Cart Abandonment: Where Did Buyers Stop?
A shopper adds a pair of shoes to a cart, sees the delivery charge only after entering an address, and closes the page. Another shopper adds the same shoes to compare sizes and never intends to buy that day. Both appear as abandoned carts, but only the first points clearly toward a checkout problem. If a store treats every abandoned cart as a missed sale, it may spend money chasing browsers while leaving the delivery surprise in place.
Cart abandonment means that a shopper put an item in an online cart but did not complete an order. The useful question is not how to force that rate toward zero. It is where shoppers who were ready to buy met a cost, requirement, or failure that changed their minds. Baymard Institute notes that browsing, comparison shopping, and saving items for later account for some abandonment; its checkout research also identifies obstacles that stores can change. Baymard’s checkout research makes those two kinds of loss worth separating.
Define the rate before deciding that it is high
The standard cart abandonment calculation is abandoned carts divided by carts created, multiplied by 100. If, in an illustrative reporting period, 1,000 eligible carts were created and 300 of those carts became orders, 700 did not. The cart abandonment rate would be 70%. Shopify describes the rate using carts as the denominator; a store must also decide how it connects an order to its originating cart and how long it allows a shopper to return before calling the cart abandoned. Shopify’s cart abandonment guide gives the basic formula.
Now suppose that only 600 of those 1,000 carts reached checkout and the same 300 became orders. The illustrative checkout abandonment rate is 50%: 300 checkout starts without an order, divided by 600 checkout starts. The cart abandonment rate is still 70%. Neither number is wrong. They answer different questions. The first covers the journey after an item reaches the cart; the second begins after a shopper starts checkout. A change from one denominator to the other can make a trend appear to improve even when the store has gained no orders.
Keep the eligibility rules as steady as the denominator. Say whether the count covers unique carts or users, which orders qualify, and whether an order placed after the reporting window can complete a cart started inside it. Repeat purchases, reopened carts, and shoppers moving between devices need a consistent treatment. These choices are not clerical details: if the counted carts and orders do not describe the same shoppers and period, the result cannot guide a decision. Google Analytics 4, for example, lets a funnel start at different steps, but its funnel exploration counts users who complete configured steps in sequence and reports only the first qualifying sequence for a user in the selected date range. A GA4 user funnel therefore should not be assumed to equal a cart-level rate from a commerce platform. Google’s funnel exploration documentation explains those counting rules.
Baymard reports a 70.19% average cart abandonment rate across the studies it tracks. That figure offers context, not a target for an individual store. A business with many shoppers using the cart as a comparison list may have a different rate from one serving repeat buyers who arrive with a product already chosen. Baymard also reports that, in one study, 42% of US online shoppers had abandoned a cart in the prior three months because they were browsing or not ready to buy. That is a share of surveyed shoppers reporting a reason, not 42% of any one store’s carts. Baymard’s findings are most useful for understanding the mix of intent behind the headline rate.
Find the stage where willing buyers stop
The next decision is where to look first. A practical view follows a shopper from adding an item, to opening checkout, to entering delivery details, to attempting payment, to receiving an order confirmation. The point is to see where the journey changes, rather than treating all unfinished carts as one pool. Google Analytics 4 allows a funnel to be built from configured steps and can show how users move through them; open and closed funnel settings change who is allowed to enter at each step. Google’s funnel documentation describes both options.
Before interpreting a large fall between two steps, try the route as a customer would. Confirm that each expected event fires and that a successful order reaches the final step. If a shopper can skip a configured step, Google says that user can appear to drop out of a sequential funnel even after completing a later action. Payment redirects and alternate checkout paths deserve the same attention: a chart showing no purchase is not proof that no purchase occurred. Compare the final funnel count with the store’s completed orders for the same eligibility window, and fix unexplained gaps before changing the page. Google’s explanation of sequential steps is the reason to check the event path.
Once the path is trustworthy, separate cart-to-checkout loss from checkout-to-order loss. A sharp loss before checkout raises questions about the cart total, delivery promise, or whether the shopper was ready to buy. A sharp loss after payment begins calls for a different inspection: payment errors, unsupported methods, or a failure to show that the order succeeded. These are leads, not automatic diagnoses. Look at the actual cart page, checkout screens, and order records where the loss appears.
Break the same path out for groups whose experience can differ: mobile and desktop shoppers, first-time and returning customers, and shipping destinations with different delivery charges. Google Analytics 4 supports segment comparisons and device breakdowns in funnel explorations. Its documentation also warns that a user who enters on mobile and later completes on desktop can remain attributed to mobile in that breakdown. Google’s breakdown rules matter when interpreting a device gap. A difference between groups tells the store where to inspect; it does not, by itself, reveal the cause.
Show the full cost before the final decision
For many stores, the first improvement should be to make the amount payable understandable earlier. Shopify identifies unexpected shipping charges, taxes, and fees among reasons shoppers abandon, and it points to unclear delivery information as another source of hesitation. Shopify’s checkout discussion supports putting those facts near the purchase decision. A product price that looks acceptable can feel different once a delivery charge appears on the last screen.
