Core Web Vitals: The LCP / CLS / INP Optimization Manual
Key takeawayCore Web Vitals is a set of core metrics for page loading stability and interaction responsiveness: LCP measures how fast the main content appears, CLS measures whether the page jumps around while loading, and INP measures how quickly interactions get a visual response. Together they determine the page experience assessment and indirectly influence crawl frequency.
When this applies
- Sites with many failing URLs in GSC's Page Experience report
- Sites with visible layout jumps while loading or laggy responses after clicks
- Sites heavy on images, animations, and third-party scripts
- Sites clearly slower on mobile than on desktop
| Metric | What it measures | Good threshold | Common causes of failure |
|---|---|---|---|
| LCP | Render time of the largest content element | ≤ 2.5 seconds | Unoptimized large images, render-blocking CSS/JS, slow TTFB |
| CLS | Layout shift accumulated during loading | ≤ 0.1 | Images without dimensions, no placeholders, ads and popups inserting content |
| INP | Delay from interaction to visual feedback | ≤ 200 ms | Long tasks blocking the main thread, heavy event handlers |
Step-by-step diagnosis
- Group the Page Experience report in GSC by issue type and fix the category affecting the most pages first.
- Test mobile and desktop separately in PageSpeed Insights to pin down the specific elements and requests involved.
- LCP: find the largest content element (usually the hero image or title block), compress it, use a modern format, and make sure it is not queued behind critical CSS/JS.
- CLS: set fixed dimensions or aspect ratios on all images and ad slots, reserving space with aspect-ratio.
- INP: split long tasks, move non-essential computation into requestIdleCallback or Workers, and cut re-renders and unnecessary event listeners.
- Compress TTFB — it is a precondition for LCP; a slow server guarantees a poor LCP.
- Re-test monthly and keep all three metrics under continuous monitoring.
Common mistakes
- Testing only desktop — on sites with a high mobile share, good desktop scores are meaningless.
- Optimizing only the homepage — LCP problems usually live in templates and affect every page at once.
- Adding heavy preloading plugins to chase scores, which ends up slowing LCP further.
- Optimizing the frontend while ignoring TTFB — when the backend is slow, frontend gains have a low ceiling.
- Blaming CLS on user actions — user-triggered shifts are exempt; script-triggered ones are not.
- Not re-testing after optimizing — metrics regress as third-party scripts update.
FAQ
Do Core Web Vitals directly affect rankings?
Not as a direct ranking factor, but they are part of the page experience assessment and indirectly influence crawl frequency and user behavior signals.
What is the passing threshold for LCP?
Good is within 2.5 seconds, needs improvement is 2.5-4 seconds, poor is above 4 seconds. Judged on field data (75th percentile), not lab data.
Did INP replace FID?
Yes. From 2024, INP replaced FID as the third core metric. FID measured only the first interaction's input delay; INP covers all interactions across the page lifecycle.
Version notes2026-11-20: first release, compiled from official Google documentation and public technical sources; updated as engine rules change. Engine behaviour details follow the latest official Google documentation.
Related reading
How Overseas Hosting Latency and Packet Loss Really Affect IndexingWhen CDN Rules Block Crawlers: A Troubleshooting GuideUncompressed and Non-Lazy-Loaded Images: How They Hurt Indexing and PerformanceNon-Standard HTTPS Configurations and the Risks They CreateWhere Internal Link Equity Leaks Out: The Common PointsNote: this page is a summary of publicly documented search-engine rules and is for self-check reference only. Engine behaviour is subject to the latest official Google documentation; verify against your own Search Console data.