LCP measures how long until the largest visible element renders. INP measures the delay between a user interacting and the interface responding — it replaced First Input Delay in March 2024 and is considerably harder to pass. CLS measures how much content shifts around unexpectedly during load.
The distinction that matters most in practice is field data versus lab data. Lab tools like Lighthouse simulate a single load on a synthetic connection. Field data — what Google actually uses — comes from real Chrome users on real devices and networks, aggregated over 28 days.
A site can score 98 in Lighthouse and fail Core Web Vitals in the field, because real users are on mid-tier Android phones on congested networks, not on a developer laptop.
Formula
Pass = LCP ≤ 2.5s AND INP ≤ 200ms AND CLS ≤ 0.1 (at the 75th percentile)
The 75th percentile is the threshold, not the average. Three quarters of your real visits must meet each figure, which is a materially stricter bar than a median.
Why Core Web Vitals matters
Core Web Vitals are a genuine but modest ranking signal — they rarely outweigh relevance. Their real value is commercial: LCP correlates strongly with conversion rate, and every additional second of load time cuts conversions by roughly 7%.
Why INP is the hard one now
INP measures every interaction across a session, not just the first. Sites that passed First Input Delay comfortably frequently fail INP, because the problem isn't initial load — it's heavy JavaScript blocking the main thread on subsequent clicks, filter changes, and menu opens. Tag managers and third-party scripts are the usual culprits.
Benchmarks
- LCP — good
- ≤ 2.5 seconds
- INP — good
- ≤ 200 milliseconds
- CLS — good
- ≤ 0.1
Ranges drawn from Digital Squad client accounts and published industry data. Treat them as orientation, not targets — your category may differ substantially.
Common mistakes
Optimizing for Lighthouse instead of field data
Lab scores are a debugging tool. Google uses CrUX field data, so that's what you should be tracking and reporting.
Ignoring the 75th percentile
Teams celebrate a good median while a quarter of real users have a poor experience — which is precisely the segment the threshold is designed to catch.
Treating third-party scripts as untouchable
Chat widgets, tag managers, and analytics are frequently the largest contributor to INP failures. They're negotiable more often than teams assume.
Not reserving space for dynamic content
Images, ads, and embeds without explicit dimensions are the dominant cause of CLS, and the cheapest to fix.
Where we work on this