Every number on this site comes from the same setup, described here in enough detail that you could run it yourself and get the same answer. If a page shows a figure that is not produced this way, that is a mistake and we want to hear about it.
The test store
One Shopify development store, used for nothing else. Twelve products, four images each, every image the same source file at 2048×2048 and the same file size, one collection containing all twelve. No apps installed at any point. Password protection disabled so the measurement tool can reach the pages.
Every theme is measured on the same product page of that same store, in the theme’s default preset with no sections added or removed and no hero video. The only thing that changes between two runs is the theme.
How speed is measured
PageSpeed Insights, mobile tab, five runs per theme, median reported rather than best. We use PageSpeed Insights rather than a local tool for one reason: you can re-run it on the same URL and check us. It runs Lighthouse on Google’s hardware with a mid-range Android profile and throttled mobile data, which is closer to your actual customer than a developer’s laptop on office wifi.
Three figures are recorded. LCP (largest contentful paint) is how long until the main product image is on screen. CLS (cumulative layout shift) is how much the page jumps while loading. Page weight is the total bytes transferred, taken from the same report’s network payload audit.
What the numbers do not tell you
A theme measured in default configuration is a floor, not a forecast. Your store will be slower than these figures once you add apps, a chat widget, tracking pixels and your own images — often by more than the gap between the fastest and slowest theme in the table. Treat the ranking as a way to avoid starting from a bad place, not as a prediction of what your store will score.
How prices are checked
Read from the vendor’s own store listing, with the date recorded on the page. Where a price differs between a vendor’s site and the Shopify Theme Store or App Store, we use the store listing and say so.
How scores are decided
A score out of 10 labelled our score is ours, awarded against fixed criteria written down before testing. A rating labelled store rating belongs to the Shopify Theme Store or App Store and is reproduced with its review count and the date we read it — we never present someone else’s rating as our verdict. Where a column is empty, we have not measured it yet and will not fill it with a guess.
What we will not do
- Score something we have not used.
- Publish a measurement we did not take, including one borrowed from another site.
- Let a vendor see a draft before it goes live.
- Accept payment for a placement, a score or a position in a table.
- Leave an expired offer up because it still converts.