A cold visitor who bounces off your homepage never cost you much. They were just looking. But a shopper who picked a product, chose a size, added it to the cart, and started typing in their shipping address? That person decided to buy. They are the most expensive visitor you have. And on the average store, about seven out of ten of them walk away at the very last step.
That is not a traffic problem or a product problem. Those shoppers are already sold. It is a checkout problem - and checkout is the one place where you lose the customers you already won. The good news is that the reasons people abandon at the end are well understood and almost entirely fixable. This is the last mile, and it is worth more than any other stretch of the road.
First, Find Out Where the Bleed Is
"People abandon their carts" is too vague to act on. There are two different leaks, and they point at two different problems.
Cart abandonment is the shopper who adds an item, looks at the cart, and leaves before starting checkout. Checkout abandonment is worse: they started the checkout, began handing over details, and still backed out. The further down they got, the more committed they were - so a leak deep in checkout is the most painful and usually the most fixable.
Pull the two numbers separately in your analytics. As a rough read, a cart-abandonment rate north of 65% points at the cart page, and a checkout-abandonment rate north of 45% points at the checkout flow itself. You are not trying to hit a magic number. You are trying to find out which step is doing the damage so you fix the right thing instead of guessing.
Most platforms will show you a step-by-step funnel: how many people reached the cart, then shipping, then payment, then paid. Look for the single step with the steepest drop. That one step is where your money is actually leaking, and nearly every fix below maps to a specific step - so once you know where the cliff is, you know which section of this article to start with.
The Number One Killer: Surprise Costs
Across years of large-scale checkout research, the Baymard Institute has found the same top reason people abandon, over and over: extra costs. Shipping, taxes, and fees that show up at the end push roughly 40% of abandoning shoppers out the door. It is not always that the shipping is too high. It is that it was a surprise.
A shopper builds a mental price as they go. When the total jumps at the final step, it does not feel like a shipping charge - it feels like a bait and switch, and trust evaporates right when you need it most.
The fix is to remove the surprise. Show costs early: put shipping and any fees in the cart, not four steps later. Even the wording matters - "shipping calculated at checkout" sets a gentler expectation than a bare "plus shipping" stamped on the product page. Better still, take the surprise off the table entirely with free shipping, or free shipping over a threshold that nudges the order size up while it removes the objection. Whatever the number is, the shopper should meet it long before the final button.
Stop Making People Create an Account
Right behind surprise costs sits a self-inflicted one: forcing shoppers to create an account before they can buy. Baymard's research pins nearly one in five abandonments on this single demand.
Think about what it asks. Someone has their card out, ready to pay, and you stop them to invent a password and confirm an email. To a first-time buyer that does not feel like a membership perk. It feels like homework standing between them and the thing they already decided to buy.
Turn on guest checkout and make it the obvious default. You still get the sale, the email, and the shipping address - everything you actually need. If you want the account, ask for it after the purchase, on the confirmation page, as a single checkbox that turns the details they just entered into a login. Offered that way it is a convenience. Demanded up front it is a closed door.
And you give up nothing by waiting. Most first-time buyers who check out as guests will happily make an account with one click once they see the order went through, and the ones who do get a faster checkout next time because their details are saved. Owners force the account up front because they want the data - but you already capture the email and address at guest checkout. You are trading a real sale today for a password you could have collected a minute later.
Shorten the Gauntlet
Every extra field and every extra step is another chance to lose someone. Open your own checkout and count the fields. If some are optional, they still add to the wall of work the shopper sees. Keep only what you truly need to take payment and ship the order, and give people a sense of progress with a simple step indicator - three steps is a good ceiling, and more starts to feel like a maze.
Two specific wins are worth calling out. First, express wallets - Apple Pay, Google Pay, Shop Pay - let a returning shopper skip the form almost entirely and pay in a tap or two. That is the single biggest piece of friction, gone. Second, the humble coupon-code box. It looks harmless, but a shopper who sees an empty "promo code" field often leaves to go hunt for a code online, and a good number never come back. If you are not running an active promotion, hide it. If you are, apply eligible discounts automatically instead of daring people to leave and search.
Earn Trust at the Exact Moment It Wobbles
The instant a shopper is asked for card details, a quiet alarm goes off: is this site safe? Baymard found that roughly one in five people abandon because they do not trust a site with their credit card information. The doubt is real, it is normal, and it peaks precisely at the payment step.
So answer it right there, where the worry lives. A short "secure checkout" line, the familiar payment icons, a visible padlock, easy links to your return and shipping policies, and the star rating with its review count near the button all do the same quiet job: they tell a nervous buyer that real people have done this before them and nothing went wrong. None of it is flashy. All of it converts the shopper who is one small reassurance away from finishing.
This is also the worst possible moment to introduce anything new. No surprise upsells, no pop-ups, no stray navigation to wander off into. By the payment step the decision is made, and your only job is to get out of the way. Think of it as a calm, sealed room with one clear exit: the buy button.
Let People Pay the Way They Already Pay
You may have a favorite payment processor with the lowest fees. That is fine for you, but it does not help if it is not how your customer wants to pay. Offer a genuine range: major cards, the digital wallets, and buy-now-pay-later options like Klarna or Afterpay for higher-priced items. With those installment providers you are paid in full up front while they carry the collection and the risk. Afterpay's own data puts the conversion lift from offering installments at more than 20% - treat that as their figure, not a guarantee, but the direction is clear. The rule is simple: meet the shopper's preferred payment method, or watch them leave to find a store that does.
This matters most on phones, where most shopping now happens and where typing a sixteen-digit card number is genuinely annoying. A wallet button that fills everything in one tap is often the whole difference between a sale and a sigh. And the loss is invisible: the shopper who only uses the one method you skipped does not email to complain, they just quietly leave.
Win Back the Ones Who Still Leave
Even a clean checkout leaks. Some people get interrupted, compare a price, or simply get distracted - and those carts are not dead, they are paused. Set up an automated nudge: an email or text about an hour after they leave, and another around the next day, each one linking straight back to the exact cart they built so they can finish in a tap. You are not starting a new sale. You are handing back the one they already assembled. It is some of the cheapest revenue in the whole store.
Then Test, One Change at a Time
Everything above is a strong starting point, but your store is not the average store. One-page versus multi-step checkout, removing a field, hiding the coupon box, reordering payment options, adding a trust line - nearly all of it can be tested. Change one thing at a time so you actually know what moved the needle, and judge each change on the numbers that matter here: cart-abandonment rate, checkout-conversion rate, and average order value. One small, verified improvement a month compounds into a very different checkout by the end of the year.
The Last Mile Is the Cheapest to Fix
You paid for the click. You earned the product decision on your product pages. Do not give it all back at the register over a surprise shipping line or a forced password. Checkout is the shortest, cheapest stretch of the whole journey to fix, and it sits right on top of the money.
If you would rather have someone dig into your cart and checkout numbers, find the exact step that is leaking, and fix it, that is the kind of work we do every day over at conversion rate optimization. And if you would rather do it yourself, start at the top of this list - the surprise costs - and work down. Every leak you close is a customer you had already won, walking back to the register.
