Online merchants: the checkout frictions that cost you the most
The average checkout, the sequence of screens between the cart and the order confirmation, displays some twenty form fields when a dozen are enough. A good share of your abandonment comes neither from the product, nor the price, nor the traffic you paid for: it comes from what you ask the customer at the moment they have already decided to buy.
- The Baymard Institute measures an average of 23.48 form elements displayed by default in a US checkout, while its testing indicates an optimal checkout can be as short as 12 to 14.
- It observes that most sites can reduce the number of form elements displayed by default by 20 to 60%.
- 17% of shoppers say they recently abandoned an order solely because of a difficult checkout process.
- Once you set aside those who were browsing with no intention to buy, the top reason for abandonment is extra costs discovered too late.
- These fixes require no advertising budget and are measured on revenue, not click-through rate.
On this page
What is checkout form load
Checkout form load is the number of input elements displayed by default in a checkout, before the customer does anything. It differs from the number of steps, which does not have the same effect: a checkout with three light steps completes better than a single overloaded page. What decides abandonment is the number of decisions asked of the customer, not the length of the path.
What the abandonment rate does not tell you
We are writing here for executives and managers of online stores, in retail as in B2B. It covers the checkout itself, not the cart, which is treated elsewhere in our corpus.
The average cart abandonment rate sits around 70%, a figure the Baymard Institute compiled from some fifty separate studies. It is quoted everywhere and it misleads, because it mixes two populations that have nothing in common.
A large share of that abandonment comes from people who were browsing with no intention to buy. According to the Baymard Institute, 42% of US shoppers abandoned a site in the past three months simply because they were browsing or not ready to buy. Those people cannot be recovered, no matter how good your checkout is.
The second group is the one that matters: customers who intended to buy and left during checkout. The recoverable revenue sits there, and it is the only population worth measuring.
A business that judges its checkout on the overall abandonment rate will conclude it is average and do nothing. Measure the abandonment rate from the moment people enter the checkout instead, step by step. It is a different number, lower, and far more actionable.
Form load
It is the simplest measurement to take and the most revealing.
Count the input elements displayed by default in your checkout: fields, checkboxes, dropdowns, radio buttons. The Baymard Institute measures an average of 23.48 elements on US checkouts, while its usability testing indicates an optimal checkout can be as short as 12 to 14.
It also observes that most sites can reduce the number of fields displayed by default by 20 to 60%, a substantial reduction that takes nothing necessary away from the customer.
Two caveats on scope. These figures come from tests and comparisons of mostly US and European sites. And they describe retail sites: a B2B checkout with purchase orders and customer accounts follows other constraints.
The reduction almost always happens the same way. First and last name merged into one field, the address completed automatically from the postal code. The second address line hidden behind a link, the billing address identical by default, the province inferred. None of these fixes removes information; they remove gestures.
Fees discovered too late
Once you set aside those who were browsing with no intention to buy, the top reason for abandonment remains extra costs, shipping, taxes and fees included, judged too high at the moment of discovery.
Everything hinges on the word discovery: the problem is not always the amount, it is the moment. A shopper who sees $14 shipping on the product page decides with full knowledge. The same shopper who discovers $14 at the second-to-last step feels they were led there under false pretences.
Three fixes, in order of effect. Announce the free-shipping threshold on the product page and in the cart, with the amount left to reach it. Show a shipping estimate before checkout begins. And never introduce new fees after the first mention of the total.
This logic lines up with the regulatory framework on price display, treated in our article on drip pricing. What is good conversion practice is also, for mandatory fees, an obligation.
Forced account creation
Of all checkout frictions, this is the most costly and the easiest to remove.
Requiring an account from a shopper who has already decided to buy means asking for an effort that does not serve them. They have to pick a password, sometimes confirm an email, for a transaction they wanted to close in two minutes.
The internal argument is always the same: the account enables loyalty and measurement. That is true, and it is not a reason to impose it before the sale. Offer account creation after the order confirmation, pre-filling everything the customer just entered. The sign-up rate you get at that moment surprises most teams.
In B2B, the question is different: the account carries negotiated prices and order history, so it has real value for the customer. Even there, separate the first order from account creation when possible.
Validation errors that make people leave
The form that wipes itself. An error that clears fields already filled in triggers an immediate abandonment. It is the most serious defect and it still exists.
The error message at the top of the page. The customer has to hunt for which of the twenty-three fields is the problem. The message belongs next to the field concerned.
The format imposed without saying so. A phone number rejected because it contains spaces, a postal code rejected without a space. Accept the formats, normalize afterwards.
The card declined without explanation. The customer does not know whether to fix a digit, switch cards or call their bank. A single sentence changes the outcome.
Validation that fires too early. A field flagged as an error before the customer has finished filling it in creates a feeling of permanent failure.
These five defects can be fixed in a few days of development and require no business decision. It is the best ratio of effort to revenue in this whole file.
Putting a number on recoverable revenue
The calculation takes four lines and turns a usability topic into an executive decision.
| Line | What you take | Scenario |
|---|---|---|
| Annual online revenue | Sales completed | $1,400,000 |
| Checkout abandonment rate | Measured from checkout entry | 68% |
| Revenue lost at checkout | Revenue divided by the completion rate, minus revenue | $2,975,000 |
| Targeted reduction in abandonment | A prudent twelve-month objective | 6 points |
A six-point reduction in this scenario represents about $61,200 in additional revenue per year, without one more dollar of media.
