Your developer shows you a 95 score on PageSpeed and swears everything is fine. Google, looking at field data from your actual visitors, may be seeing a failing website. That gap between lab and field is where most B2B sites lose position without anyone noticing.
In this guide you will understand what each metric measures, the official targets, and how to diagnose your site in 15 minutes.
What Core Web Vitals Are
Core Web Vitals are the three official metrics Google uses to measure the real experience of people visiting your site: LCP (loading speed), INP (click response speed), and CLS (visual stability of the page). They are collected from real Chrome users, feed into the ranking algorithm, and show up directly in Search Console, which makes them the most concrete experience indicator a manager has on hand.
Each one answers a simple question. LCP measures how long the largest element on screen takes to appear. INP measures how long the page takes to react when the user clicks, taps, or types. CLS measures how much the layout jumps while loading, that button that moves right as you click it.
INP is the newest of the trio: in March 2024 it replaced FID as the official metric. If your last performance audit predates that, it is out of date.
One detail many people overlook: Google evaluates these metrics with field data from CrUX (Chrome User Experience Report), collected from real visitors over a 28 day window. A lab test with a green score guarantees nothing if the real experience, on your customer's 4G connection, is poor.
And it is worth calibrating expectations. Core Web Vitals is a confirmed ranking signal, but it works as a tiebreaker between pages with comparable content. Fixing the metrics does not make up for weak content, and ignoring them hands an advantage to the competitor who did the homework.
Why Marketing Managers Should Look at Core Web Vitals Before the IT Team Does
Your site opens fast on your laptop, with corporate wifi and a full cache. The customer arrives on 4G, on a mid range Android, clicking an ad you paid for. That is the scenario where the metrics fall apart, and you never see it.
The Impact on Revenue, Not Just Ranking
Google measured it: 53% of mobile users abandon a page that takes more than 3 seconds to load. The Milliseconds Make Millions study (Google and Deloitte, 2020) found that shaving 0.1 second off load time lifted retail conversion by 8.4%.
Translated to your dashboard: same media budget, same creative, higher cost per lead. Site optimization gives that margin back before any campaign adjustment.
Where Core Web Vitals Fits Into Technical SEO in 2026
Within Page Experience, the three metrics act as a tiebreaker: between two pieces of content of similar quality, the faster one tends to come out ahead. And the effect goes beyond ranking.
A slow site burns more crawl budget, indexes fewer pages, and makes it harder for ChatGPT, Gemini, and Perplexity to read. A page that AI cannot process well is unlikely to be cited in an answer.
LCP, INP, and CLS Explained Without Jargon
LCP: How Long Until the Main Content Appears
LCP (Largest Contentful Paint) measures when the largest visible element finishes loading, usually the hero image or the headline on your landing page. Classic culprits: a 2 MB banner image, a slow server, and a font that blocks rendering.
INP: The Delay Between the Click and the Page's Response
INP (Interaction to Next Paint) measures the time between the interaction and the visual reaction. The user taps "add to cart" and nothing happens for a second: that is poor INP. The most common cause is third party scripts, pixels, chats, and media tags all running at the same time.
CLS: The Layout That Jumps and Makes Users Miss the Button
CLS (Cumulative Layout Shift) measures how much the page moves while loading. A banner that suddenly appears and pushes the content down, an image with no reserved dimensions, the click meant for the menu that lands on the ad.
The Passing Thresholds for Each Metric
| Metric | Good | Needs improvement | Poor |
|---|---|---|---|
| LCP | up to 2.5s | 2.5s to 4s | above 4s |
| INP | up to 200ms | 200ms to 500ms | above 500ms |
| CLS | up to 0.1 | 0.1 to 0.25 | above 0.25 |
The detail that changes everything: Google evaluates the 75th percentile. In other words, 75% of your real visitors need to have that experience, not the average.
How to Measure Your Site's Core Web Vitals in 15 Minutes
Open PageSpeed Insights, paste the URL of your most important page, and look first at the top block, "discover what your real users are experiencing." That is the data Google uses for ranking.
Then, in Search Console, open the Core Web Vitals report. It groups URLs with the same problem, so 400 failing pages may be a single broken product template. Fixing the template resolves the entire group at once.
Field Data or Lab Data
PageSpeed shows two different numbers on the same screen, and that confuses a lot of managers. Field data comes from CrUX, the Chrome panel that collects the experience of real users over the last 28 days. Lab data (Lighthouse) is a simulation run on the spot, in a controlled environment.
Business decisions are made with field data. Lab data is there for the developer to diagnose the cause.
Prioritize by Revenue, Not by Score
Do not start fixing page by page. List the templates that concentrate conversion (home, paid media landing pages, product pages) and attack those first. Anyone running a proper website management routine tracks these metrics every month, before paid traffic takes a hit.