Show the product subtotal, any known fees, and the shipping rule in the cart. If a precise tax or delivery amount depends on an address, say so and give the shopper a clear way to see the estimate before completing a long form. Put an estimated delivery date or a meaningful delivery range beside each shipping choice where the store can support it. “Standard shipping” alone leaves the customer to guess when the item will arrive. A visible return policy can also answer a question that otherwise sends the shopper away to search for it; Shopify lists unsatisfactory returns among reported reasons for leaving a cart. Shopify’s guide covers delivery clarity and returns.
Transparency costs less than automatically offering free shipping, but it can expose an unattractive total. That is useful information. If the charge is genuinely too high for the product, changing the copy will not make it affordable. The store must then decide whether it can reduce the delivery cost, set a clearly stated free-shipping threshold, change the product offer, or accept that some orders are uneconomic. Hiding the charge until payment postpones the decision and adds frustration at the point where a buyer is closest to ordering.
Remove requirements that do not help complete the order
When shoppers have reached checkout, make the route to an order obvious. Requiring a new account before a first purchase asks a shopper to make a second commitment before finishing the one they came for. Shopify names required account creation and complicated checkout among reasons for abandonment. Shopify’s guide recommends a guest option. A store that needs an account for later service can explain that benefit after the order or offer account creation without blocking the purchase.
Inspect each field from the shopper’s side. Ask whether a field is needed to deliver the item, take payment, or meet a real order requirement. If it is only useful for marketing or internal convenience, move it out of the mandatory path. Make errors specific enough to correct and preserve entries when a shopper returns to a previous step. Those are direct design choices, not a promise that shorter checkout always wins. A field that prevents delivery mistakes may be worth keeping; an optional company name does not need the same status as a delivery address.
Baymard’s research gives the scale of the issue without prescribing a universal form length. It reports that 17% of US online shoppers in a cited study said they had abandoned an order in the previous quarter because checkout was too long or complicated. Its benchmark also found an average of 23.48 form elements displayed by default in US checkout flows it examined. Those figures describe Baymard’s study and benchmark, not the effect a particular store will get by deleting fields. Baymard’s checkout research is a reason to inspect the actual form and remove needless work.
The payment step deserves separate attention. Shopify lists limited payment options, card declines, and website errors among potential reasons for abandonment. Shopify’s guide also advises choosing payment methods with the shopper’s country in mind. If buyers repeatedly reach payment and fail to order, check which methods they can use, what happens after a decline, and whether the page offers a clear way to retry without rebuilding the cart. Adding a payment method has integration and operating costs, so choose it for the shoppers actually reaching that step rather than filling the page with every available option.
Use reminders to recover interest, not to hide friction
A reminder can help a shopper who meant to return. Shopify discusses abandoned-cart email and messages that bring a shopper back to the items left behind. Shopify’s guide gives those as recovery options. A useful message shows the items, links back to the cart, and gives the customer the information needed to decide: the current total, delivery terms, or where to get help. It should stop when the order is completed. Otherwise, the store may tell a recent buyer that an already purchased cart is still waiting.
Recovery has a narrower reach than the headline abandonment rate suggests. A store cannot email a cart for which it has no usable address, and a shopper who was only comparing products may not be ready to purchase because a message arrived. It is also poor economics to offer a discount to every abandoner before checking whether the checkout itself is creating avoidable loss. A discount may convert an order while reducing the margin on it. If the problem is an unexpected delivery fee or a broken payment step, fix that experience first; the same obstacle will confront shoppers who return through an email.
Treat the recovery message as a continuation of the buying decision. If the cart was left before checkout, the message can clarify price or delivery. If it was left after a failed payment attempt, a route back to a working payment choice is more relevant than a generic promotion. The store needs a reliable way to distinguish those situations; otherwise, the safer message simply restores the cart and gives the buyer a clear path to complete it.
Judge changes by orders, not by a prettier rate
Suppose a store removes a required account step and its checkout abandonment rate falls. That is promising, but it is not the whole result. Count the completed orders from the same eligible checkout starts, and check whether payment success and order value changed. A lower rate can arise because the entry mix changed or because fewer low-intent shoppers started checkout. If a free-shipping offer was part of the change, weigh the added orders against the delivery cost. A rate is useful only when it leads back to real orders and their economics.
Compare like with like over a stable window, and prefer a controlled comparison when the store can run one. If a new checkout goes live at the same time as a sale, a campaign, or a change in shipping prices, the before-and-after difference cannot be assigned confidently to the new checkout. Keep the cart and checkout definitions fixed while comparing versions. Look at the step that was changed, then at completed orders; a change that moves people one screen farther but fails to produce more orders has not solved the purchase problem.
The sensible order of work is usually to make the count trustworthy, inspect where buyers stop, reveal the full price and delivery promise, remove avoidable checkout work, and then use reminders for shoppers who still leave. Baymard’s research shows why no store should expect every cart to become an order: some carts belong to people who are still browsing. Baymard’s findings also show that checkout obstacles are real. A useful cart abandonment program accepts the first fact and works steadily on the second.