If you run an online store, you have probably watched this happen: solid traffic, a healthy number of items added to cart, then a disappointing number of actual sales at the end of the day. That gap has a name, and it has been studied in more depth than almost anything else in ecommerce. Here is what the research actually shows, and what to change in your checkout before you spend money on a redesign.
- Around 70 percent of online shopping carts are abandoned before the order is placed, according to Baymard Institute's ongoing research across dozens of studies.
- Extra costs that appear late, shipping, tax, fees, are the single biggest reason shoppers give for walking away.
- Forcing a new account before checkout pushes away roughly a quarter of buyers who would otherwise pay.
- Fixing common checkout problems can lift conversion by around a third, based on a decade of large scale checkout testing.
The scale of the problem
A Baymard Institute analysis of more than 50 separate abandonment studies puts the average rate at just over 70 percent. That figure only counts shoppers who added an item to their cart, so it filters out casual browsers and reflects people who showed real intent to buy. For a Sydney store running 1,000 checkouts a month, that is roughly 2,300 people who started to buy and then stopped. Not all of that is recoverable. Some visitors were comparing prices or waiting for payday. But a meaningful share of it is lost to problems inside the checkout itself, and those are fixable.
The costs nobody mentioned until the last screen
Baymard's checkout research consistently finds one reason ahead of everything else: shoppers reach the final page, see shipping, tax or a service fee added on top of the price they expected, and leave. Close to half of shoppers who abandon for a fixable reason, as opposed to just browsing, point to this. It is an easy problem to create by accident. A store that shows a tidy price on the product page and then adds $12 shipping and a card surcharge at the very last step is training customers to distrust the final total. Showing the full price, or at least a realistic shipping estimate, in the cart rather than at the last step removes the surprise and, with it, the reason to bail.
Making someone sign up before they can pay
Roughly a quarter of shoppers abandon a purchase specifically because a site insists on creating an account before it will take payment, per Baymard's research. That is a strange trade for most small stores to make. An account has real value for repeat customers and email marketing, but making it compulsory at the exact moment a buyer is trying to hand over money works against you. A checkout built with fewer steps asks for an account after the sale, using the details already entered, and keeps both the order and the chance to build a mailing list.
Slow, long and untrustworthy checkouts
The same body of research also flags checkouts that stretch across too many pages, ask for information that is not needed to complete the order, or do not visibly reassure a buyer that their card details are handled properly. None of these show up as dramatically as the cost surprise, but they compound. Every extra field or page is another point where someone can change their mind, especially on a phone, where most Australian online shopping now happens. A date-of-birth field nobody needs, a phone number marked required when it isn't, or a captcha that fails on a small screen: any one of these is a small thing, and each one is also a place where a real customer, halfway through paying you, just closes the tab.
What actually fixes it
None of this requires a rebuild. The fixes are specific and mostly configuration, not design work:
- Show the real total early. Put an honest shipping estimate in the cart, not the final screen.
- Turn on guest checkout. Ask for an account after the sale, not before it.
- Cut it to one or two screens. Merge shipping, delivery and payment onto as few pages as the platform allows.
- Test it on a phone, properly. Autofill, number pads for card fields and large tap targets matter more than they look like they should.
- Put trust signals near the payment button, not buried in the footer: card logos, a security note, a link to the returns policy.
What this is worth
Baymard's long running checkout testing across major retailers found that fixing common problems like these lifts conversion by around 35 percent on average, built from a decade of large scale testing. Most small Australian stores will not see exactly that figure, results depend on the platform and the traffic, but the direction is consistent. Checkout is one of the few pages on a website where a handful of specific, unglamorous fixes reliably turn into more completed sales, not just more visitors. It is also cheaper to fix than almost anything else on the site. There is no new photography to shoot, no homepage to rewrite, and usually no developer needed beyond a few settings changes in Shopify, WooCommerce or whatever platform the store already runs on.
Where to start today
Open your own checkout right now on your phone and count how many screens it takes to pay from an empty cart. If it is more than two or three, or if shipping only appears at the very end, that is your starting point. It is usually a smaller job than it sounds, and platforms like Shopify let you turn on guest checkout and reorder steps without touching the rest of the site.
