Does page speed affect Google rankings?
Short answer: yes, but less than most SEO content suggests. Core Web Vitals are one of hundreds of ranking signals, and content quality beats speed almost every time.
Where page speed really matters is as a tiebreaker.
What Google has actually said
Google's public stance, from its own documentation:
Content quality is more important than page experience. A great page doesn't have to be fast, and a fast page can't rescue bad content.
But also:
When multiple pages have similar relevance, page experience helps us identify the best ones.
Translation: if you have great content and terrible speed, you might still rank. If you have terrible content and great speed, you won't. If you have decent content like everyone else in a competitive SERP, speed helps break the tie.
What the data shows
Studies (Backlinko, Ahrefs, Search Engine Land) consistently find a weak positive correlation between Core Web Vitals and rankings — sites with better vitals rank slightly better on average, but the correlation is swamped by content and backlink signals.
The clearest ranking impact is at the "poor" threshold. Pages in the "poor" bucket for LCP, INP, or CLS tend to lose positions in competitive queries. Moving from "poor" to "needs improvement" unlocks that penalty; moving from "needs improvement" to "good" is smaller.
What actually matters
The practical hierarchy for SEO, in rough order of impact:
- Content quality and relevance — does the page actually answer the query?
- Topical authority and internal linking — does the site have depth on this topic?
- Backlinks — who cites this page from where?
- On-page SEO basics — title, meta description, heading structure, URL, image alt text
- Core Web Vitals — as the tiebreaker
Sites that obsess over Lighthouse scores and ignore content usually don't rank. Sites that produce genuinely useful content and have decent (not perfect) vitals usually do.
What to do
If your Core Web Vitals are in the "poor" bucket, fix them — that's where the ranking cost is highest.
If they're "needs improvement," fix them when you have time, but don't drop content work to do it.
If they're "good," stop optimizing and write more articles.
For most sites, the ROI curve flattens sharply once vitals are green. Going from 2.8s LCP to 2.2s LCP won't meaningfully change rankings. Going from 5.1s LCP to 2.8s LCP might.
Tools for a quick check
- PageSpeed Insights — shows both lab (Lighthouse) and field (CrUX) in one view
- Google Search Console → Experience → Core Web Vitals — your own site's CrUX aggregated across pages
- This tool — type in any domain, see CrUX data immediately
By Paulo de Vries · Published