Shopify Store Benchmarks — what we measured across 198 real stores

These are our own readings, not a survey and not someone else's report. Every figure below was measured by reading a store's own public pages, the same way a shopper's browser would. No store is named anywhere on this page.

198

Stores measured

4,603

Product pages read

22,772

Images checked

682 KB

Median homepage weight

How to read this, and what the sample is. These are mostly established direct-to-consumer brands — the kind of store with staff looking after it. That matters: it is a demanding comparison set, not a cross-section of Shopify. If your store measures well against these, it is measuring well against stores with engineering teams. It also means these figures are not a claim about Shopify stores in general, and we do not present them as one. 198 stores is still a small sample. Each row states what it was measured over, because not every reading succeeds on every store — where one failed it is excluded, never counted as a zero.

What most stores get wrong

Finding	Figure	Measured over
Homepage images with no alt text	34%	7,732 of 22,772 images, 194 stores
Stores with at least one image missing alt text	94%	194 stores
Stores with no H1 heading at all	20%	198 stores
Stores with exactly one H1	52%	198 stores
Stores where a refund policy could not be found	13%	198 stores
Stores where a shipping policy could not be found	19%	198 stores
Products with only one image	18%	823 of 4,603 products, 161 stores
Stores with at least one single-image product	62%	161 stores
Products with no description at all	11%	490 of 4,603 products
Stores with at least one product missing a description	40%	161 stores
Homepages over 1 MB	26%	198 stores
Stores with no meta description	5%	198 stores

The typical store, in numbers

Measure	Median	Worst seen
Homepage load time	520 ms	5,126 ms
Homepage page weight	682 KB	8,025 KB
Third-party scripts on the homepage	5	29
Average product description length	61 words	across 161 stores

Method, in full

Each store was read once from its own public pages — no store login, no app install, nothing that a shopper could not see. The homepage was fetched and its size, time to arrive, third-party script tags, headings, meta description and images recorded. Up to thirty product pages were then read for description length and image count, and the refund and shipping policies were looked for where a store normally links them.

Where a reading failed — a store that would not load, or one whose product pages could not be read — that store is left out of the affected rows rather than recorded as a zero, which is why the samples above differ from row to row. Percentages are rounded to whole numbers. These are single readings taken at one moment, not averages over time, and a store measured today may read differently tomorrow.

Figures on this page are generated from the measurements themselves each time the page is served, so they move as more stores are scored. Nothing here is typed in by hand.

One question at a time

Each of these answers a single question from the same measurements, with the sample stated beside every figure.

Want to know where your own store sits against these?

Score my store — free

It reads your store the same way and tells you which of the problems above you have, ranked by what each one costs you. No sign-up and no store login. Your score is shown against this same set of stores.

A store does not stay fixed. Products change, apps get added, themes get updated, and the things you cleared this month come back next month. Store Watch re-reads your store every month and emails only what changed since last time — newly broken products, ones you fixed, and which way each score moved against these same 198 stores. $49 a month, cancel any time.

Prefer a single look? The Full Store Audit is a one-off $49, offered on your own result page after you have seen the free score.