If you want core web vitals statistics you can quote without getting laughed at, you have to be careful about where the numbers come from. Most roundups recycle a single blog post from 2021 and present it as current, so this page pulls every figure from a named study and tells you the year it was measured. The short version of the core web vitals statistics below: fewer than half of mobile sites pass, the bar moved when Google swapped in a new metric, and the link between these scores and rankings is real but weaker than vendors imply.
Read the sources at the bottom before you cite anything. Where two studies disagree, that’s usually because they measured different things in different years — and I’ve flagged it rather than blended them into a fake average.
The thresholds, so the percentages mean something
Core Web Vitals are three field metrics, each scored at the 75th percentile of real page loads, according to Google’s developer documentation on web.dev. A page is “good” when it clears all three: Largest Contentful Paint (LCP) at 2.5 seconds or faster, Interaction to Next Paint (INP) at 200 milliseconds or under, and Cumulative Layout Shift (CLS) at 0.1 or lower. Miss any one and you don’t pass overall. That “all three or nothing” rule is why the combined pass rate is always lower than any single metric’s score.
How many sites pass Core Web Vitals
The most authoritative recent number comes from the HTTP Archive Web Almanac, which analyzes millions of origins in the Chrome UX Report. In its 2024 Performance chapter, 43% of websites achieved good Core Web Vitals on mobile and 54% on desktop. So the majority of the mobile web still fails — the metric that matters most, because Google indexes mobile-first.
That 43% is progress, not a plateau. Per the same Web Almanac data, mobile pass rates sat around 37% in 2022 and 2023 (measured with INP) before climbing to 43% in 2024. Desktop is comfortably ahead at 54%. Compare that against Semrush‘s earlier 2021 study of roughly 78,000 URLs per platform, which found just 16.2% of mobile URLs and 34.7% of desktop URLs passing all metrics — a reminder of how far the baseline has moved in three years, and how much of that gain came from faster devices and better defaults, not heroic optimization.
The metric-by-metric breakdown
Averages hide where the pain actually is. Here are the individual “good” pass rates from the HTTP Archive Web Almanac’s 2024 Performance chapter:
| Metric | Mobile (good) | Desktop (good) |
|---|---|---|
| Largest Contentful Paint (LCP) | 59% | 72% |
| Interaction to Next Paint (INP) | 74% | 97% |
| Cumulative Layout Shift (CLS) | 79% | 72% |
Read that table and the story is obvious: LCP is the bottleneck on mobile. Loading is where most sites lose their pass, not interactivity or layout stability. Desktop interactivity is nearly universal at 97% good INP, which makes sense — desktops have more processing headroom than the mid-range phones most of your traffic actually uses.
An older but larger page-level view backs the same shape. Backlinko’s study, run on 208,085 pages via WebCEO in 2021, found LCP “good” on 53.77% of pages, the then-current First Input Delay (FID) good on 53.85%, and CLS good on 65.13%. Average LCP came in at 2,836 milliseconds — just over the 2.5-second line — with average FID at 137.4 ms and average CLS at 0.14. Those page-level averages sat right on the wrong side of “good,” which tells you how many sites were close but not clearing the bar.
The INP switch changed the numbers
If you’re comparing statistics across years, know that Google changed the scoring. INP officially replaced FID as a Core Web Vital on March 12, 2024, per Google’s announcement on web.dev. INP is a harder test — it measures responsiveness across every interaction on the page, not just the first input.
The Web Almanac quantified the hit directly: had FID stayed as the interactivity metric, 48% of mobile sites would have passed Core Web Vitals in 2024. Under INP, that figure dropped to 43%. So a five-point chunk of the “failing” mobile web didn’t get slower — the ruler got stricter. Any pre-2024 pass rate you see was measured with the easier FID metric and isn’t directly comparable to today’s.
Do Core Web Vitals affect rankings?
This is where honesty matters most, because it’s the most oversold claim in the field. Core Web Vitals are a confirmed part of Google’s page experience signals, but the measured correlation with rankings is weak.
Semrush’s 2021 analysis of 10,000 keywords found that URLs with “good” vitals were already performing well in search, with position 2 showing the highest optimization levels on both mobile and desktop. That’s a correlation, and a modest one — better-run sites tend to both rank well and load well, which isn’t the same as speed pushing you up the results. Backlinko’s study was blunter still, reporting only “a weak correlation between Core Web Vitals scores and UX metrics,” with LCP and CLS showing no significant relationship to bounce rate, session time, or page views. Treat Core Web Vitals as a tie-breaker and a user-experience floor, not a ranking lever you can crank.
Mobile versus desktop, home versus deeper pages
Two gaps in the data are worth planning around. The first is the mobile-desktop split you’ve already seen: 43% versus 54% overall in the 2024 Web Almanac, driven mostly by LCP. If you only ever test on your laptop, you’re auditing the version of your site that passes and ignoring the one Google ranks.
The second is subtler. The Web Almanac found that secondary pages perform notably better than home pages — roughly 20 percentage points higher for desktop LCP and about 13 points higher for overall mobile Core Web Vitals. Home pages tend to be the heaviest, most script-laden page on a site, so a green homepage doesn’t guarantee your money pages pass. Sample the pages that actually earn revenue.
What this means for your SEO
Put the numbers to work instead of framing them. A few concrete moves the data supports:
- Fix LCP first. It’s the most common mobile failure across every study here, so your largest image or hero element is the highest-leverage target.
- Test on mobile field data, not desktop lab scores. The 43% versus 54% gap lives almost entirely on phones, and Google scores you at the 75th percentile of real visits.
- Audit your revenue pages, not just the homepage. Deeper pages score better on average, but that average hides the templates that quietly fail.
- Don’t expect a ranking jump. Both the Semrush and Backlinko data show a weak link — pass the threshold to stop losing tie-breakers, then spend your real effort on content and links.
This is exactly the kind of check SEO Rocket is built to run for you. Instead of opening five tabs, you ask in plain language which of your pages fail Core Web Vitals and why, and it pulls the field data, flags the LCP or INP offenders, and tells you whether it’s actually worth fixing before your rankings will move. The statistics say most sites are one or two loading problems away from passing — the job is finding which two on your site, then getting back to the content and links that move the needle.
