Products SEO & GEO Agent Paid Ads Agent Website Agent Conversion Agent
How it works Pricing FAQ Blog Book a call

Core Web Vitals Scores: What Good, Needs Work and Poor Mean

Updated · 6 min read

Short answer Good means LCP of 2.5 seconds or less, INP of 200 milliseconds or less and CLS of 0.1 or less. Poor starts above 4 seconds, 500 milliseconds and 0.25. Anything between is needs improvement. Each grade comes from the 75th percentile of real visits, so three of every four visits must hit the mark.

Want help applying this to your business? Talk it through on a free 30-minute call.

Book a call

Core Web Vitals scores: the cutoffs for each metric

Each of the three metrics gets one of three grades. These are the cutoffs Google's PageSpeed Insights uses, as of November 2026 (PageSpeed Insights: About PageSpeed Insights):

MetricWhat it timesGoodNeeds improvementPoor
LCP (Largest Contentful Paint)When the biggest image, text block or video on screen finishes showing2.5 s or lessOver 2.5 s, up to 4 sOver 4 s
INP (Interaction to Next Paint)How long the page takes to react to a tap, click or key press200 ms or lessOver 200 ms, up to 500 msOver 500 ms
CLS (Cumulative Layout Shift)How much things jump around on screen0.1 or lessOver 0.1, up to 0.25Over 0.25

The edges matter. An LCP of exactly 2.5 seconds is good; 2.6 is not. An LCP of exactly 4 seconds is still "needs improvement"; poor starts just above it. A millisecond (ms) is a thousandth of a second, so 200 ms is a fifth of a second.

Three details owners get wrong:

  • CLS has no unit. It isn't seconds or percent. It's a score built from how much of the screen moved and how far, added up over the worst burst of shifts in a visit (web.dev: Cumulative Layout Shift).
  • INP is close to your slowest tap, not the average. Google takes "the longest interaction observed, ignoring outliers," dropping one top value per 50 interactions (web.dev: Interaction to Next Paint). Scrolling doesn't count.
  • Mobile and desktop use the same cutoffs. Google kept one set because people expect the same speed on any device, even though phones have a harder time reaching it (web.dev: Defining the Core Web Vitals metrics thresholds).

What does "good" mean: the 75th percentile

Your grade isn't the average visit. Google lines up every real Chrome visit from the past 28 days (from users who opted in) and reads the one three-quarters of the way down the list, called the 75th percentile:

"In other words, if at least 75 percent of page views to a site meet the "good" threshold, the site is classified as having "good" performance for that metric." (web.dev: Defining the Core Web Vitals metrics thresholds)

The same page adds the other side: if at least a quarter of visits land in poor, the metric is graded poor. So one slow phone on a bad connection won't fail you, but a page that's slow for one visitor in three will.

A page passes the Core Web Vitals assessment only when all three metrics are good at the 75th percentile. If there aren't enough INP visits to measure, it passes on LCP and CLS alone. If LCP or CLS lacks data, there's no grade at all (PageSpeed Insights: About PageSpeed Insights).

How to read your Core Web Vitals report in PageSpeed Insights

Enter your page in PageSpeed Insights and read it top to bottom:

  1. Pick the Mobile tab first. Most service-business visitors arrive on a phone, and the two tabs are graded separately.
  2. Find the real-user section at the top. It shows "Passed" or "Failed" for the Core Web Vitals assessment, from Chrome users over the previous 28 days.
  3. Check whether it says this URL or origin. If one page has too few visits, the tool shows your whole site (the origin) instead. If the whole site is too small, there's no real-user data at all.
  4. Read the big number on each metric. That's the 75th percentile, graded with the cutoffs above.
  5. Read the colored bar under it. Green, amber and red show what share of visits fell in each grade. Google's own example: 11% in LCP's amber bar means 11% of visits loaded the main content in 2.5 to 4 seconds.

Steps 2 to 5 follow PageSpeed Insights: About PageSpeed Insights.

The section also lists First Contentful Paint (when anything first appears; good is 1.8 s or less, poor over 3 s) and Time to First Byte (when your server starts answering; good is 0.8 s or less, poor over 1.8 s). Neither is a Core Web Vital. Treat them as clues: a poor Time to First Byte usually drags LCP down with it.