Two precautions. Revenue lost at checkout includes people who would never have bought, so never present it as fully recoverable. And a six-point reduction is a prudent objective when form load sits well above average, more ambitious when it is already low.
The order of the fixes matters, because the first ones cost little and return the most.
Validation errors first: they require no business decision and are settled in a few days. Then form load, aiming for a one-third reduction, which is almost always attainable. Then fee display, which requires a commercial decision on the shipping threshold. Finally the mandatory account, which touches internal processes and needs sign-off from more than one person.
Measure after each step, not at the end. Otherwise you will not know which one produced the effect, and you will not be able to decide whether the next one is worth it.
One last point of method: track the number of completed orders, never the click-through rate on the payment button. A fix that filters out over-budget shoppers earlier pushes the second down and the first up.
What to establish before increasing the acquisition budget
These five numbers can be pulled in one day. Without them, extra media budget feeds a checkout that loses part of what it receives.
- How many input elements does your checkout display by default?
- What is your abandonment rate measured from checkout entry, not from the cart?
- At which step does the customer discover the full amount, shipping fees included?
- Is account creation required before the order, and why?
- If the reduction in abandonment misses your target after two quarters, what changes, and who decides?
A good answer gives a field count and a rate measured in the right place. A weak answer quotes the industry's average abandonment rate. A vendor who proposes increasing your ad budget without having counted your form fields is selling media to offset a friction they never looked at.
Counting your form load and putting a number on recoverable revenue is part of what we deliver in a paid audit.
What is yours to own: the decision on the free-shipping threshold, the decision to remove the mandatory account, and the trade-off between the information you want to collect and what you need to deliver. What gets delegated: counting fields, measuring step by step, fixing validations, reworking templates and tracking completed orders. A business that fixes its five validation errors sees an effect within three weeks for a few days of development. A business that answers the problem with more advertising pays twice: the media, then the revenue it still loses at the last step.
What heat maps actually show about that friction is set out in what heat maps show.
This point sits inside the approach described in what conversion rate optimization really fixes.
The frictions specific to crossing a border are gathered in the three frictions of a first international sale, and declined cards abroad in declined payments abroad.
Turning a site into a tool that closes the sale rather than a catalogue that lets it slip is at the heart of the Improve your site's conversion goal.
The problem sits after the click and you already have a team? See our work on conversion rate optimization.
Other sales are lost well before checkout: the calculation is in our article on site search for a large catalogue.
Frequently asked questions about the checkout
Is the 70% abandonment rate a good benchmark?
No, because it mixes two populations. A large share of that abandonment comes from people who were browsing with no intention to buy and cannot be recovered. Measure abandonment from checkout entry instead, step by step. That number is lower, specific to your site, and it shows where to act.
Do you need a single-step checkout?
The number of steps is not what decides, the number of decisions asked is. A checkout with three light steps completes better than a single step overloaded with twenty-five fields. Count your input elements displayed by default before debating the layout.
Can you really remove the mandatory account?
In nearly every retail case, yes. Offer account creation after the order confirmation, pre-filling what the customer just entered. The sign-up rate you get at that moment surprises most teams, and the sale is no longer at stake. In B2B, the account carries negotiated prices and follows a different logic.
Which fields can you remove without losing information?
Most reductions remove nothing; they remove gestures: first and last name merged, address completed from the postal code, second address line hidden behind a link, billing identical by default, province inferred. The Baymard Institute observes that most sites can reduce fields displayed by default by 20 to 60%.
How long before you see an effect?
Validation fixes produce an effect within two to three weeks, because they touch situations that repeat every day. Reducing form load is measured over one or two months. Measure after each step rather than at the end, otherwise you will not know which one worked.
Which indicator should you track?
The number of completed orders, never the click-through rate on the payment button. A fix that filters out over-budget shoppers earlier, by showing shipping fees on the product page for example, pushes the second down and the first up. Judging on clicks leads to the wrong conclusion.
- Baymard Institute, UX Statistics, accessed July 2026. Source of the 23.48 form-element average, the optimal checkout at 12 to 14 elements, the possible 20 to 60% reduction, the 17% share of abandonment caused by a difficult checkout and the 42% share of shoppers who were browsing with no intention to buy. Data covering mostly US shoppers and sites.
- Baymard Institute, E-Commerce Cart and Checkout Usability Research, accessed July 2026. Source of the average abandonment rate compiled from some fifty separate studies.
- Falia working framework, recoverable-revenue arithmetic. Amounts are explicit scenarios, to be redone with your revenue and your measured rate.
Geneviève puts the strategy for your engagement into action. She leads all our web development projects: Shopify, WordPress and the new ways of building a site with AI. She manages our team of developers and translates your business needs into technical language. She runs your organic search (SEO), your visibility in AI answers (GEO) and your site's conversion rate optimization (CRO). Her work is at the heart of three goals: Attract customers with SEO and AI, Improve your site's conversion, and Strengthen your visibility in AI answers. With Gabriel, she also builds the landing pages for your advertising campaigns. She writes mainly about SEO, AI visibility and web design.
About Falia →
SEO
Paid advertising
Conversion
AI visibility
Web design