Skip to content

Why Nearly Half Your Shoppers Never Add to Cart (And How We Fixed It)

Real data from two online stores: 49% and 41% of shoppers who viewed a product never added it to cart. What was stopping them, what we changed, and the scary checkout error that turned out to be harmless.

Published 5 min readModulyth Systems

What the numbers showed

Two online stores we work with had the same problem in the same week: plenty of visitors, not enough orders. Both run our visitor analytics, which counts how many visits reach each step of buying: view a product, add to cart, check out and pay.

At the first store, 349 visits viewed a product and 179 added something to cart. That is 51% added and 49% that did not: 170 visits stopped at the product page. The drop was worst on computers.

At the second store, 504 visits viewed a product and 297 added to cart. That is 59% added and 41% that did not, so 207 visits stopped there. That drop was worst on phones.

Different stores, different devices, same step. When nearly half of the people who look at a product never add it to cart, the product page is where the money leaks, so that is where we started.

Fix one: the button was below the fold on laptops

At the first store we measured the live product page at common laptop screen sizes. The Add to cart button sat 909 pixels down the page at every one of them, below the bottom of the screen. A shopper on a laptop saw the photo, the name and the price, but had to scroll to find a way to buy.

We moved the volume discount tiers under the Add to cart row and trimmed the space above the product. The button now sits 672 pixels down, on screen without scrolling on any laptop screen 768 pixels tall or more. We also added a slim buy bar along the bottom of the screen on computers. It appears only while the real button is scrolled out of view and uses the same form, so prices and limits work exactly as before. Phones were already fine and did not change.

Fix two: shoppers had to sign in before they could tap Add to cart

At the second store, every order needs a free account, which is a sensible fraud control. But a shopper who was not signed in never saw an Add to cart button at all. They saw a link asking them to sign in first, and many of them simply left.

We kept the account rule and changed the order of things. Now a signed-out shopper taps a real Add to cart button. We remember the product and the option they picked, send them to sign in or create an account, and add the item to their cart as soon as they come back. A short line under the button says a free account is needed to order, so nobody is surprised.

The lesson for any store with an account wall: let people choose first and sign in second. The choice is the commitment. The login is paperwork.

The checkout error that was not costing sales

The same week, the analytics flagged a page error on the second store's checkout: a payment check timing out, 45 times in 42 visits. That looks alarming. Before building anything, we looked at what those shoppers did next.

Across the 48 checkout visits that hit the error over a slightly longer window, 40 placed a paid order seconds later. That is 83%, well above the store's normal checkout rate. The error came from the payment provider's own background check, and the payment carried on without it.

So we did not rewrite the checkout. We stopped the harmless error from being counted as a problem, and fixed the one real issue we found nearby: images from one payment app were being blocked by the site's security settings.

Not every error is losing you money. Check what the people who hit it actually did before you pay for a fix.

Speed, and a guard against double orders

Real visitors at the second store waited 9 seconds on average for the main content of the contact page to show. Under 2.5 seconds counts as good. Only 8 page views were measured, so we treated it as a lead, not a verdict.

The page itself was light. The weight came from things every page loads. The header logo had no size hint, so phones downloaded a file of about 48 KB to show a small logo, and a font used only on the homepage was preloaded everywhere. Sizing the logo correctly cut it to about 6 KB. With the extra font preload gone, page weight fell from 588 KB to 509 KB and font downloads from 119 KB to 83 KB. In lab tests the main content now shows in 2.23 seconds instead of 2.32. The real-world wait was most likely first visits taking an extra redirect on slow phone connections, so we are watching the next report rather than calling it solved.

While in the checkout, we also traced a case where one payment was submitted twice, 2 seconds apart, creating two orders. The payment provider declined the first charge and refused the second, so nobody was charged, but it should never happen. Checkout now allows one payment submission at a time and each payment can only be used once, both on the page and on the server.

Catching the visitors who are not ready yet

The second store had 1,520 visits in one week and 0 email sign-ups. Its only sign-up form sat at the very bottom of the homepage, and 94% of visits came from phones, where few people scroll that far. The store had 15 sign-ups ever.

We added a short restock alerts sign-up to the footer of every store page, and a small card that slides up after 45 seconds on a page or 60% of the way down it. It never shows during checkout or sign-in, and closing it hides it for 30 days. The email box uses a 16 pixel font so phones do not zoom in when you tap it.

What to take from this

If you sell online, measure the steps, not just the total. Find the step where the most people stop and fix that one first. For both stores it was the product page, for different reasons: one button was out of sight, the other was hidden behind a login.

Then check your alarms before acting on them. The scariest-looking error here was harmless, and the real wins were a button position and the order of two steps.

If you want to see where your own visitors stop, run the free website audit. Or tell us how your store runs and we will show you what we would look at first.

Next steps

Let's design the system your business runs on.

Thirty minutes on how your business runs. You leave with a plan, the package that fits and a fixed quote within days. Most businesses start with Scale, from $3,000, with the first month of the Growth care plan free.