements of how a page feels to use. They are not a score out of a hundred and they are not about how fast your server is. Each one describes a specific frustration a visitor might have.
This article explains them without code.
The three core web vitals
Each measure answers one question about the experience of loading a page.
- Largest contentful paint. How long until the main thing on the page appears? Usually that is a big image or the headline.
- Interaction to next paint. When somebody taps or clicks, how long until the page visibly responds?
- Cumulative layout shift. Does the content jump around while it loads, so people tap the wrong thing?
Notice that none of them measures total load time. A page can take a while to finish loading and still feel fast, provided the important part arrives early and nothing moves once it has.
What usually causes poor core web vitals
The causes are boringly consistent across sites, which is good news, because it means the fixes are too.
Large images cause most slow paint scores. A photograph exported straight from a camera can be several megabytes, and the page waits for it. Compressing images and serving them at the size they are actually displayed fixes more sites than any other single change.
Too much script causes most poor interaction scores. Every tracking tag, chat widget, popup tool and testing script competes for the same attention, and while they run, taps do nothing.
Missing dimensions cause most layout shift. When an image or advert has no size declared, the browser leaves no room for it, then shoves everything down when it arrives. Declaring width and height solves it.
Where to see your own numbers
Search Console has a core web vitals report built from what real visitors experienced, grouped into good, needs improvement and poor. That is the version that counts, because it reflects real devices on real connections.
Testing tools that score a single page in a lab are useful for diagnosis and misleading as a target. A lab test runs on a machine in perfect conditions; your visitors are on a phone with three tabs open and patchy signal.
How much do core web vitals actually matter
Honestly, less than most articles imply, and more than nothing.
They are a genuine ranking signal. However, they are a tie breaker rather than a lever. A page that answers the question badly will not outrank a good one by loading faster. Meanwhile, between two pages that both answer well, the faster one has an advantage. Weighing one signal against another gets easier once you know which parts of the work carry the most weight.
The stronger argument for fixing them has nothing to do with rankings. People leave slow pages. A visitor who gives up before your page paints costs you the same whether or not a search engine noticed.
The repairs that move the numbers
Work in this order and stop when the numbers turn green.
- Compress and correctly size every image on your busiest templates.
- Declare width and height on images and embeds so nothing jumps.
- Remove tracking and widget scripts nobody uses. Most sites carry several.
- Load whatever is left after the page has painted rather than before.
- Only then consider fonts, caching and hosting.
The first three cost nothing and need no developer on most platforms.
Core web vitals on a small site
If your site has few pages and modest traffic, Search Console may not have enough data to report on them at all. That is not a problem to solve. It simply means you are below the threshold where real user measurement becomes reliable.
In that case, test your three most important pages by hand on a phone, on mobile data rather than office wifi, and trust what you experience. If it feels slow to you it feels slow to everybody.
Core web vitals and the rest of the site
One measurement trap catches people out. Core web vitals are reported per page group rather than for the site as a whole, so a template used by hundreds of pages drags the whole report with it while your carefully tuned home page looks fine.
Check which template is failing before optimising anything. On most sites it is the blog article layout or the product page, simply because those exist in the largest numbers and nobody tests them. Fixing one template can move thousands of pages at once, which is the best return available in this whole area, and it is invisible to anybody testing the home page alone.
What not to do
Do not chase a perfect score. The last few points typically cost far more than they return, and they are usually spent on details no visitor would ever notice.
Equally, do not install a plugin that promises to fix everything automatically. Those tools frequently break layouts and disable scripts you needed, and diagnosing the damage afterwards takes longer than the original work would have.
Frequently asked questions
What are core web vitals?
Three measurements of how a page feels to use: how long until the main content appears, how quickly the page responds when somebody taps, and whether content jumps around while it loads. None of them measures total load time, which surprises people.
Do core web vitals affect rankings?
Yes, as a genuine signal, but as a tie breaker rather than a lever. A page that answers the question badly will not outrank a good one by loading faster. Between two pages that both answer well, the faster one has an advantage.
What is the most common cause of a poor score?
Large images, by a wide margin. A photograph exported straight from a camera can be several megabytes and the page waits for it. Compressing images and serving them at the size they are actually displayed fixes more sites than anything else.
Where can I check my core web vitals?
Search Console has a report built from what real visitors experienced, which is the version that counts. Lab testing tools that score one page are useful for diagnosis but misleading as a target, because they run in perfect conditions your visitors never have.
Why does my site show no core web vitals data?
Usually because there is not enough traffic for real user measurement to be reliable. That is not a fault to fix. Test your most important pages by hand on a phone using mobile data instead, and trust what you experience.
What is cumulative layout shift?
It measures whether content moves while the page loads, which is what makes people tap the wrong button. The usual cause is an image or advert with no size declared, so the browser leaves no room and then pushes everything down when it arrives.
Should I aim for a perfect score?
No. The last few points usually cost far more effort than they return and get spent on details no visitor would notice. Getting each measure into the good range is worth doing; polishing beyond that is rarely the best use of the time.
Do speed optimisation plugins work?
They can help, and they frequently break layouts or disable scripts you actually needed. Diagnosing that damage afterwards often takes longer than doing the original work by hand. Compressing images and removing unused scripts is safer and usually enough.