What Are Core Web Vitals?

The decision comes first, because it decides whether you need the rest of this at all. Then the plain English version of what the three metrics actually measure.

Open the Core Web Vitals report in Search Console and look at one thing only. Have you got any URLs marked poor.

If you have not, and your pages are split between good and needs improvement, walk away. You will spend an absurd amount of time fixing those needs improvement pages when you do not need to. As long as the page loads, it is fine.

If you are flooded with poor URLs, that is a different conversation. Spend the time and do what needs doing, because at that point real visitors are having a genuinely bad experience on your site.

This is not us being contrary. Google’s own help page for the report says to fix everything labelled poor first, and that URLs labelled needs improvement could be improved but are less important to fix. The industry reads that as a to-do list. It is a priority order, and the bottom of it is optional.

LCP is how long until the biggest thing on screen has loaded, usually your hero image or your headline. Google wants that inside 2.5 seconds. Past 4 seconds it counts as poor, and on a phone people leave before they have seen anything at all.

INP is how long the page takes to react when somebody taps or clicks. Good is 200 milliseconds or less. A page that looks loaded but ignores your first three taps has an INP problem, and it is nearly always too much JavaScript.

CLS is the annoying one. It measures how much the page shifts about as things load, so the button you were aiming at moves and you tap an advert instead. Good is 0.1 or less, and images without a width and height set are the usual culprit.

That is the whole vocabulary. You do not need to know more than this to make a sensible decision about whether to act.

The scores in Search Console are not a test that runs on demand. They come from the Chrome User Experience Report, which is real data from real Chrome users, gathered over a rolling 28 days and reported at the 75th percentile.

That last part matters more than it sounds. The number you see is what three quarters of your visitors experienced or better, so one slow visit does not sink you and one fast one does not save you.

It also means a small site with modest traffic often has no data at all, and Search Console simply says no data available. That is not a failure and it is not something to fix. Google’s documentation confirms only indexed URLs appear in the report, and that a group of URLs needs a minimum amount of data before it will be shown.

If you want a number anyway, run a single page through PageSpeed Insights. Just remember that is a lab test on a simulated device, and it is not the thing Search Console is grading you on.

Google recommends good Core Web Vitals and says page experience aligns with what its core ranking systems seek to reward. That is true, and it is also about as soft as Google’s language ever gets.

What it does not say anywhere is that a needs improvement score is holding you back. Relevance, content and links do the heavy lifting. Speed is a tiebreaker, and it matters far more for a shop taking payments than for a five-page service site.

If you are not ranking, speed is rarely the reason. A page nobody links to, that answers nothing, will not rank at two seconds either.

Here is the part the industry never puts in the proposal. People will spend hours fault finding, going through each individual page one at a time, trying to understand how to get it from needs improvement to good.

They might spend two or three weeks on that, when they could have spent those two or three weeks on outreach, or on work that would have brought paying customers to their door. That is the real price of a green score, and it never appears on the invoice.

Fix poor. Ignore needs improvement. Go and do something that brings in money.