Page speed guide
Core Web Vitals: what they are and how to pass.
What the three metrics LCP, INP and CLS measure, which scores count as good, and how to find and fix what slows your pages.
Short answer
What are Core Web Vitals?
Core Web Vitals are Google's metrics for real-world page experience: loading (LCP), interactivity (INP) and visual stability (CLS). Good scores are an LCP within 2.5 seconds, INP of 200 milliseconds or less and CLS of 0.1 or less.
In brief
Five things to know first.
Three metrics
Loading, interactivity and visual stability, measured as LCP, INP and CLS on real visits.
Clear thresholds
Good means LCP within 2.5 seconds, INP at 200 milliseconds or less and CLS at 0.1 or less.
Real users decide
Pass or fail is judged at the 75th percentile of page loads, on mobile and desktop.
A search signal
Google highly recommends good scores for success with Search and a great user experience.
Fix the slowest part
Find which metric fails, then target its cause instead of tuning everything at once.
The basics
What the three metrics measure.
The three metrics measure real-world user experience for loading performance, interactivity and visual stability. Google's Chrome Team introduced the set in early 2020.
They are part of the wider Web Vitals initiative, which gives unified guidance on quality signals for a great experience on the web.
| Metric | What it measures | |
|---|---|---|
| LCP | Largest Contentful Paint | How fast the main content loads |
| INP | Interaction to Next Paint | How fast the page responds to input |
| CLS | Cumulative Layout Shift | How much the layout jumps around |
Thresholds
The Google Core Web Vitals thresholds.
Each metric has one target. Check pass or fail at the 75th percentile of page loads, segmented across mobile and desktop.
| Metric | Good score |
|---|---|
| LCP | Within 2.5 seconds of when the page first starts loading |
| INP | 200 milliseconds or less |
| CLS | 0.1 or less |
Search
Page speed SEO and Core Web Vitals.
Google says it highly recommends good scores for success with Search and a great user experience. Its documents say users prefer sites with a great page experience.
Google also notes that helpful content generally offers a good page experience. Speed supports good content; it does not replace it.
Recommended by Google
Google highly recommends good scores for success with Search.
Experience, not only rank
Users prefer sites with a great page experience, so speed helps people too.
Content still comes first
Google says helpful content generally offers a good page experience.
See how your site reads to Google and AI.
LLMReach, an SEO and GEO agency, starts with a free AI audit reviewed live with you. All it needs is your website URL and the names of two or three competitors.
Measuring
Google site page speed: how to measure it.
Use real-user data first. The Chrome User Experience Report collects anonymized, real user measurements for each Core Web Vital.
PageSpeed Insights measures FCP, LCP, CLS and INP. Lab scores vary with network, hardware and resource contention, so rerun tests before you trust one number.
| Tool | Data type | Use it to |
|---|---|---|
| Chrome User Experience Report | Real users | See how real visitors experience pages |
| PageSpeed Insights | Real users and lab | Check metrics and find issues per page |
Steps
How to improve site speed, metric by metric.
To optimize site speed, work metric by metric. Start with the one failing for the most visitors.
Measure on real users
Check real-user data and note which of LCP, INP and CLS misses its good threshold.Fix the failing metric
Find the cause behind that metric and fix one thing at a time.Stop layout shifts
Find the elements that move as the page loads and reserve space for them.Retest and compare
Rerun PageSpeed Insights after each change, remembering that lab results vary between runs.Watch the field data
Check real-user data over time to confirm the improvement reaches visitors.
Mistakes
Common mistakes and myths.
Treat the scores as a guide to the visitor's experience, not the goal itself. The metrics that make up the set will evolve over time, so build habits, not one-off fixes.
| Common mistake | Better approach | |
|---|---|---|
| One lab test | Trusting a single score | Rerun tests; lab results vary |
| Lab only | Ignoring real-user data | Use field data at the 75th percentile |
| Everything at once | Tuning every metric together | Fix the failing metric first |
| Speed over content | Treating speed as the whole job | Pair speed with helpful content |
FAQ
LCP, INP and CLS: common questions.
What do LCP, INP and CLS measure?
LCP measures loading performance, INP measures how quickly a page responds to interaction, and CLS measures visual stability. Together they describe real-world user experience on a page.
What are the thresholds for good scores?
A good LCP is within 2.5 seconds of page start, a good INP is 200 milliseconds or less and a good CLS is 0.1 or less. Google suggests judging them at the 75th percentile of page loads, on mobile and desktop.
Do Core Web Vitals matter for SEO?
Good scores are recommended for SEO. Google says it highly recommends good scores for success with Search and a great user experience, and notes that helpful content generally offers a good page experience.
What does INP measure?
INP, Interaction to Next Paint, measures how quickly a page responds when people interact with it. Google advises an INP of less than 200 milliseconds for a good user experience.
How to improve Core Web Vitals and site speed?
Improving the three metrics starts with real-user data. Identify which of LCP, INP and CLS fails, fix its cause, retest in PageSpeed Insights and confirm the change in field data over time.
Where can I check my page speed?
PageSpeed Insights reports FCP, LCP, CLS and INP for a page. The Chrome User Experience Report supplies anonymized real-user data for each Core Web Vital, which shows how actual visitors experience your pages.
Why does my PageSpeed score change between tests?
PageSpeed scores vary because of local network availability, client hardware and resource contention. Rerun a test several times and compare it with real-user data before you judge a change.
Free AI Audit
Find out how your site reads to Google and AI.
You leave knowing where the gap is, what is causing it, and which changes would matter first, whether we work together or not.
What to expect
Audit
Before the call, we run your buyers' questions through ChatGPT, Claude, Perplexity and Gemini.
Map
On the call, we show you where competitors are cited and you are not, and what's causing it.
Decide
An honest read on whether the Citation Stack fits. If it doesn't, you'll hear it on the call.
Don't see a time that works? Email Karim directly: contact@llmreach.ai