The 0–100 score is a different grade

Lower on the page sits the Performance score from Lighthouse, Google's lab test. It colors scores the same way, with different numbers (Chrome for Developers: Lighthouse performance scoring):

Lab Performance scoreColorGrade
90–100GreenGood
50–89OrangeNeeds improvement
0–49RedPoor

This score comes from one simulated load (a mid-range phone on the Mobile tab), not from your visitors. Google's own note: good lab data "does not necessarily mean real-user experiences will also be good" (PageSpeed Insights: About PageSpeed Insights). When the two disagree, the real-user grades are the ones that describe your customers. Search Console's Core Web Vitals report uses the same real-user data, grouped across your site; see Google Search Console for a small business.

Why the cutoffs are 2.5 seconds, 200 ms and 0.1

Google picked numbers real sites can hit. A "good" cutoff only counts as achievable if at least 10% of websites already meet it. "Poor" is set so that, without research saying otherwise, the worst 10–30% of sites fall into it (web.dev: Defining the Core Web Vitals metrics thresholds). For LCP, Google found 2.5 seconds was the fastest target top sites hit consistently.

Google calls these thresholds stable and says updates come with notice, on a predictable yearly cycle (web.dev: Web Vitals). The last change replaced FID (First Input Delay) with INP on March 12, 2024 (web.dev: Interaction to Next Paint becomes a Core Web Vital on March 12). Any score guide that still lists FID is out of date.

Does "needs improvement" hurt my Google ranking?

Somewhat, and only on close calls. Google says it highly recommends "good Core Web Vitals for success with Search" (Google Search Central: Understanding Core Web Vitals and Google search results), but it also says it "always seeks to show the most relevant content, even if the page experience is sub-par" (Google Search Central: Understanding page experience in Google Search results). How much speed weighs against content is covered in Core Web Vitals: does website speed affect Google ranking?

Example

A plumber's home page, LCP 2.8 seconds

The bar shows 68% green, 25% amber, 7% red. The 75th-percentile visit sits in amber, so LCP is "needs improvement" and the assessment fails. Getting about 7 more visits in 100 under 2.5 seconds flips it to good.

A dental office, INP 180 ms on mobile, 90 ms on desktop

Both are good. The lab Performance score says 62, orange. The lab score doesn't change the grade: the real-user assessment passes.

A new landscaping site, no real-user data

PageSpeed Insights shows only the lab score. The site has too few Chrome visits to grade. Use the lab score to spot obvious issues, and check again once traffic grows.

FAQ

Is a CLS of 0.1 good or needs improvement?

Good. The cutoff includes 0.1 itself; 0.11 is needs improvement.

Why does my score change every time I test?

The 0–100 lab score varies run to run with network and test-machine conditions, even if you changed nothing (PageSpeed Insights: About PageSpeed Insights). The real-user numbers move slowly because they cover 28 days.

Are the thresholds different for desktop?

No. Same cutoffs for both, graded separately (web.dev: Defining the Core Web Vitals metrics thresholds).

Is FID still a Core Web Vital?

No. INP replaced it in March 2024. INP's good mark is 200 ms or less.

What's a good PageSpeed score?

90 or above is green. It's a lab diagnostic, though, not the Core Web Vitals grade.

Different for you?

Keep learning

NextAbove-the-fold content: what to load first for speedWhat to load first on a page: find your LCP element, put it in the HTML at high priority, never lazy-load it, and lazy-load what sits below the fold.

Do it yourself

Crawlability checklist: make every page reachable by GoogleA crawlability checklist for a small business site: find orphan pages, robots.txt blocks, broken links, redirect chains and server errors, with free tools.

Ask us one question

One line is enough. We reply by email.

Sources

  1. PageSpeed Insights: About PageSpeed Insights
  2. web.dev: Defining the Core Web Vitals metrics thresholds
  3. web.dev: Web Vitals
  4. web.dev: Largest Contentful Paint (LCP)
  5. web.dev: Interaction to Next Paint (INP)
  6. web.dev: Cumulative Layout Shift (CLS)
  7. web.dev: Interaction to Next Paint becomes a Core Web Vital on March 12
  8. Chrome for Developers: Lighthouse performance scoring
  9. Google Search Central: Understanding Core Web Vitals and Google search results
  10. Google Search Central: Understanding page experience in Google Search results