I ran Lighthouse on 269 Shopify stores. The median mobile score was 48.

I spent a July weekend running the Google PageSpeed Insights API against 269 live Shopify and DTC storefronts, one mobile audit each. I expected mediocre scores. I did not expect the biggest brands to be the worst offenders.

Here is what the data says, and what almost every store gets wrong.

The headline numbers

• Median mobile Performance score: 48 (out of 100)
• 53.4% of stores score under 50. Only 9.4% hit 90 or above.
• Median Largest Contentful Paint: 10.0 seconds. Google's "good" threshold is 2.5s. 88.3% of stores are in the Poor bucket on LCP.
• Median SEO score: 92. The same stores that fail performance ace the SEO checklist.

That last pair is the story, I think. Stores have clearly invested in titles, meta descriptions, and structured data, and then lost (or never fought) the speed battle. They optimize for the crawler and lose the customer.

Not one recognizable brand cracked 50

I seeded the list with brands you know. Every single one scored under 50 on mobile:

Steve Madden's product-listing page takes 23 seconds to render its largest element on a throttled mobile connection. These are companies with real engineering budgets.

What actually fails (ranked by how many stores fail it)

• Network dependency tree: 93.7%
• First Contentful Paint: 91.4%
• Efficient cache lifetimes: 91.4%
• Largest Contentful Paint: 91.1%
• Render-blocking requests: 90.7%
• Reduce unused JavaScript: 85.5%
• Reduce unused CSS: 83.3%

No exotic problems in that list. It is the same four causes over and over: app scripts stacked on app scripts, render-blocking theme CSS/JS, oversized hero images, and long critical-request chains. If you run a Shopify store, your failing audits are almost certainly these exact ones.

One genuinely useful nuance: layout stability is mostly fine (82.2% of stores have good CLS). The popular advice to obsess over layout shift is solving a problem most stores do not have. The problem is loading speed, specifically whatever blocks the main image and the main thread.

Methodology, briefly

PageSpeed Insights API (Lighthouse 13.4), mobile strategy, one run per store, July 2026. Store list from the public TeamDukaan/performance dataset seeded with recognizable brands, so read it as a broad cross-section of real Shopify/DTC storefronts, not a global top-269 ranking. Lab scores, standard Good/Needs-improvement/Poor thresholds.

The full writeup with charts, the complete failed-audit ranking, and per-audit fix guides is here: Mobile Lighthouse performance across 269 Shopify stores. The raw dataset is a downloadable CSV, CC BY 4.0, so you can rerun or extend the analysis.

The question I am left with

Is this a Shopify problem or an ecommerce problem? My hunch is the latter: the app-store economy (every tool is one more render-blocking script tag) exists on every platform. But I only measured Shopify/DTC stores, so I genuinely do not know.

If you run or build Shopify stores: does the 10-second LCP match what you see in the field? And is it the theme, the app stack, or the image pipeline in your experience? Curious what people fighting this daily would say.