Web · 5 min read · October 2026

What your PageSpeed score really means

Someone runs your website through Google PageSpeed Insights, sees a red score of 38 on mobile, and sends you a screenshot with an offer to fix it. Before you panic or pay, it helps to know what the score measures, which parts matter for your customers and which are noise.

Here is how to read a PageSpeed report, what the main numbers mean in plain language, why mobile scores are almost always lower, and which improvements are usually worth paying for on a small business site.

Two kinds of data on one page

PageSpeed Insights can show two things. Field data, labelled as what real users experienced, comes from actual Chrome users visiting your site over recent weeks. Lab data is a single simulated test run on a throttled mobile connection. Small sites often have no field data because they do not get enough visitors, so only the lab test appears.

Field data reflects your real customers. The lab score is useful for diagnosis, but it is one simulated visit, and it varies from run to run.

The numbers that matter

Largest Contentful Paint (LCP) is how long the main content, usually the big image or heading, takes to appear. Interaction to Next Paint (INP) is how quickly the page responds when someone taps. Cumulative Layout Shift (CLS) is how much things jump around while loading. Google groups these as Core Web Vitals and publishes thresholds for good, needs improvement and poor.

For a small business site, LCP is usually the one to watch. If the first screen appears quickly on mobile, most visitors are happy.

Why mobile scores are lower

The mobile test simulates a mid-range phone on a slow connection. The same page that scores 95 on desktop can score 50 on mobile. That is expected. Do not compare your mobile score with a competitor's desktop score, and do not chase 100 on mobile; well-built sites often score in the 80s or 90s, and many perfectly usable sites score lower.

Run the test three times and look at the typical result. Single runs vary.

Fixes that usually matter

Oversized images are the most common problem, and the easiest to fix: resize, compress and use modern formats. Too many plugins or scripts on WordPress, especially sliders, page builders and chat widgets, slow pages noticeably. Slow or overloaded hosting delays everything. Large web fonts loaded in many weights add up. Fixing these usually produces a visible improvement for real visitors.

Fixes that often do not

The report lists many technical suggestions, and some offer tiny gains for a lot of work. Shaving a few milliseconds off a script on a site that already loads in two seconds rarely changes enquiries. Prioritise what makes the first screen appear faster and the page respond quickly. Ignore cosmetic score chasing.

Third-party tools cost speed

Chat widgets, social media feeds, embedded maps, video embeds, tracking pixels and review widgets all load code from other companies' servers. Each one may be worth having, but each slows the page. Add them deliberately, load them only on the pages where they are needed, and remove the ones nobody uses.

Speed and search rankings

Google uses page experience signals, including Core Web Vitals, as part of its ranking systems, but relevance and content quality matter far more. A slow page with exactly what a searcher wants can still rank well. The bigger cost of a slow site is visitors who leave before it loads, especially on mobile data.

Test the way customers visit

The most honest speed test is your own phone on mobile data, opening the site from a Google search or a WhatsApp link. If the first screen appears in two or three seconds and buttons respond immediately, the site is fast enough for most purposes, whatever the score says.

Fix what customers feel, not what the score colours red.

Be wary of cold emails offering to "fix your score" for a fee. Ask what specifically they will change and what difference it will make for visitors.

Reading your report in five minutes

Check for field data. If present, it matters most.
Look at LCP on mobile. Is the main content appearing quickly?
Scan the diagnostics. Images, unused scripts, server response time.
Run it three times. Use the typical result, not the worst.
Test on your phone. Mobile data, from a search result.

Fix the top one or two issues, then test again. Most small sites get the biggest improvement from images and hosting alone.

Quick answers

What is a good PageSpeed Insights score?

Scores of 90 or above are labelled good, but many well-built sites score lower on mobile. Focus on real-user Core Web Vitals and how quickly the first screen loads on a phone.

Why is my mobile PageSpeed score lower than desktop?

The mobile test simulates a mid-range phone on a slow connection, so scores are almost always lower than desktop. It is expected.

Does PageSpeed score affect Google ranking?

Page experience signals, including Core Web Vitals, are part of Google's ranking systems, but relevance and content quality matter much more. Speed matters most for keeping visitors.

Fast where it counts.

I build websites that load their first screen quickly on mobile data, with compressed images and only the scripts they need. Send me your site and I'll tell you which fixes would make a real difference.

Send a brief → +91 70102 66975 whatsapp