Shoppers Aren't Abandoning Your Checkout. They Never Found the Product.
Nearly every store we've looked at fails the most basic test of navigation: does the menu use words a shopper would actually use? In the stores where we could test this, 22 of 24 failed, a 92% failure rate, and that's before anyone has touched a filter, a search box, or a product page.
That number matters more than it looks like it should, because it sits upstream of everything else. A store can have a fast checkout, a clean cart page, a generous return policy stated three times, and none of it helps if the shopper never gets past the main menu with the right product in front of them.
The menu doesn't speak the shopper's language
When we say navigation fails on customer language, we mean a specific, visible thing: a shopper opens the nav expecting to see themselves described, and instead sees a label built around how the merchandising team organizes inventory. A shopper looking for a warm jacket sees "Outerwear." A shopper looking for a gift sees a brand's internal collection name instead of "Gifts under $50." Nothing is broken in the technical sense, the link works, the page loads, but the shopper has to translate the store's language into their own need, on a phone, in a few seconds. Most of the time in our audits, that translation didn't happen. The shopper picked wrong, or picked nothing.
Filtering and search: the backup that also fails
If the menu doesn't get a shopper to the right shelf, filtering is supposed to catch them once they're on it. In the stores where we could observe this, 12 of 16 failed, 75%. That means once a shopper lands on a category with more products than fit a single screen, there's no way to narrow by size, price, or the other one or two things that actually decide a purchase. The shopper is left scrolling and eyeballing, on mobile, which is the worst possible place to do that.
Search is the other backup, and it also breaks more often than not: 15 of 25 stores failed the check for search being available and accessible, 60%. A shopper who knows exactly what they want, and types it, still hits a wall in more than half the stores we audited. That's not a discovery problem in the abstract sense, that's a shopper with money in hand and no path to spend it.
Product cards ask for a tap before they've earned it
Even when a shopper does reach a list of products, the cards themselves often withhold the information needed to choose. 13 of 19 stores failed the check for product cards carrying decision information, 68%. No visible price range, no rating, no indication of what sizes are actually in stock. So the shopper has to open a product just to find out whether it's worth considering, then go back, then open the next one. Each of those taps is a chance to give up, and on a small screen with a spotty connection, giving up is easy.
Across the stores we've scored, Product Discovery as a category averages 69.1, from 23 stores. That's not the lowest average we've measured, Product Presentation and First Impression both score lower, but the average hides how concentrated the failures are within a handful of specific, high-impact checks. A store can score reasonably on discovery overall and still fail navigation language, filtering, and card information all at once, because those three checks are doing three different jobs and a store can lose all three.
Why this beats checkout as the place to fix first
We don't have a checkout category in this data, and we're not going to pretend we do. What we can say is that none of the failures above happen at checkout. They happen earlier, at the exact moment a shopper is trying to decide what they even want to look at. If a shopper never finds the product, the condition of your checkout is irrelevant to them, because they never get there. Fixing navigation language, adding filters, and putting decision information on product cards addresses a problem that happens before the cart exists, which is why it's worth looking at first.
The sample, plainly
This comes from 25 audits across 23 sites we selected for outreach, not a random sample of the market, so treat these as a pattern worth checking against your own store rather than a market benchmark. Each audit is one mobile visit as a first-time shopper, not a full site review, which is exactly the position most of your traffic is actually in.
If you want to see where your own store's discovery path breaks, you can run the same audit at dropiq.ai.