We checked 249 Shopify stores. Here’s what was leaking sales.

What we found in 249 stores

1. 89%221 of 249 stores had at least one product with fewer than 3 photos

   Shoppers who arrive from an ad can’t touch the product; one or two photos rarely answer their questions.

2. 83%207 of 249 stores still listed sold-out products

   Ads and collections keep sending shoppers to products they can’t buy. The median store had 7.

3. 18%45 of 249 stores promised delivery of more than 10 days in their shipping policy

   Long delivery is one of the most common reasons first-time buyers leave at checkout. The longest was 65 days.

4. 16%41 of 249 stores showed no product reviews we could see

   A first-time visitor from an ad has little reason to trust a store with no reviews.

5. 12%31 of 249 stores had at least one of these we couldn’t find: contact page, refund policy, shipping policy or terms

   Shoppers and payment providers look for these before trusting a new store. The contact page was missing most often.

6. 2%5 of 249 stores loaded the Meta pixel twice, from theme code and from Shopify’s Facebook & Instagram app

   Counts page views and add-to-carts twice, so Meta optimises on inflated numbers.

7. 2%5 of 249 stores had a test product visible to shoppers

   Looks unfinished, and a test order can end up in sales figures.

What to do with this

• Check the shop before the tracking. Doubled pixels were rare (2%); sold-out products and thin photo sets were the norm.
• Hide or restock sold-out products every week, and pause ads that point to them.
• Give every product at least 3 photos: on a person or in use, close up, and to scale.
• If you promise more than 10 days’ delivery, say so on the product page, not only in the policy.

Who was in the check

We checked 376 store addresses on 27 October 2026. 84 weren’t Shopify stores or didn’t share their product list publicly, and 43 showed no Meta pixel on the homepage. That left 249 Shopify stores with a Meta pixel, each checked in full. A typical store listed 154 products.

Each store was looked at once, from its public pages, the way any shopper sees it. No logins and no private data. A store counts once per finding, however many products it affects.

Limits

• One check of each store’s public pages on one day. Stores change, and some fixes may already be made.
• Stores were small and mid-size brands in five niches, mostly fashion and jewellery, in four countries. It is not a random sample of all Shopify stores, so treat the percentages as a guide for similar stores.
• Only what any visitor’s browser can see. Server-side tracking, checkout pages and anything behind a login weren’t checked, so tracking problems there aren’t counted.
• “No reviews” means none were visible to us; a store whose reviews load late may be counted.
• 43 Shopify stores showed no Meta pixel on the homepage and were left out. Some load the pixel only after cookie consent, so this isn’t a count of stores without tracking.