My Shopify store scores 34 on mobile PageSpeed but 92 on desktop, what's going on?
A 30 to 60 point gap between mobile and desktop PageSpeed scores is normal, not proof your store is broken. Here is what actually drives it and how to fix what is fixable.
What's going on
You run your store through Google PageSpeed Insights and see a mobile score in the 30s while desktop sits in the 90s, same theme, same page, same content, wildly different numbers. It's tempting to assume something is broken, especially when the mobile score barely moves no matter what you tweak.
In most cases nothing is actually broken. PageSpeed Insights runs Google Lighthouse under very different simulated conditions for each device type: the mobile test throttles the CPU and simulates a slower network connection, while the desktop test runs largely unthrottled. That difference alone accounts for a large chunk of the gap, because anything CPU-heavy, such as third-party app scripts, large JavaScript bundles, or unoptimized images, costs far more time under mobile throttling than it does on a fast desktop connection.
That said, a very wide gap like 34 versus 92 usually does mean there is real room to improve, not just a quirk of the test. The fix is to separate the test's built-in pessimism from genuinely fixable problems, and to check what your real visitors are actually experiencing rather than relying on one lab score.
Why it happens
Lighthouse's mobile emulation profile intentionally simulates a mid-tier phone with constrained CPU and network conditions, while its desktop profile does not apply the same throttling. This means identical page weight and script execution time translate into very different scores by design. It is testing two different user experiences on purpose, not grading the same page twice.
On top of that baseline gap, Shopify stores accumulate real mobile-specific weight over time. Every installed app can add its own JavaScript through script tags or theme app embeds, hero and product images are often uploaded at full camera resolution, and custom theme edits sometimes skip responsive image markup. A desktop CPU and connection can absorb that overhead without much visible slowdown, but a throttled mobile simulation cannot, so the same page-weight problems show up as a much lower score on mobile.
5 ways to fix it
Understand that PageSpeed Insights runs two different tests, not one score split in half
Google Lighthouse, the engine behind PageSpeed Insights, throttles the mobile test to simulate a mid-tier phone on a slower connection, applying a CPU slowdown on top of added network latency. The desktop test runs with a fast connection and little to no throttling. A gap of 30 to 60 points between the two, such as 34 versus 92, is common and does not mean desktop is fine while mobile is broken. It means the mobile test is surfacing render-blocking scripts, oversized images, and slow third-party tags that a fast desktop connection was masking. Treat the two scores as separate diagnostics and prioritize mobile, since it reflects the harsher conditions most real shoppers browse under.
Check Shopify's own Web Performance report before trusting a single Lighthouse run
In Shopify admin, under Online Store, the Web Performance report uses real-user data collected from actual visitors over a recent trailing period, not one simulated test run. It scores Core Web Vitals such as Largest Contentful Paint, Cumulative Layout Shift, and Interaction to Next Paint, broken out by device, and shows how theme or app changes have shifted real results over time. This is a more trustworthy signal than a single Lighthouse score because it reflects your actual traffic and device mix rather than one lab snapshot from one location and device, and real-user data of this kind also carries more weight for search ranking purposes than lab scores do.
Audit and trim installed apps and theme app embeds first, since they are usually the biggest mobile cost
Every app that injects a script or theme app embed adds JavaScript that a throttled mobile CPU has to parse and run far more slowly than a desktop CPU does. In Shopify admin, go to Online Store, then Themes, then Customize, and open the app embeds section to see and toggle off embeds you are not actively using. Review widgets, upsell popups, chat launchers, and tracking pixels are common offenders. Also check for legacy tracking snippets pasted directly into theme settings. Removing or disabling even two or three unused apps often recovers more mobile score than any single theme tweak.
Compress and properly size images, and let Shopify serve them responsively
Oversized hero images and product photos are consistently the top Largest Contentful Paint offender on mobile. Shopify's CDN can automatically serve resized, compressed versions of images when your theme requests them with appropriate width parameters and responsive markup, but themes with hardcoded image tags or apps that inject full-resolution images can bypass this. Re-upload oversized source files so hero and banner images are as small as possible without visible quality loss, confirm your theme's image tags use responsive sizing, and make sure images have explicit width and height so the browser does not reflow the layout while they load, which also helps your layout shift score.
Get a developer to run a mobile-specific trace and fix the top few flagged items, rather than chasing the number itself
PageSpeed Insights ranks its diagnostics and opportunities by estimated time savings, usually pointing at render-blocking resources, unused JavaScript or CSS, and font loading. Fixing the top two or three items, such as deferring a non-critical script or changing how a custom font loads, typically moves the score more than a dozen minor tweaks. If your theme code itself is the bottleneck rather than installed apps, this is worth paying a developer for a targeted look, since blind edits to theme code carry real risk of breaking something.
Bottom line
A 34 versus 92 split usually is not a bug in your store, it is Lighthouse's mobile throttling doing its job and surfacing app and image bloat that desktop hides. Start with Shopify's own Web Performance report to see how real visitors on real devices actually experience your site, then work down your app and image list before touching theme code. If mobile scores stay poor after trimming apps and images, that is a signal to bring in a developer for a targeted performance pass rather than installing a general speed booster app, since most mobile score problems trace back to app and asset choices you already control directly.
Browse the rest of the problem library, run a free storefront scan to catch issues like this automatically, or email us and we'll work out a custom solution for it.