Mobile Speed vs Desktop Speed: Why the Gap Matters

Run any website through PageSpeed Insights and you'll usually see two different scores — one for mobile, one for desktop. The desktop score is almost always significantly higher. That gap isn't a bug. It's the most important thing the report tells you.

Here's why mobile scores are lower, what that gap means, and which number actually matters for SEO.

Why mobile scores are always lower

Lighthouse runs mobile audits with deliberately throttled conditions:

These settings approximate the median real user on mobile — not the best case, not the worst. The desktop test, by contrast, simulates a developer-class machine on broadband.

The result: the same site will load fast on desktop because there's plenty of CPU and bandwidth, and slowly on mobile because there isn't. The gap is the difference between your dev machine and the real world.

A typical gap

A 15–30 point gap is normal. A 40+ point gap is a sign of too much JavaScript and too many images for mobile bandwidth.

Which one matters more?

Mobile, by a wide margin.

Three reasons:

1. Mobile-first indexing. Since 2019, Google uses the mobile version of your site for ranking and indexing. The mobile audit is closer to what Googlebot sees.
2. Real traffic. Most websites get 50–70% of their traffic from mobile devices. If your mobile experience is bad, most of your visitors are having a bad experience.
3. Stricter Core Web Vitals thresholds in practice. Mobile devices struggle more with LCP and INP than desktops, so failing CWV is more likely on mobile and that's what Google ranks against.

A 95 desktop / 55 mobile site is functionally a slow site to Google and to most of its users.

How to close the gap

If your gap is wide, the issue is almost always one of these:

Too much JavaScript

Mobile CPUs take 3–5× longer to parse and execute the same JS bundle. Cut what you don't need. See how JavaScript slows down websites.

Images sized for desktop

A 2000px-wide hero image is overkill on a 360px-wide phone. Use srcset to serve smaller versions on mobile:

<img
  srcset="/hero-800.webp 800w, /hero-1600.webp 1600w"
  sizes="(max-width: 768px) 100vw, 800px"
  src="/hero-800.webp"
  width="800"
  height="400"
  alt="..."
>

Render-blocking resources

The same CSS file that loads in 50ms on desktop takes 500ms on simulated 4G. Inline critical CSS so mobile users see content immediately.

Third-party scripts

A 200KB analytics script costs barely 50ms on desktop and 400ms on mobile. Defer everything non-essential.

Variance in mobile scores

Mobile scores fluctuate more than desktop scores because:

• Simulated CPU throttling introduces randomness
• Network conditions are emulated with packet loss simulation
• Lighthouse runs are inherently single-sample lab tests

A site that scores 78 might score 73 or 82 on consecutive runs. Always average 3–5 runs for a reliable number, especially if you're making small optimizations.

For real-world data instead of lab data, look at the field data section in PageSpeed Insights — actual measurements from real Chrome users over the last 28 days.

When desktop score still matters

A high desktop score isn't useless:

• B2B SaaS with desktop-heavy users still benefits from desktop optimization.
• Internal tools and dashboards used on laptops should be fast on desktop.
• Long-form content (research, technical writing) tends to be read on desktop.

But for almost any consumer-facing site, mobile is the number you optimize for.

How to check both

PageSpeed Insights shows both by default. RateMySite.io shows them side-by-side with a weighted "overall" grade that accounts for both. Chrome DevTools lets you simulate mobile audits locally with the same throttling Lighthouse uses.

The bottom line

Mobile scores are lower because mobile conditions are harder. Google uses mobile for ranking. Most of your real users are on mobile. Your job is to make mobile fast — and the desktop score will take care of itself.

For a fix-it action plan, read how to improve your PageSpeed Insights score.