28 August 2026 · 4 min read
Checkout is the product
Most stores spend their budget on the homepage and lose the money three clicks later.
The homepage gets the photos, the copywriter and three rounds of changes. The checkout gets whatever the platform came with. That is backwards. Nobody has ever paid you on a homepage. Every taka a store earns goes through the checkout. Most stores I audit have never looked at it as a page at all.
Here is the shape of the problem. Take a store with 10,000 visits a month. 3 percent add to cart. 40 percent of those finish the checkout. That is 120 orders. Lift that 40 percent to 55 percent and you get 165 orders. Same traffic, same ad spend, same homepage. Nothing else on the site moves the number that much for so little work.
Where it breaks
The problems are boring. That is why they last. A phone field that rejects a leading zero. A district dropdown with 64 entries and no search. A shipping cost that only shows on the last step. A coupon field that sends people away to hunt for a code. A page that reloads on every error and jumps back to the top.
Each one costs a few percent. Together they cost a third of the people who wanted to buy. The founder never sees it. The founder has never bought from their own store on a mid-range Android phone over mobile data.
The cost of a field
Every field is a question. Every question is a chance to leave. Baymard’s checkout research puts the average form at around 12 fields. The ideal is seven or eight. On a Bangladeshi store the least you need is name, phone, address and area. Email is optional. A second address line is optional. Company name should not exist unless you sell to businesses.
Shopify lets you make most of these optional or hidden in the checkout settings. On WooCommerce a few lines in a child theme remove them. It is a 20 minute job. Stores put it off for years.
Guest checkout is not optional
Forcing people to make an account is the most expensive checkbox in commerce. People do not want a relationship. They want the thing. Let them buy as a guest. Offer the account after the order is placed, when they have a reason to want one: tracking.
Cash on delivery changes the maths
In a card-first market, the checkout ends when the payment clears. Here, most orders are cash on delivery. So the checkout ends when the courier hands over the parcel. That changes the goal. A fast checkout that makes orders nobody confirms is worse than a slightly slower one that makes orders that get delivered.
So the checkout has two jobs. It must be easy enough that a real buyer finishes. It must be firm enough that a bored one does not. In practice that means three things. Phone verification by OTP on cash on delivery orders. A clear delivery charge before the button. A confirmation screen that says exactly when the call will come. On the stores where I added OTP confirmation, fake and duplicate orders dropped by roughly half in the first month.
What to measure
Three numbers. First, the rate from checkout start to checkout finish. Shopify Analytics and GA4 both give you this once the funnel is set up. Second, time in checkout. It shows where people stop and think. Third, the delivery success rate on cash on delivery orders. It shows whether the checkout is letting the wrong orders through.
Watch them per device. Desktop is usually fine. Mobile is where the money leaks. Mobile is 85 to 90 percent of traffic on every Bangladeshi store I have worked on.
The order of fixes
Do not redesign the checkout. Shopify will not let you on a standard plan anyway. On WooCommerce a redesign is where the bugs come from. Remove fields first. Turn on guest checkout second. Show the delivery charge on the cart page third, so nobody meets it as a surprise on the last step. Add OTP confirmation fourth. Then measure for two weeks before you touch anything else.
Each of those is a settings change or an afternoon of work. Together they are worth more than the last three homepage redesigns combined. And unlike a redesign, they cannot make the store look worse.
Tomorrow morning
Open your store on a phone that is not yours. Buy something. Time it. Count the fields. Write down every moment you had to think. Then open the checkout settings and remove every field you did not need to place that order. It takes one hour. It costs nothing. It is the best hour you will spend on the site this quarter.