Why Website Speed Isn't Magic

Chasing a PageSpeed score is one of the more effective ways to spend time on something that doesn't move the needle. Here's what actually matters.

The PageSpeed score is not what Google measures.

At some point someone told every small business owner that if you fix your Google PageSpeed score, your rankings improve. The score is a useful diagnostic tool — but it's a lab measurement, simulating your page on a throttled connection with a mid-range device. Google uses real-world Chrome User Experience (CrUX) data from actual visitors to assess Core Web Vitals. A page can score 45 in PageSpeed Insights and still pass Core Web Vitals in the field if real users on good connections have fast experiences. Chasing a high lab score isn't the goal — passing the field thresholds is.

Lab data vs field data

PageSpeed Insights shows two sets of data: lab data (simulated, from Lighthouse) and field data (real, from CrUX). The field data — the coloured bar at the top — is what Google actually uses for ranking signals. A green bar in field data with an average lab score is fine. A red bar in field data with a perfect lab score is a problem.

The three metrics that actually matter.

Core Web Vitals are the specific signals Google uses for page experience ranking. Not the overall score — these three:

< 2.5s

LCP

Largest Contentful Paint

How fast the main content loads

< 200ms

INP

Interaction to Next Paint

How fast the page responds to input

< 0.1

CLS

Cumulative Layout Shift

How much the layout jumps during load

Speed is a ranking signal. A weak one.

Google confirmed Core Web Vitals as a ranking factor in 2021. What this means in practice: page experience signals are used as a tiebreaker when content quality is roughly equal. If your page is significantly slower than all competing pages targeting the same keyword, it can hurt you. But fixing a slow page will not cause a slow-loading site with weak content to outrank a fast site with excellent content. Speed is a floor, not a ceiling.

Common mistake

Spending several hundred dollars to get a PageSpeed score from 58 to 79, then expecting a ranking jump that doesn't come. If you're passing Core Web Vitals in the field, the marginal gains from further score optimisation are unlikely to outweigh investing that effort in content or links.

Where speed actually does matter: conversions.

The strongest argument for a fast site isn't rankings — it's that slow sites leak leads. Pages taking more than 3 seconds to load on mobile lose a significant portion of visitors before they ever see the content. For a local service business, that means potential customers bouncing before they see your offer, call your number, or fill out your form. The conversion impact is immediate and direct. The SEO impact is real but indirect — via engagement signals, dwell time, and reduced bounce rates.

What to actually do about speed.

If your site is failing Core Web Vitals in the field — fix it, starting with LCP (usually the biggest culprit). Image optimisation, lazy loading, and caching are legitimate improvements. Switching to a faster hosting setup makes a real difference for sites on slow shared hosting. Check your field data first in Google PageSpeed Insights — if you're already passing, put that effort into content quality and link acquisition instead.

Check before you fix

Most speed work gets done because someone showed a client a red PageSpeed score. That's not always the right trigger. Here's the right order:

Check your field data

Go to Google PageSpeed Insights, enter your URL, and look at the CrUX field data at the top — not the lab score. Are you passing, failing, or 'needs improvement' on LCP, INP, and CLS? That's the only number that actually matters for rankings.

Act on what you find

If failing — fix LCP first (usually the largest image or hero element load time), then INP, then CLS. If passing — move on. Invest the next hour in content or outreach instead. Every site we build passes Core Web Vitals in the field by default, because it's baked into how we build — not retrofitted after launch.

Key takeaways

PageSpeed Insights score is a lab measurement — field data (CrUX) is what Google uses
Core Web Vitals (LCP, INP, CLS) are the specific metrics that affect rankings, not the overall score
Speed is a ranking tiebreaker — strong content on a passing site outranks great speed on thin content
Slow sites hurt conversions more immediately and directly than they hurt rankings
If you pass Core Web Vitals in the field, invest effort in content and links first

Keep reading