Why is your mobile PageSpeed score lower than desktop?

Key takeaways

A mobile PageSpeed score lower than desktop is normal, because the mobile test deliberately simulates a slower phone on a slower connection. A gap of 20 or 30 points does not mean the site is broken. What matters is whether real visitors on phones get a fast page, which a different part of the same report shows.

• The mobile test simulates a mid-tier phone on a throttled mobile network.
• Scores of 90 or more are good, 50 to 89 need improvement, under 50 are poor.
• Field data from real users over 28 days matters more than the lab score.
• Images, scripts and fonts cause most of the mobile gap.

Why are the two scores different?

The scores differ because the tests differ. PageSpeed Insights runs its mobile lab test by simulating a mid-tier phone on a mobile network, and its desktop test by simulating a desktop computer on a wired connection. The same page, with the same images and scripts, takes longer on the slower device and connection.

That is the point of the mobile test. It represents a visitor with an ordinary handset and ordinary signal, not you in the office on Wi-Fi with a new phone. A page that scores 95 on desktop and 60 on mobile is telling you it depends on power and bandwidth that many visitors do not have.

• Why mobile-friendly web design is essential for UK businesses
• How fast a business website should load

Which number should you trust?

Trust the field data at the top of the report over the score. PageSpeed Insights shows two kinds of result. Field data comes from real Chrome users visiting your page over the previous 28 days. Lab data is a single simulated load. The Core Web Vitals assessment, which is what Google's ranking systems use, is based on the field data.

The assessment passes when the 75th percentile of Largest Contentful Paint, Interaction to Next Paint and Cumulative Layout Shift are all in the good range. A page can have a lab score of 65 and pass, or a score of 92 and fail. Small sites may see no field data at all, because there are too few visits to report. In that case the lab result is your only guide.

• Field data: real visitors, 28 days, what Google uses. Read this first.
• Lab data: one simulated visit, useful for diagnosing and testing fixes.
• Score: a summary of lab metrics, useful for trend, not a target in itself.

• Core Web Vitals for business websites

What makes mobile slower?

Three things: bytes, scripts and render-blocking resources. Phones download more slowly on mobile data, so large files hurt more. Phone processors are slower, so JavaScript that a laptop handles in a blink can freeze a phone for a second or more. And anything that must load before the page can draw delays the first screen.

• A large hero image or slider served at desktop size to phones.
• Too many scripts: chat widgets, tracking tags, pop-up tools, social embeds.
• Page builders and heavy themes that load code the page does not use.
• Web fonts that block text from showing until they arrive.
• Video or map embeds that load on arrival instead of on tap.
• Slow hosting, which delays everything else.

• Is your page builder slowing your website down?
• Carousels on mobile websites: keep, fix or remove?

Which fixes move the mobile result most?

Fix the largest element on the first screen, then remove scripts. On most business sites, the Largest Contentful Paint element is the hero image. Serving it at the right size for a phone, compressed and in a modern format, is often the single biggest gain. After that, every script you can remove or delay helps responsiveness.

• Serve responsive images so phones get a smaller file.
• Preload the main image and avoid lazy-loading it.
• Remove plugins and tags nobody uses or can explain.
• Delay chat widgets and embeds until the visitor scrolls or taps.
• Reserve space for images and adverts to stop layout shift.
• Use caching and a content delivery network through your host.

• Optimise website images for speed and search
• Website hosting for UK small businesses

What does it cost, and when should you stop?

Most mobile speed work on an existing site falls within an indicative £500 to £3,000: image handling, script clean-up, caching and font loading. If the theme or page builder is the cause, optimisation reaches a ceiling, and the choice becomes living with a mediocre result or rebuilding on a lighter foundation.

Stop when the Core Web Vitals assessment passes on mobile and the page feels quick on a real phone. Chasing a score of 100 is poor value. The last ten points usually cost more than the first forty and make no difference a visitor would notice. Removing a chat tool that brings in leads to gain three points is a bad trade.

• Slow website: fix it or rebuild it?
• What it costs to make a website mobile-friendly in the UK

Why do scores change between runs?

Lab scores vary from run to run because each test is a single load affected by server response, network conditions and third-party scripts at that moment. A swing of five points either way is ordinary. Run the test three times and take the middle result before concluding that a change helped or hurt.

Test the pages that matter, not only the homepage. Service pages, landing pages and product pages often have different templates and different problems. A fast homepage tells you little about the page your adverts point to.

What does this look like in practice?

A pattern we see with owners who have just run PageSpeed Insights: the desktop score is in the 90s, the mobile score is in the 40s, and the web designer says mobile scores are always low and can be ignored. The first part is partly true. The second is not.

Looking closer, the hero is a 2,400-pixel-wide image sent to every phone, and four marketing scripts load before anything is drawn. Fixing the image and deferring the scripts lifts the mobile result substantially in an afternoon. Global Bridge Labs (GBL) reports field data first and the score second, so the conversation is about visitors, not points.

• See client work
• Website Design & Development for UK businesses
• Make your website work properly on mobile

Next step

Run your top landing page through PageSpeed Insights and read the Core Web Vitals assessment on the mobile tab before you look at the score. If it fails, the first item in the diagnostics list is where to start.