Outrings
Search engine optimisation

Does page speed affect SEO?

Yes, modestly and as a tie-breaker. It affects whether people stay far more than it affects where you rank.

3 min read
Short answer

Speed is a confirmed but small ranking factor, mostly acting as a tie-breaker between pages of similar quality. Its larger effect is on behaviour: slow pages are abandoned, and abandonment costs you conversions and eventually rankings by a longer route.

What Google actually measures

Core Web Vitals, measured from real users in Chrome rather than from a lab test:

MetricWhat it measuresGood
LCPWhen the largest visible element finishes renderingUnder 2.5s
INPHow quickly the page responds to interactionUnder 200ms
CLSHow much the layout jumps around while loadingUnder 0.1

Field data — what real visitors experienced — is what counts. A perfect lab score with poor field data means your test conditions are not your users' conditions.

What actually makes pages slow

In practice, and in rough order of impact:

  • Images. Almost always the largest thing on the page. Serve modern formats, size them for the display, lazy-load below the fold, and set width and height so nothing shifts.
  • Third-party scripts. Analytics, chat widgets, tag managers, ad networks. Each one blocks or delays something, and they are the usual reason a fast site becomes slow.
  • Render-blocking CSS and JavaScript. Anything in the head that must load before painting.
  • Fonts. Web fonts delay text rendering. Use font-display: swap and preload the critical one.
  • Server response time. If the first byte takes 800ms, nothing downstream can rescue it.
  • Layout shift. Images without dimensions, ads injected above content, fonts swapping at different metrics.

The order worth fixing things in

  1. Measure field data in Search Console, not just a lab score. Fix what real users experience.
  2. Optimise the largest image on the page. Frequently this alone fixes LCP.
  3. Audit third-party scripts and remove the ones nobody looks at. This is usually the biggest single win and the most resisted.
  4. Set explicit dimensions on every image and reserve space for anything injected. This fixes CLS cheaply.
  5. Defer non-critical JavaScript.
  6. Only then consider a CDN, caching layer or hosting change.
A caution about lab scores: a Lighthouse run on a fast laptop over office broadband is not what your users have. Chase the field data. It is also normal for a genuinely fast site to score imperfectly in a lab test, and to matter not at all.

Keeping it in proportion

Speed will not rescue a page that does not answer the question, and slowness will not sink a page that is the best answer available. The realistic framing: it is a tie-breaker in search and a significant factor in whether people stay. The second is why it is worth doing.

A useful reference point — every additional second of load time measurably reduces conversion, across every study that has looked. That effect is larger and more immediate than the ranking effect.

What our audit reports about this

Every item below is measured directly, not inferred. Run it against your own site and the result names the exact rule or header responsible.

  • Server response time and transfer size, measured directly during the audit.
  • Compression, caching headers and HTTP protocol version.
  • Render-blocking resources, third-party script count and their origins.
  • Images without explicit dimensions, and images not lazy-loaded below the fold.

For agents and scripts, the same measurement is at /api/v1/infra?url=yoursite.com — see the API documentation.

Related questions

Do I need a perfect PageSpeed score?

No, and chasing 100 is usually a poor use of time. Getting into the "good" band for the three Core Web Vitals captures nearly all the benefit; the last few points cost disproportionately.

Is lab or field data more important?

Field data. It is what Google uses and what your users actually experienced. Lab tests are for diagnosis — useful for finding the cause once field data tells you there is a problem.

Will a CDN fix my speed?

It helps with geographic distance and static assets. It does not fix a slow server-side render, oversized images or twelve third-party scripts. Fix those first, then a CDN adds to an already-fast site.

Read next

All 50 guides · How every check works · API for agents