What to Fix First and What to Delegate to the Developer
High Impact, Low Effort Fixes
Start with what can be solved in days: compressing images to WebP, enabling cache, cutting scripts for tools nobody uses anymore in Tag Manager. Image compression alone usually brings LCP down noticeably. In the SP Labor project, 6,433 images went from 4.0 GB to 1.4 GB, and PageSpeed reached 97 on mobile.
When the Problem Is the Theme, the Plugin, or the Hosting
Some signs point to a structural bottleneck: TTFB above 800 ms (slow server), WordPress with 40 active plugins, a theme loading 2 MB of CSS on every page. In that scenario, spot optimization is like mopping the floor with the tap running. The way forward is a site optimization project, with a review of the stack and hosting.
The Brief Your Technical Team Needs to Receive
Do not ask them to "make the site fast." Ask with acceptance criteria: "home page LCP below 2.5 s in PageSpeed field data, measured over 28 days." Without a measurable target, you end up in the "we optimized and nothing changed" loop.
And remember: a good metric today degrades in three months without monitoring. A website management routine monitors all three metrics continuously and fixes issues before Search Console complains.
Where to Start This Week
That site that opens fast on your laptop keeps fooling you. The 4G user, on a mid range Android, feels every second of LCP and every button that does not respond. And Google measures all of it in the field, every day, with data from real people.
Now you know how to read those numbers. Three things to take away:
- Field data rules. The real user block in PageSpeed Insights and the Search Console report are the numbers that feed into ranking. Lab tests are for diagnosis, never for celebration.
- Images and scripts solve most of it. WebP compression, caching, and Tag Manager cleanup bring LCP down in days, without rewriting the site.
- Structural bottlenecks require engineering. Slow servers, heavy themes, and legacy code do not go away with a plugin. That calls for a site optimization project with an owner and a deadline.
Tomorrow's step fits in 15 minutes: run PageSpeed on the page that gets the most paid clicks, screenshot the field data, and bring it to the meeting with your team. If nobody on the other side of the table can say who monitors this every month, the problem is website management, and it will not fix itself.
At SP Labor, this work took PageSpeed to 97 on mobile and 99 on desktop. The method exists, and it is repeatable.
Want to know where your site is losing revenue to performance? Talk to an Expert and get a diagnosis of your case, no strings attached.
Core Web Vitals measure one thing only: how much your site respects the time of the people who pay your bills.
Frequently Asked Questions
What are Core Web Vitals?
They are the three official metrics Google uses to measure the experience of real users on your site: LCP measures loading speed, INP measures click response time, and CLS measures the visual stability of the page. The data comes from real Chrome visitors, feeds into the ranking algorithm, and appears in Search Console.
What are Google's targets for each metric?
LCP up to 2.5 seconds, INP up to 200 milliseconds, and CLS up to 0.1. Google evaluates the 75th percentile, meaning 75% of visits need to fall within the target. Above that, the page lands in the "needs improvement" or "poor" band, and you see it grouped by URL in the Search Console Core Web Vitals report.
Do Core Web Vitals affect Google rankings?
Yes, they are a confirmed ranking factor, although content and authority weigh more. In practice, the bigger impact tends to be indirect: a slow page increases abandonment before the content even appears. Google measured that 53% of mobile users give up when loading passes 3 seconds, which hurts conversion from paid and organic traffic at the same time.
What is the difference between field data and lab data?
Field data comes from real Chrome users over the last 28 days, on the phones and connections they actually have. That is the data Google uses. Lab data is a simulation run at the moment of the test, useful for diagnosing the cause but not representative of the real experience. In PageSpeed Insights, the top block is field, the bottom one is lab.
Why does my site feel fast to me but fail Core Web Vitals?
Because you test it on a laptop, with corporate wifi and a cache full from previous browsing. Your customer arrives on 4G, on a mid range Android, loading everything from scratch. Field metrics capture that scenario, your manual test does not. If Search Console flags failing URLs, trust the report, it reflects who actually visits.
How can I improve LCP quickly?
Start by compressing images to WebP, enabling cache, and cutting abandoned scripts in Tag Manager. Image compression alone usually brings LCP down noticeably: in the SP Labor project, 6,433 images went from 4.0 GB to 1.4 GB and PageSpeed reached 97 on mobile. Slow servers and render blocking fonts are the next steps.
How long does it take for Core Web Vitals to update after a fix?
Field data uses a rolling 28 day window, so the improvement shows up gradually in Search Console and PageSpeed Insights. Fixes applied today start moving the report within one or two weeks and consolidate in about a month. Use the lab test to validate on the spot whether the fix worked technically.
