Most checkout losses are decisions the webshop made, not shoppers changing their minds
Roughly seven in ten online carts never become an order. Baymard Institute puts the documented average at 70.22%, calculated across 50 studies, and has tracked the rate for fourteen years. A number that high invites a shrug: shoppers browse, compare and leave, and nothing a webshop does will change that. Baymard’s checkout research makes the opposite case. A large share of the loss happens after the shopper has decided to buy, and it traces back to how the checkout is built.
The evidence behind that claim is broad. Baymard has run 25 rounds of moderated usability testing with more than 4,400 participant and site sessions, benchmarked 344 leading US and European sites against more than 110 cart and checkout guidelines, and audited the checkouts of over 150 sites. One finding keeps returning: for many shoppers who abandon, the checkout itself is the whole reason. Some lose patience with the process and walk away. Others stall at a field they cannot work out how to fill in.
Separate the shoppers you were never going to win
Not every abandoned cart is a failed checkout. In Baymard’s survey of US online shoppers, 42% say they have left a cart because they were only browsing or not yet ready to buy. Those visitors use the cart as a wishlist, a price check or a way to see delivery costs. No change to your form will convert them, and a report that counts them alongside genuine failures sends the team after the wrong problem.
Set those browsers aside and the reasons change character. At the top are extra costs for shipping, tax or fees that were too high (40%), followed by slow delivery (20%), not trusting the site with card details (19%), being made to create an account (18%), a checkout that felt too long or complicated (17%) and errors or crashes on the site (17%). Further down, 12% could not see or calculate the total cost up front, 10% had a card declined and 9% did not find enough payment methods. Most of these come down to what the shop shows, asks or offers, and at which moment.
That split also tells you how to read your own analytics. Abandonment counted from add-to-cart mixes browsers with buyers. Abandonment counted from the first checkout step, broken down per step, comes much closer to measuring the checkout itself, and shows you which screen loses people.
Leading sites are further from good than they look
Baymard’s benchmark makes uncomfortable reading. Of the 344 sites, 65% score mediocre or worse on checkout UX, 35% reach decent or better and only 2% reach good. None reaches the top two ratings. These are large retailers with dedicated teams, not template shops. The average site still has 32 specific improvements to make.
The institute estimates that the average large site can potentially gain 35% in conversion rate through better checkout UX, a figure based on its usability test sessions. Read that as a potential, not a forecast for your shop. Even a fraction of it matters, because it applies to visitors you have already paid to bring in. A better checkout does not need more traffic to show up in revenue.
Where the fixes usually sit
Start with the order of information. Delivery costs, fees and the total belong on the product page or in the cart, before the first checkout field. A shopper who meets a new cost on the payment step reads it as a surprise, and a share of those shoppers leave at that point.
Next, look at what you ask before you allow a purchase. If creating an account is the default path, guest checkout becomes a link most people miss. Make the guest route at least as visible as logging in, and offer the account after the order, when the shopper has a reason to say yes.
Then count your fields. Every field is a question the shopper has to answer, and every unclear label is a chance to stall. Use one name field where you can, drop the second address line if nobody needs it, and let the postcode fill in the street where your market allows it. Error messages should say what is wrong and how to fix it, next to the field, without wiping what the shopper already typed.
Finally, the payment step. Show the methods your shoppers expect, in the order they use them, with names and logos they recognise. In the Netherlands that starts with iDEAL | Wero. A shopper who does not see their method, or does not recognise the label, hesitates at the last step before the order. When a card is declined, the message should offer a next step, such as another method, rather than a dead end.
The simplest test costs nothing. Go through your own checkout as a first-time guest, on a phone, with no saved details, and write down every moment you hesitate. Hold that list against the reasons above. A large part of what you find is wording, order and defaults rather than development work.
Sources: baymard.com, checkout usability research and baymard.com, cart abandonment rate
