Run your website through Google PageSpeed Insights and you'll likely see two very different numbers: a comfortable green score on desktop, and a red or orange one on mobile. A site that scores 92 on desktop can score 38 on mobile.
That gap isn't a glitch — it's the whole point of the test. And the mobile number is the one that matters, because 64% of global web traffic is now mobile as of mid-2025. For a local business, the phone is where most "near me" searches begin.
Here's exactly why the mobile score is so much lower, what's dragging it down, and what actually fixes it. Every figure is sourced.
Why Mobile Scores Are So Much Lower Than Desktop
Lighthouse — the engine behind PageSpeed Insights — runs two completely different tests. The desktop test assumes a fast wired connection and a capable computer. The mobile test deliberately simulates a cheap, slow phone on a weak network. Two factors do the damage:
Throttled network. The mobile test simulates "Slow 4G" — roughly 1.6 Mbps download and 150 ms of latency — while desktop runs on a fast wired connection. (Lighthouse Throttling docs)
Throttled CPU. The standard mobile test applies a CPU slowdown to mimic a mid-tier phone, where desktop applies none. A mid-range Android chip is dramatically slower at parsing and running JavaScript than a laptop. (CPU Throttling in Lighthouse, DebugBear)
The effect on the metric that punishes heavy JavaScript is enormous. Across the whole web, the median desktop page has a Total Blocking Time of 67 ms; the median mobile page is 1,209 ms — roughly 18× more blocked. (Web Almanac 2024, HTTP Archive)
So mobile isn't scoring lower because the test is unfair. It's scoring lower because it's the honest test — it measures your site on the device most of your visitors are holding. The desktop score is the flattering one you can safely ignore.
Why Builder-Built Sites Score Worst of All
If your site was built on Wix, Squarespace, or a WordPress page-builder theme, the throttled mobile test hits especially hard, because those platforms ship a lot of code to every visitor. DebugBear's October 2025 comparison ran real Lighthouse tests across the major platforms. The mobile scores:
| Builder | Mobile Score | Mobile LCP |
|---|---|---|
| Webflow | 77 | 4.95 s |
| Wix | 72 | 5.24 s |
| GoDaddy | 63 | 3.93 s |
| WordPress.com | 34 | 5.54 s |
| Squarespace | 31 | 8.79 s |
Squarespace's median mobile LCP of 8.79 seconds means the average phone visitor waits almost nine seconds for the main content to appear. Google's "good" threshold for LCP is 2.5 seconds. Even Wix, the strongest of the mainstream builders at 72, has a mobile LCP of 5.24 seconds — more than double the good mark.
Real-user data tells the same story. Across all sites measured by Google's Chrome User Experience Report, only 57% of Wix sites and 60% of Squarespace sites passed all three Core Web Vitals on mobile in 2024, and WordPress sat at 40% — versus a global average of about 51%. To the platforms' credit, every one of those numbers is trending up year over year. But "improving" and "fast" aren't the same thing, and a throttled phone test doesn't grade on a curve.
The short version: a drag-and-drop editor has to ship a general-purpose rendering engine to every visitor's phone, whether your page uses it or not.
What's Actually Dragging Your Mobile Score Down
Four things account for most of the damage. None of them are exotic — they're the default state of the modern web.
1. Too much JavaScript
JavaScript is the single biggest mobile killer because the phone has to download it, parse it, and run it — all on that throttled CPU. The median mobile page now ships 558 KB of JavaScript, and 44% of it goes completely unused. The median mobile page runs 14 "long tasks" that freeze the main thread for a combined 2,366 ms. (Web Almanac 2024 JavaScript Chapter)
2. Render-blocking resources
When a script or stylesheet in the page's <head> has to load before the browser can paint anything, the visitor stares at a blank screen. It shows up in the real-user numbers: only 51% of mobile pages hit a "good" First Contentful Paint, versus 68% on desktop. (Web Almanac 2024 Performance Chapter)
3. Heavy, unoptimized images
The median mobile page weighs 2.3 MB, up 357% over the last decade, and images are the largest slice. A hero photo exported straight from a phone or a stock library at full resolution is often the single element Lighthouse flags as your LCP — the thing the visitor is waiting for. (Web Almanac 2024 Page Weight Chapter)
4. Third-party scripts
Live chat widgets, booking embeds, analytics, review badges, font loaders, ad pixels — each adds code you don't control. Over 90% of pages load at least one third-party resource, and at the heavy end third-party JavaScript alone runs to 1,292 KB. (Web Almanac 2024 JavaScript Chapter) On a builder site you often can't remove these even if you wanted to.
