The checkout is where the money leaks

Every part of the budget invested into ads, SEO and content has a couple of goals. One of them is getting the customer to the checkout. And then, at the very moment when the goal is almost reached, most of them leave.

The whole phenomenon is studied every year by different companies. The Baymard Institute puts the average abandonment rate at just over 70%, and other industry series confirm similar numbers year after year. Marketing has become more sophisticated, storefronts prettier, delivery faster, and still seven out of ten baskets end up as an empty promise. An order that will sit in the basket for weeks.

Some of that is simply because of how people shop. In Baymard's US survey, 42% of shoppers had abandoned their cart only because they weren't ready to buy or were just browsing. The modern basket has become a wishlist and a price comparison tool as much as a signal of intent. That is not something a checkout can change, and it should not try to.

The rest are a different story. They wanted the product and still left, because something in the checkout itself got in their way. It may have been a cost that appeared late, a payment method that was missing or a form that asked too much. Some of it was designed in, some came with the platform. The bottom line, however, is that the shop has to fix it.

The leaks in the checkout you can plug

When we set aside the people who are just browsing, a clear pattern appears. The biggest fixable cause is unexpected cost: 40% left because extra charges were too high. Not trusting the site with card details follows at 19%. Then being forced to create an account (18%), a long or complicated checkout (17%) and not seeing the total cost upfront (12%).

What do all those causes have in common? None of them are about the product or its price. People left because the final step made the whole purchase feel riskier than it needed to be.

That is not a marketing problem, it is a psychology problem. People feel a potential loss more sharply than an equivalent gain, so any surprise or ambiguity at the moment of paying tips the balance towards walking away, even when the product still appeals. Fraud, hassle, a difficult return: the checkout is where all of those possibilities become vivid.

In checkout audits, this is the single most common finding. Clients usually walk in with a pricing or assortment hypothesis, then the session recordings show something else entirely. For example, when the customer adds the item, reaches the last screen and hesitates over a cost or a step that was not part of their mental model of the purchase. Loss aversion is not an abstract concept here. It shows up as a specific click, at a specific moment, and it is measurable. Treating it as a UX detail rather than a structural risk is usually the first mistake we correct.

~ Grzegorz Sperczyński, Business Development Manager

Baymard estimates that fixing documented checkout usability problems would lift the average large site's conversion rate by around 35%, which it translates into roughly 260 billion dollars of recoverable orders across the US and EU. This figure is built on assumptions and is best read as a rough scale rather than a precise number.

Fewer steps, fewer fields, fewer surprises

In 2026 there is a clear move towards simpler checkouts. Polish market reports point out that single-page flows convert better than multi-page ones, and that conversion drops with every additional screen. Most checkouts could cut the number of fields shown by 20 to 60% without losing anything the order actually needs.

The practical checklist is short.

Show the full cost early. Delivery costs belong on the product page and in the basket, not behind the address form. If the shop sells across the EU, all-inclusive prices are simply what customers expect. Make guest checkout the default. Offer an account later, once the order is confirmed and the customer has a reason to want one.

Ask only for what the order needs. Every optional field is one more place to hesitate.

Put trust where the doubt is. Payment logos, the returns policy and a support contact belong next to the pay button. Not in a footer that nobody scrolls to on a phone. McKinsey's research on returns shows that a guaranteed refund and the cost of sending something back matter more to shoppers than almost anything else. One line about returns at the checkout is one of the cheapest trust signals there is.

Handle errors like a human would. An unclear message at the payment looks, to the customer, like a broken site.

Payment and delivery are part of the checkout, not something that comes after

Here the Polish market goes its own way, and generic international advice stops being useful.

BLIK is the clear favourite. According to Gemius, 72% of Polish online buyers use it and for 56% it is the method they reach for first. Among 15 to 24 year olds the share climbs to 90%. Having several payment methods to choose from is itself a reason to buy for around half of internet users. So what does a Polish checkout without BLIK on top looks like? Like a leak built in on purpose.

Delivery works the same way. Most Polish online shoppers prefer parcel lockers, and having a locker nearby actively encourages people to order. If choosing a locker means leaving the checkout, hunting through a map and coming back, we have just put the friction back that we removed two steps earlier.

There is a wider European angle too. Speed is the main reason people take up digital payment methods at all. Nobody chooses a wallet because it is clever. They choose it because it ends the transaction faster.

The B2B checkout leaks in its own way

Business buyers abandon too. They just do it for different reasons.

A purchasing manager who cannot see the negotiated prices, cannot pay by invoice or purchase order, cannot split one delivery between two warehouses, or cannot get a VAT invoice without writing an email, will pick up the phone. The order may still happen. The digital channel, however, has failed, and every phone order costs exactly what the online store was supposed to save.

The fixes mirror the consumer ones: show everything, ask for less, offer the right way to pay. The difference is in the details. "The right way to pay" means payment terms and PO numbers, not BLIK. "Show everything" means contract prices and stock levels, not delivery fees.

This is the part B2B clients underestimate most. A phone order looks like a success in any report that only counts revenue booked. Nobody puts a line item on the sales rep’s time, the manual invoice, the price re-entered by hand. Once that channel is costed properly, against what a working self-service checkout would have taken, the case for fixing B2B checkout usually closes itself. It stops being a UX argument and becomes a margin argument, which is the version finance actually acts on.

~ Grzegorz Sperczyński, Business Development Manager

A word of caution about the thank-you page

There is a growing idea that the moment right after purchase is a media opportunity. The customer has just shown trust, so why not show them partner offers?

Maybe. But the same logic that clears distractions out of the checkout should make us careful about what we add back the second the order is confirmed. Fix the leak first. Sell the thank-you page, only once we are sure it is not opening a new one.

What comes next?

One in five consumers would already let an AI agent finish a purchase for them, according to McKinsey's latest payments report. That is a small number today. It is also a warning. A checkout that confuses a human will fail an automated buyer outright. Software has no patience and no goodwill to spend.

Which brings us back to the start. The checkout is not a form to get through. It is the last place a shop can lose a customer who has already decided to buy, and the cheapest place to win one back.

Start with an audit. Count the steps. Count the fields. Check where the delivery cost first shows up. Then try to pay with the method most of your customers use on their phones. Most stores find their leak in an afternoon.