What is page speed, in SEO terms?
When people ask "does page speed affect SEO", two different things get mixed together. Page speed in the everyday sense is how fast a site feels. Page speed in the SEO sense is narrower: a specific set of measurements called Core Web Vitals that Google collects from real Chrome users and feeds into its ranking systems. The two overlap, but Google is not timing your site with a stopwatch; it is scoring three defined metrics covering loading, responsiveness, and visual stability.
That distinction matters because it tells you what to optimise and when to stop. If you want the three metrics unpacked without jargon, our guide to Core Web Vitals in plain English covers each one with the thresholds Google publishes.
What Google has actually said about speed and rankings
Google is unusually clear on this subject, which makes the confident myths around it all the stranger. Its Search documentation on Core Web Vitals states that the metrics are used by its ranking systems, and encourages good scores "for success with Search". The supporting engineering guidance at web.dev defines the thresholds: main content within 2.5 seconds, interaction response within 200 milliseconds, layout shift below 0.1, each judged at the 75th percentile of real page visits.
At the same time, Google's spokespeople have repeatedly said the signal is modest. The consistent public line, from the 2021 page experience update onwards, has been that relevance wins: a slow page with the best answer will outrank a fast page with a worse one. The fairest reading of everything Google has published is that speed behaves like a tiebreaker. When two pages compete closely on relevance and authority, the better experience can edge ahead. That is real, but it is not a lever that lifts weak content.
The direct ranking effect: real, modest, threshold-based
Three properties of the direct effect are worth having straight. First, it is measured from field data: real visitors' experiences collected through the Chrome User Experience Report, not the simulated lab test you run on demand. Second, it appears to be threshold-based: Google's guidance is about passing "good" thresholds, and there is no evidence that being twice as fast as the threshold earns extra credit. Third, it is small relative to content and links, which no amount of speed work substitutes for.
The practical conclusion for a UK service business: if your pages fail Core Web Vitals and your closest competitors pass, you are giving away a tiebreaker in every close contest, and fixing that is worthwhile. If you already pass, chasing further milliseconds for ranking reasons alone is effort spent in the wrong place.
The indirect effects: where speed really earns its keep for SEO
The direct signal gets the headlines, but the indirect effects are where speed does most of its SEO work, and they operate whether or not rankings move.
Crawling. Google discovers and refreshes your pages with a finite crawling budget per site. Slow server responses mean Googlebot fetches fewer pages per visit, so new content is found later and updates take longer to register. On small sites this rarely bites; on larger sites with hundreds of pages, a sluggish server measurably delays indexing.
Engagement. Visitors abandon slow pages, and visitors who arrive from search and immediately leave are visitors your content never got the chance to satisfy. Whatever weight search engines do or do not give behaviour signals, a page nobody waits for cannot earn links, mentions, or return visits, and those absolutely are ranking inputs.
Conversion. This one is not a ranking factor at all, and it is still the strongest commercial argument. SEO exists to bring visitors; a slow site then wastes them. Slowness sits high on our list of reasons websites fail to convert because it taxes every enquiry the rankings deliver. Doubling your conversion of existing traffic is worth the same as doubling your traffic, and it is usually cheaper.
Speed factors and their SEO effect, direct or indirect
Here is how the pieces line up when you separate what Google measures from what merely follows:
| Speed factor | Direct or indirect SEO effect | What it actually influences |
|---|---|---|
| Core Web Vitals (field data) | Direct | A confirmed ranking signal; a modest tiebreaker between comparable pages |
| Server response time | Both | Feeds the loading metric, and governs how efficiently Googlebot can crawl |
| Lab scores (PageSpeed 0 to 100) | Indirect only | A diagnostic aid; Google's rankings do not use the lab score itself |
| Perceived speed and bounce | Indirect | Whether visitors stay long enough for content to earn links and return visits |
| Mobile speed specifically | Both | Google indexes the mobile site first, and most field data comes from phones |
| Speed's effect on conversion | None (commercial) | Not a ranking input, but decides what your existing rankings are worth |
Common myths about site speed and rankings
"A 100 PageSpeed score means better rankings." No. The 0 to 100 number is a lab score from a simulated device. Google ranks on field data, so a 65 with passing real-user metrics beats a 100 nobody actually experiences. Chasing the round number is a vanity project.
"Speed does not matter, content is everything." Half true. Content dominates, but between close competitors the experience signal exists, and the indirect costs of slowness, poorer crawling and abandoned visits, apply to everyone regardless of ranking position.
"My site feels fast to me, so this is handled." Your browser has the site cached and your connection is unrepresentative. Only field data from real visitors, or a proper test, can tell you the truth, and the mobile numbers usually tell a harsher story than your desktop impression. Our post on why websites are slow walks through the ten usual causes and how to test properly.
"Speed work should come before everything else in SEO." Rarely. If your pages fail the thresholds badly, fix that. Otherwise, targeting, content quality, and internal structure move rankings far more per hour invested, which is why speed is one strand of a proper SEO engagement rather than the whole of it.
What to prioritise: a sensible order for speed and SEO work
First, get the verdict. Check your field data in PageSpeed Insights or Search Console's Core Web Vitals report. Passing already? Bank it, monitor it quarterly, and spend your energy on content. Failing? Identify which metric fails and on which pages, since the fix for slow loading is entirely different from the fix for layout shift.
Second, fix to the threshold, not to perfection. The goal is moving real-user measurements into Google's good range on your important templates. That usually means images, caching, hosting, and script weight, in roughly that order of effort against return.
Third, treat everything beyond the threshold as conversion work, not SEO work, and judge it by enquiries rather than rankings. If you want to know where your site stands on speed alongside clarity, trust, and structure, the free audit scores all of it in about a minute and shows you which problem deserves your next pound of effort.
Keep a record of the page type, test date, field-data status, and change made. Speed scores fluctuate between runs, so a single green test is not proof that the underlying experience improved. Compare like with like and wait for enough real-user data before making a ranking claim. The useful outcome is a consistently responsive site, not a perfect screenshot from one test.
Frequently asked questions
Is page speed a confirmed Google ranking factor?
Yes. Google has used speed in rankings since 2010 on desktop and 2018 on mobile, and since 2021 the specific measurements have been Core Web Vitals. Google's own documentation describes the effect as one signal among many, and its spokespeople have repeatedly characterised it as more of a tiebreaker than a dominant factor. Relevance and content quality still matter far more.
Will improving Core Web Vitals move my rankings?
Sometimes, and usually modestly. If your pages currently fail the thresholds and your competitors pass them, fixing speed removes a disadvantage, and sites in competitive niches do report movement. But nobody can honestly promise a specific ranking change from speed work alone, because content and links dominate. The dependable wins are in engagement and in what your existing traffic does after it arrives.
Does Google use lab data or field data for rankings?
Field data. Google's documentation is explicit that the ranking signal comes from real-user measurements collected through the Chrome User Experience Report, not from the simulated lab test that PageSpeed Insights runs on demand. That means a poor lab score with healthy field data is not a ranking problem, and a glowing lab score cannot rescue pages that real visitors experience as slow.
How fast does my site need to be for SEO?
Google's published thresholds are the sensible target: main content visible within 2.5 seconds, interaction responses under 200 milliseconds, and minimal layout shift, each measured at the 75th percentile of real visits. There is no evidence that being dramatically faster than the thresholds earns extra ranking credit, so once your pages pass, further speed work is for visitors rather than for Google.