The biggest leak in your mobile checkout is a number you're hiding

A brand I work with sent me five screenshots of their mobile checkout and asked, more or less, "is this fine?" Most of their buyers are on a phone, so this little three-screen flow is where the actual money is made or lost. It was not fine. And the single biggest leak was a number they were deliberately hiding.

At the very top of the first screen sat a grey warning banner, something like "customs and VAT may apply." No figure. Just a vague threat, planted at the exact moment a shopper is deciding whether to trust you with their card. It's the digital equivalent of a shop assistant whispering "there might be extra charges, who knows" as you reach for your wallet.

The thing shoppers hate most is a surprise cost

This isn't a hunch. Baymard Institute has been testing checkouts for over a decade, and across their data the number one reason people abandon a cart — once you set aside the window-shoppers — is "extra costs too high (shipping, tax, fees)," at 39%. eMarketer ran the same finding under the headline "extra costs are the No. 1 reason consumers abandon online carts." It's not even close to the next reason.

What kills the sale isn't the cost itself. It's the surprise. A vague banner with no number is the worst of both worlds: it creates the fear of a surprise cost without giving the shopper anything concrete to make peace with. You've triggered the abandonment trigger and withheld the one thing that could defuse it.

A scary banner with no number doesn't manage the fear. It manufactures it.

The fix is a strategy decision before it's a design one. Ship duty-paid so the customs question disappears entirely, and flip that grey threat into a green reassurance — "duties and VAT included." If the business genuinely can't absorb that yet, then show the estimated landed cost as a real number in the summary. Either way, you replace dread with a fact. People can accept a fact. They cannot accept a lurking maybe.

The second leak: buying blind

The order total didn't appear until after two full pages of forms. So on the smallest screen, with the highest stakes, people were filling in their address while committed to a figure they couldn't see. That's backwards. The summary — the total, what's in the cart — should be there from the first step, as a collapsible bar you can tap open whenever doubt creeps in.

And there were no express wallets. No Apple Pay, no Google Pay, no PayPal button. On mobile, asking someone to hand-type a sixteen-digit card number is a tax you're charging your most ready-to-buy customers. Baymard's wider work is brutal on this: the average checkout shows roughly twice as many form fields as it needs to, and that bloat is recoverable conversion. Every field you remove is a small refund on abandonment.

None of this came from a dashboard

Here's the part I keep coming back to. You could stare at this brand's analytics for a month and see a soft mobile conversion rate, and "mobile converts worse than desktop" is true of nearly every store, so you'd shrug and move on. The dashboard tells you that it's leaking. It will never tell you it's a colour choice on a banner and a hidden subtotal. For that you have to actually walk through the thing on a phone, as a nervous first-time buyer, and feel where your own hand hesitates.

That's the same lesson as a dashboard that lies by omission — the metric flags the symptom and stays silent on the cause. And it's why I keep arguing that the judgement is the leverage, not the tooling. An AI can build the before-and-after mockups and write the platform code in an afternoon — and it did. What it can't do is feel the flinch.

One honest caveat

I'd never ship these changes blind and call it a win. Every "obvious" checkout improvement should be A/B tested and properly instrumented, because checkout is exactly where confident assumptions go to embarrass you. The customs reframe in particular can't go live until the business can actually ship duty-paid — promise "included" before it's true and you've built a returns problem instead of a conversion one.

But the direction isn't in doubt. Stop hiding the number. Show the total early. Let people pay with a thumbprint. Your best customers are the ones already reaching for their wallet — the least you can do is get out of their way.

Sources & further reading

External
Baymard Institute — Cart & checkout abandonment statistics
eMarketer — Extra costs are the No. 1 reason consumers abandon online carts

Related posts
Your analytics dashboard is lying to you by leaving things out
The death of generic AI: why deep domain expertise is the only real leverage left