Form abandonment statistics, and what each one counts

Published form abandonment statistics run from about 30% to 70%, and they disagree because they divide by different things. Baymard Institute’s average cart abandonment rate is 70.22%, from 50 studies between 2006 and 2025. Zuko’s 2025 benchmark puts starter to completion for all forms at 51.71%, which means roughly half of the people who start a form never submit it.

Three denominators, three answers

Every abandonment figure is a share of some starting group, and the groups are not the same:

• Cart abandonment starts from shoppers who put something in a cart. Baymard compiles 50 published studies, from 2006 to 2025, and averages them to 70.22%. Many of those shoppers were never going to buy, which is why the number is so high.
• Starter to completion starts from people who interacted with the form, usually by focusing a field. Zuko’s 2025 table puts all forms at 51.71%.
• View to completion starts from everyone who loaded the page with the form. Zuko reports 37.2% on desktop and 31.3% on mobile.

So a form can lose 30% of its viewers before anyone types, half of its starters before they submit, and still sit inside a sector where carts are abandoned seven times in ten. Before comparing your number with a benchmark, check that both divide by the same group.

What the checkout research blames

Baymard asks US online shoppers why they abandoned. The largest group, 42%, say they were just browsing. Leaving those aside, the reasons are mostly about cost and effort:

• 40% said extra costs such as shipping, tax or fees were too high.
• 18% said the site wanted them to create an account.
• 17% said the checkout was too long or complicated.
• 12% could not see or calculate the total cost up front.

Baymard’s 2024 benchmark also counts the fields in a checkout: an average of 11.3, down from 12.7 in 2019, while it says most sites need only 8. The page does not state the sample size for that figure, so it is best read as a direction of travel rather than a precise industry average.

What the form research shows by industry

Zuko’s 2025 table is drawn from its form optimisation database and reports overall conversion by industry. Property is the lowest at 34.55% and media close behind at 36.29%. Travel, at 60.51%, is the highest figure Zuko does not qualify; forex trading shows 67.35%, but Zuko says many of those forms were multi step forms logged as single forms, so true completion is lower. Healthcare has the lowest view to completion rate, 21.4%.

The pattern makes sense. Property and healthcare forms ask for the most personal and documentary detail, such as income, identity, history and medicines. Travel forms ask for things people know by heart. Effort, not industry, is what moves the number.

What is not published

Several figures that appear in round ups could not be traced to a page that states them, so they are left out here. Zuko does not publish a single “average form abandonment rate”; the 30% to 35% figures quoted elsewhere are conversions of its completion rates by third parties. Baymard does not give the sample size or year of its abandonment reasons survey on the public page. Neither source measures non commercial forms such as consent or intake forms separately.

How the forms on this site compare

The 280 form examples on this site ask a median of 12 questions, one more than Baymard’s average checkout, and the 30 survey forms among them ask 8.5. The online form statistics post breaks those counts down by industry. They are finished examples, not live forms, so they carry no abandonment data of their own; what they show is how long a complete form for each job tends to be.

Three changes the studies agree on

1. Remove fields you will not act on. Baymard’s gap between 11.3 fields and the 8 most checkouts need is the clearest finding in the set. The GOV.UK service manual makes the same case for every form: know why you need each answer.
2. Show costs and commitments early. Late surprises drive abandonment more than length does.
3. Split long forms into short steps with a visible progress indicator, and measure each step. The multi step form best practices post covers how to split, and how to increase form completion rate covers the fixes in order.

The response rate against completion rate page explains the two measures, and the multi step form page defines the format.

Measuring your own abandonment

A benchmark is only useful beside your own number, and most small businesses do not have one. Three counts are enough to start. Count how many people load the page with the form, which any page analytics tool reports. Count how many submit, which the responses table gives you. If your analytics can record the first interaction with a field, count starters too. Divide submissions by page views for view to completion, and by starters for starter to completion, and you have the two measures Zuko publishes.

Then look for the step where people leave. On a paginated form, compare the number who reach each page. On a one page form, look at the last field people touched before leaving, if your analytics can see it. The usual culprits are a field asking for something people do not have to hand, such as a policy number or a document, a question that feels intrusive before the purpose is clear, and a long text box early in the form. Moving the hard question later, or making it optional, often recovers more completions than any design change.

Measure for a few weeks before and after each change, and change one thing at a time. A form that loses ten people a week will not show a clear difference in a few days.

In the product

A classic form can be paginated into steps with a progress bar in the accent colour and a Page X of Y badge, and a conversational form asks one question per screen with a progress bar. Nothing is stored before the respondent submits, so the product records completed responses only; to measure abandonment you need page analytics on the form’s page. The multi step form builder page shows the paginated layout, and the guide to paginating a classic form shows how to split one.