Google has told us, in writing, that website speed affects rankings. What it hasn't done is explain it in words a business owner can act on — the official documentation reads like a physics paper. So here's the translation: the three numbers Google actually measures, the thresholds that count as passing, why most agency-built sites fail them, and how to test yours in the next two minutes without installing anything.
The three numbers, in plain words
Google's "Core Web Vitals" sound intimidating and measure mundane things:
- LCP (Largest Contentful Paint) — how long until the main content is visible. The bar: under 2.5 seconds. This is the one that correlates with people giving up.
- INP (Interaction to Next Paint) — when a visitor taps something, how fast the page reacts. Bar: under 200 milliseconds. Fails when a page is busy running scripts.
- CLS (Cumulative Layout Shift) — does the page jump around while loading (you aim at a button, it moves, you tap an ad). Bar: under 0.1. Pure annoyance metric.
Pass all three and speed stops being your problem. Fail them and you're paying twice: a ranking handicap in Google, and — the bigger invoice — visitors who leave before the page finishes. On mobile connections, every extra second of load time measurably increases abandonment; that's not a Google penalty, that's human patience.
Why most agency sites fail this
Not incompetence — economics. The standard agency stack (WordPress + page-builder theme + 20–30 plugins) is fast to build with and slow to load: typical pages ship 400–800 KB of JavaScript across dozens of files, and every plugin adds its slice. The build is profitable at €800; the loading time is your problem, not theirs, and it arrives after the invoice is paid.
You can rescue such a site up to a point — caching, image compression, plugin diets — and sometimes that's genuinely enough. The architectural fix is different: sites compiled to static pages (that's our approach — the full comparison is in WordPress vs Next.js) start life inside the thresholds instead of dieting toward them.
Test your site in 2 minutes, free
- Open pagespeed.web.dev (Google's own tool — no signup).
- Enter your URL, wait ~30 seconds.
- Read the mobile result first — it's the one Google weighs and the one that's usually worse. Look for the three metrics above and whether they're green.
Two readings matter: the "field data" up top (real visitors, if your site has enough traffic) beats the "lab score" below it. And don't chase the 100/100 vanity score — a 95 and a 88 are functionally identical. What matters is green thresholds on mobile, nothing else.
While you're there, run your two strongest competitors. Nothing motivates a speed budget like seeing who's already faster.
What actually moves the needle (in order)
If your test came back orange or red, the fixes rank like this — first ones give the most seconds per euro:
- Images. The #1 culprit on small-business sites: a 4 MB hero photo doing the job of a 150 KB one. Modern formats and correct sizing routinely halve load time on their own.
- Script weight. Every plugin, tracker, chat widget and slider ships JavaScript. Audit ruthlessly: each one should justify its milliseconds. (A lean site has a pleasant side effect: no tracking scripts, no cookie banner.)
- Hosting and caching. A €3 shared host with no caching adds a flat second to everything. Cheap fix, instant result.
- Fonts. Three font families in five weights is a wardrobe, not a design system. One family, two weights, properly loaded.
What we build ships inside the thresholds by construction — that's a property of the fixed-price packages, not an upsell.
When speed is NOT your problem
Honesty corner: if your site already passes all three thresholds on mobile, further speed work is procrastination with a technical accent — spend the effort on your offer or your Google Business Profile instead. And speed never rescues weak fundamentals: a 0.9-second page with confusing prices and stock photos converts exactly as badly, just faster. Speed removes a handicap; it doesn't create a reason to buy.
Frequently asked questions
How much does speed actually affect my Google position?
It's a real ranking signal, but a moderate one — content relevance and local factors weigh more. Think of it as a tiebreaker among comparable competitors, plus a multiplier on everything else: faster pages get crawled more, bounce less and convert better, which feeds back into rankings. Passing the thresholds is the goal; obsessing beyond them isn't.
My PageSpeed score is 60. Is that a disaster?
The 0–100 score is a lab estimate — check the three thresholds instead. A "60" that passes LCP/INP/CLS in field data is fine; a "90" that fails LCP for real visitors isn't. Mobile field data is the truth; everything else is simulation.
Can you speed up my existing site without rebuilding it?
Sometimes — image optimization, caching and script cleanup can move a WordPress site from 4 seconds to ~2.5. Whether that's smarter than rebuilding depends on how far you are from the bar and what maintenance costs you're carrying; send us your PageSpeed link on WhatsApp and we'll tell you which side of the line you're on, no charge.
The short version
Three numbers decide whether speed is your problem: LCP under 2.5 s, INP under 200 ms, CLS under 0.1 — checked free at pagespeed.web.dev, mobile tab, field data first. Most agency sites fail because of image bloat and plugin JavaScript; fixes rank images → scripts → hosting → fonts. Pass the bars and move on with your life. Fail them and fix it once, properly — our pricing includes the thresholds by construction, and the honest diagnosis of your current site costs one WhatsApp message.