How to fix your checkout
A checkout in three or four steps, thumb-friendly, with the typing taken out. The full checklist.
Most people who reach your checkout already want the thing. They picked the size, read the reviews, and tapped buy. Then the form asks for one field too many, or the shipping cost shows up for the first time on the last screen, and they close the tab.
The Baymard Institute puts the average documented cart abandonment rate near 70 percent. A good share of that happens inside the checkout itself, for reasons that have nothing to do with price or intent. Close to one in five people who leave during checkout say it was too long or complicated. More than one in five leave because they could not see the full cost up front. Those are design problems, and design problems get fixed.
This is the checklist we work through with a store, roughly in the order that matters.
Keep it to three or four steps
The shape that holds up: cart, shipping, payment, confirmation. Four screens, or three if payment and shipping share one. Every step past that is another place to reconsider, and another place to lose people who were ready to pay.
Show a progress indicator so a shopper can see the end from the start. Three marks and a label is enough. The point is not decoration. Someone who knows there are two steps left will finish two steps. Someone with no idea assumes it is more and leaves.
Take the typing out
Typing is where mobile checkouts die. A street address on a phone keyboard is a dozen taps and two fights with autocorrect. So take the typing out wherever you can.
- Offer Apple Pay and Google Pay first. They skip the form entirely. Name, address, and card come from the phone, and for a lot of mobile shoppers that is the whole checkout, done in one tap.
- Turn on address autocomplete. One line of typing, a tap on the suggestion, and the street, city, and postcode fill themselves.
- Detect the card type from the number. The Visa or Mastercard mark should appear as they type, not sit in a dropdown they have to set by hand.
- Let people paste. Every field. Blocking paste on a card number or a discount code is a small cruelty that costs real orders.
Build for the thumb
Most of your traffic is on a phone, and that is where checkout goes worst. The rate runs higher on phones than on desktop, and the reason is physical: the targets are too small and the keyboard eats half the screen. The device-by-device numbers are in the statistics post.
- Make every tappable thing at least 44 pixels tall. That is about the width of an adult thumb. Anything smaller gets mis-tapped, and a mis-tap on the pay button reads as a broken page.
- Bring up the right keyboard for the field: a number pad for the card, an email keyboard with the at sign for email.
- Test on a real phone, held in one hand, not a desktop browser dragged narrow. The emulator will never mis-tap the way a thumb does.
Ask for less
The average checkout asks for far more than it needs. Baymard finds most run around a dozen form fields when half that would do. Every field is a small tax on finishing.
- Only ask for what you will use. A digital product does not need a phone number. A physical one might, for the courier, so say that is why.
- Default the obvious. Country from the connection, billing address the same as shipping unless they say otherwise.
- Mark the optional fields optional, not the required ones required. A field with no label reads as mandatory, and mandatory fields are where people stall.
Let people check out as guests
About a quarter of shoppers who abandon during checkout do it because they were made to create an account first (Baymard). That is a quarter of the people who reached the form, gone over a password they did not want.
Make guest checkout the default path. Offer the account after the order is placed, when you already have their email and the sale is safe. Most will say yes then. None will leave over it.
Show the whole price early, and errors in place
Two small things carry a lot of weight.
Put the full price, shipping and tax included, in front of people before the last screen. The single most common reason a cart gets abandoned is a total that jumps at the end. No badge fixes a surprise. Showing the number early does.
And when a field is wrong, say so next to the field, the moment they leave it, in plain words. Not a red banner at the top of the page after they press pay. An error they can see and fix where they are is half a second. An error they have to hunt for is a lost order.
When the blocker is a question, not a field
Sometimes what stalls a checkout is not a form field. It is a question. Does this ship to Austria. Is this the 30 ml bottle. Can I pay in four. If answering it means backing out to a product page and losing the cart, a good share of people just go.
This is where Linra sits. A shopper asks at the last step, the copilot answers in the same thread, and the cart is right there to check out. No new tab, no lost place, no retyping.
Question to checkout, one thread
Where to start
You do not have to do all of this at once. Take the two with the widest reach first: put the full price up front, and turn on Apple Pay and Google Pay. Then shorten the form. Baymard estimates a well-built checkout is worth about a 35 percent lift in conversion for a typical large store. Most of that is not a trick. It is just not making people work for something they already decided to buy.