What's a good PageSpeed Insights score?

90 or higher on mobile is the green band and the practical target. Above 80 is workable. Below 70 means real customer friction and likely ranking impact. The headline score should be looked at alongside the Core Web Vitals pass or fail.

Why is my mobile score so much lower than my desktop one?

PageSpeed Insights simulates a mid-range Android phone on a 4G connection for mobile, and a fast computer on broadband for desktop. The same site will always score lower on mobile because the test conditions are harder. Mobile is what Google uses for ranking, so it's the score to focus on.

Do PageSpeed Insights scores actually affect Google rankings?

Indirectly, yes. The headline score isn't a direct ranking factor, but the Core Web Vitals (LCP, INP, CLS) are. PSI is the easiest way to see how your CWVs are performing. Passing all three is part of Google's ranking signals.

My PSI report has 20 issues listed. Where do I start?

At the top. The opportunities are roughly sorted by impact. Fix the first 2 or 3 (usually image optimisation, unused JavaScript, and server response time), then re-test. Most sites can recover 15 to 25 points by addressing the top three items.

What Google PageSpeed Insights Actually Measures

PageSpeed Insights is Google's free website speed tool. Paste your URL, click run, get a report. For most small business owners, that report is intimidating. Green and red bars, words like "Largest Contentful Paint" and "Cumulative Layout Shift," numbers in milliseconds. Most people just look at the headline score, then either celebrate or panic. This post is the plain-language version of what's actually in that report, what matters, and what doesn't. By the end of it you should be able to read your own PageSpeed report and know what to focus on. Run pagespeed.web.dev with your homepage URL in another tab if you want to follow along.

What PageSpeed Insights Actually Tests

The tool does two things:

**Field data:** real performance from real Chrome users who visited your site recently (when there's enough traffic)
**Lab data:** a single simulated test that PageSpeed runs right now from Google's servers

Field data matters more, because it reflects what your actual customers experience. Lab data is useful for debugging because it's reproducible.

For most small business sites, field data isn't available because there aren't enough visits to generate the sample. In that case, you only see lab data, and that's still useful.

The headline score is calculated from lab data. The Core Web Vitals pass or fail above it comes from field data when available.

The Three Core Web Vitals

The three metrics Google actually weights for ranking.

**Largest Contentful Paint (LCP)**

What it measures: how quickly the main visible content of your page appears.

The threshold: under 2.5 seconds is good. 2.5 to 4.0 is "needs improvement." Above 4.0 is poor.

What slows it down: large images, slow servers, render-blocking scripts.

**Interaction to Next Paint (INP)**

What it measures: how quickly the page responds when users tap or click.

The threshold: under 200 milliseconds is good. 200 to 500 is "needs improvement." Above 500 is poor.

What slows it down: heavy JavaScript, long-running scripts blocking the main thread.

**Cumulative Layout Shift (CLS)**

What it measures: how much content jumps around as the page loads.

The threshold: under 0.1 is good. 0.1 to 0.25 is "needs improvement." Above 0.25 is poor.

What causes it: images without specified dimensions, fonts swapping mid-load, ads or embeds appearing late.

If all three are green, your Core Web Vitals are passing. Google says this affects ranking.

The Headline Score

Above the Core Web Vitals, you see a big number (the PageSpeed score). It runs from 0 to 100.

90 to 100: green, "fast"
50 to 89: amber, "moderate"
0 to 49: red, "slow"

This score is a weighted blend of several lab metrics: LCP, Total Blocking Time, CLS, First Contentful Paint, and Speed Index. Google's algorithm for the score has changed over the years.

Some details to understand:

The score is not the same as Core Web Vitals. You can have a 75 score with passing Core Web Vitals, or a 90 score with one failing CWV.
The score is more sensitive to small changes than CWV bands. A 200ms speed improvement can shift the score by 10 points without changing the green/amber/red CWV outcome.
Mobile scores are almost always lower than desktop scores. Mobile testing simulates a slower CPU and 4G network.

