In this guide 6 sections
Core Web Vitals are three measures of real visits. LCP is how fast the main content appears. INP is how quickly the page reacts to a tap. CLS is how much the layout jumps. Real users supply the numbers, so a failure points at specific templates.
The three measures in plain terms
Largest Contentful Paint, or LCP, times how long the biggest visible block takes to appear, usually the hero image or the main headline. Interaction to Next Paint, or INP, times the delay between a tap and the page responding. Cumulative Layout Shift, or CLS, scores how far things move while loading.
Think of them as three complaints. The page is slow to show. The page ignores my tap. The button moved just as I pressed it.
- LCP: the main block appears late
- INP: taps and clicks feel stuck
- CLS: content jumps and you hit the wrong thing
Field scores and lab scores are different things
Search Console shows field data, gathered from real Chrome users over the last four weeks on their own phones and connections. Lighthouse and PageSpeed lab runs use one simulated device on one afternoon.
A lab run that passes can sit beside a field report that fails, because your visitors use older Android phones on patchy mobile data. Field data decides what Google sees. Lab data helps you reproduce and fix.
Which template or image usually causes each failure
Slow LCP is nearly always an oversized hero photo, an image loaded late by a slider script or a slow server response before anything starts. Poor INP tends to come from heavy JavaScript: chat widgets, tag manager bundles, filter scripts on product lists.
High CLS comes from images without width and height, ad slots that open late, cookie banners that push content down and web fonts that swap after the text appears.
The report groups URLs by template, so a failing group such as product pages tells you where to look first.
- Hero image saved as an uncompressed PNG or very wide JPEG
- Third-party chat, review or analytics scripts
- Images, embeds and banners without reserved space
- Custom fonts without a fallback font set
Who on the team should fix what
Split the work by cause. Images and layout belong to the designer or content editor who uploads them. Scripts and plugins belong to the developer. Server speed, caching and the CDN belong to whoever manages hosting.
Marketing owns the tag manager. Every pixel added there costs speed, so ask for a review of what is still in use.
What to do first this month
Open the Core Web Vitals report in Search Console and note which group of URLs fails and on which measure. Open one URL from that group in PageSpeed Insights and read its list of opportunities.
Send the developer that list with the group name. Ask for a before and after measure on the same page.
After a fix ships, the field report takes weeks to catch up, so check lab results the same day and field data a month later.
Web Vitals triage sheet
A table with a row per failing URL group from Search Console. Columns hold the failing measure, the probable cause, an example URL, the person responsible, the date of the fix and the lab score before and after.
Open a blank worksheet to printQuestions owners ask
Do Core Web Vitals affect my search ranking?
They are one signal among many. A page with strong content can still rank with weak scores, but slow, jumpy pages lose visitors before they ever read it.
Why does PageSpeed Insights show two different results?
One block is field data from real users and the other is a lab run on a simulated phone. The field block is what Search Console reports.
My site has little traffic, so why no field data?
The report only covers a URL group once enough real Chrome visits exist. Without it, rely on lab tests and fix the obvious image and script problems.
Can a plugin fix all three?
Caching and image plugins help LCP. They cannot repair a heavy script bundle or a layout that shifts, which need template edits.