Why a Bad Mobile Score Costs You Real Customers
A low PageSpeed number isn't a vanity metric. On mobile, speed maps directly to whether a potential customer stays or bounces — and for a local business, that visitor was often about to call or walk in.
- Slow pages get abandoned. Google's widely cited 2016 study found 53% of mobile visits are abandoned if a page takes longer than 3 seconds to load. (It's an old figure and networks are faster now, but the direction hasn't changed.)
- Every second cuts conversions. Portent's 2022 analysis found a page that loads in 1 second converts at 3.05%, versus 1.12% at 3 seconds — a 1-second site converts about 2.5× better than a 5-second one.
- Even tiny gains matter. The Google-commissioned Deloitte study "Milliseconds Make Millions" found a 0.1-second mobile speed improvement lifted retail conversions 8.4% and lead-gen form submissions 21.6%.
For a local business this is sharper than for an online store, because local mobile searches are high-intent. Google's data has long held that 76% of people who search for something nearby on a smartphone visit a related business within a day. If your site takes nine seconds to paint on their phone, a chunk of that ready-to-buy traffic is gone before they ever see your hours or your phone number.
Mobile PageSpeed Is Also a Ranking Factor
Speed doesn't only lose the visitors who arrive — it can reduce how many arrive in the first place. Two things make the mobile score a search-ranking concern, not just a UX one:
- Mobile-first indexing. Google has moved to mobile-first indexing for the web — it crawls and ranks the mobile version of your site, not the desktop one. The mobile version is the version Google sees.
- Core Web Vitals feed ranking. Google says Core Web Vitals "aligns with what our core ranking systems seek to reward" as part of the page-experience signal — a tiebreaker between pages of comparable quality.
For a local business competing for "[service] near me," where several competitors offer similar content, a faster mobile site is exactly the kind of tiebreaker that decides who shows up first.
What Actually Fixes a Low Mobile Score
The fixes fall into two buckets: things you can do on your existing site, and the structural fix.
On your existing site
- Compress and resize every image. Serve images at the size they actually display, in a modern format like WebP. This alone often moves the needle most, because images are usually the LCP element.
-
Defer or remove non-critical JavaScript. Add
deferto scripts that don't need to run before paint, and delete the ones you don't use. Remember: 44% of mobile JS goes unused. - Audit your third-party scripts. Every chat widget, pixel, and embed has a cost. Keep the ones that earn their weight; cut the rest.
- Self-host fonts and limit weights. Loading fonts from a third party adds a render-blocking round trip; self-hosting two weights instead of six is faster.
- Test on mobile, not desktop. Always read the mobile tab in PageSpeed Insights. The desktop score is the one lying to you.
The structural fix
There's a ceiling to what you can tune on a builder site, because the platform's own overhead is baked in — you can optimize your images and still be shipping a rendering engine you never use. The structural fix is to serve the visitor's phone exactly what the page needs and nothing more.
That's what a hand-coded static one-page site does. No general-purpose editor runtime, no plugin stack, no unused JavaScript — just the HTML, the minimal CSS, and the images for your page. It's why a static site can score in the high 90s on the same throttled mobile test that leaves the mainstream builders anywhere from the low 30s (Squarespace) to the low 70s (Wix).
The Short Version
- Mobile scores lower than desktop on purpose — the test simulates a slow, throttled phone, which is what 60% of your visitors use.
- Builder sites score worst because they ship code to every phone: Squarespace's median mobile LCP is 8.79 s, Wix's 5.24 s, against a 2.5 s "good" mark.
- The usual culprits are excess JavaScript, render-blocking resources, heavy images, and third-party scripts.
- It costs customers (53% bounce past 3 seconds) and rankings (mobile-first indexing + Core Web Vitals).
- Tuning helps; a hand-coded static site removes the ceiling entirely.
How to Check Your Own Score
- Go to pagespeed.web.dev
- Paste your website URL
- Select the Mobile tab — that's the one that matters for rankings
- Click "Analyze"
If you're on Wix, Squarespace, or WordPress and the mobile score is below 75, there's a hard ceiling on what you can fix from inside the dashboard. You can compress images and cut widgets, but the editor infrastructure and heavy frameworks load for every visitor no matter what you change.
Stefan Gabos is a senior web developer with 15+ years of experience building fast, accessible websites. He runs PageDrop — affordable one-page websites for small businesses, built with clean code and no monthly fees.
Top comments (0)