For ranking and customer experience, Core Web Vitals matter more than the headline score. But the headline score is easier to talk about and track.

Lab vs Field Data

The two halves of your report. Understand the difference.

**Field data (Chrome User Experience Report):**

Based on real Chrome visits to your site over the previous 28 days
Only shows if your site has enough traffic (typically 1,000+ visits a month)
Shows the 75th percentile experience (most visitors had this or better)
This is what Google uses for ranking

**Lab data:**

A single simulated load from Google's testing infrastructure
Always available, regardless of traffic
Useful for debugging because it's repeatable
Not directly used for ranking, but correlates strongly with real-world performance

For sites with low traffic, you'll only see lab data. That's still actionable. Fix the lab issues and your real-world performance usually follows.

If you have field data showing all-green CWVs, you're in good shape regardless of the headline score.

Reading the Diagnostics

Below the headline numbers, PageSpeed Insights lists specific opportunities and diagnostics:

**"Reduce unused JavaScript"** (almost always present, usually fixable)
**"Eliminate render-blocking resources"** (deferring CSS or JS that doesn't need to load first)
**"Serve images in next-gen formats"** (use WebP instead of JPEG)
**"Properly size images"** (don't load a 3000px image to display at 300px)
**"Reduce server response times"** (TTFB, often a hosting issue)
**"Avoid enormous network payloads"** (page is too heavy overall)
**"Use efficient cache policy on static assets"** (browser caching settings)
**"Avoid layout shifts"** (explicitly size images and ad slots)

Each item shows the estimated time you'd save by fixing it. Items at the top of the list usually have the biggest impact.

Don't try to fix everything at once. Pick the top 2 or 3 and address them. Re-test. Then pick the next 2 or 3.

Mobile vs Desktop

The default view is desktop. Switch to mobile (the toggle at the top), because mobile is usually:

Where most small business traffic comes from (typically 60 to 75 percent of UK small business visits)
Where the worst scores hit (mobile testing simulates a mid-range Android on 4G)
Where Google primarily ranks you (mobile-first indexing has been default since 2019)

If your desktop score is 95 and your mobile is 52, your real ranking signal is the mobile score. Fix that, and the desktop will look after itself.

Only worry about desktop if your audience is genuinely desktop-heavy. For most small businesses, mobile is the score that matters.

What "Average" Looks Like in 2026

For UK small business websites I audit in 2026:

Median WordPress site: PSI mobile score of 52, 1 to 2 Core Web Vitals failing
Median Wix or Squarespace site: PSI mobile score of 65, often passing CWVs
Well-built hand-coded site: PSI mobile score of 92+, all CWVs passing comfortably

If you're in the median band, you're behind the curve enough that Google notices. If you're in the green band, you're outranking 75 percent of your competition on speed alone.

For background on why this matters commercially, see why website speed matters more than ever.

What Customers Actually Feel

PageSpeed scores measure things customers experience. The translation:

LCP affects how long the customer waits to see your headline
INP affects whether tapping a button feels instant or laggy
CLS affects whether things stay still while they're trying to read

A site with a perfect score feels snappy and confident. A site with poor scores feels broken, even if it eventually loads. Customers don't usually talk about "your LCP." They talk about your site feeling slow or annoying.

PageSpeed Insights is the closest thing to a measurable proxy for how your site feels to use.

A Quick Action Plan

If you've just run PageSpeed Insights and you're below 70 on mobile:

Check whether field data exists and which CWVs are failing
Look at the top 3 opportunities in the diagnostics
Pick the one with the biggest time saving and fix that first
Re-test, measure the change
Repeat with the next opportunity
If you can't get above 70 after 3 rounds, the foundations probably need rebuilding

Want help interpreting your PageSpeed report or improving it? Contact me for a free 15-minute review.