Form abandonment: how to measure it and find the question people quit on
Form abandonment is when someone starts a form and leaves without submitting it. Zuko, a form analytics company, reports that 34% of people who start a form don’t complete it. To reduce it, measure drop-off question by question, fix the question where most people leave, and start with the best-supported fixes: fewer fields, and errors shown as people go.
Last reviewed · Sources
What is form abandonment?
Form abandonment is when someone starts filling in a form and leaves before submitting it. The form abandonment rate is the share of people who started and didn’t finish: (starts − completions) ÷ starts.
The short version
- Abandonment rate = (starts − completions) ÷ starts. People who open a form and never answer haven’t abandoned it.
- A fair reference: Zuko reports that 34% of people who start a form don’t complete it.
- Measure each question: find the one where the most people leave, and fix that first.
- Best-supported fixes: cut fields, show errors as people go, and don’t ask for an account.
- Layout matters less: one page or one question per screen makes little difference to breakoff on its own.
Two things often get mixed up with it. People who open a form and never answer anything haven’t abandoned it; they never started, which points at the invitation or the page around the form. Cart abandonment in online shops measures a different action. Form drop-off is the same idea as abandonment, counted question by question.
| Measure | What it counts | How to work it out |
|---|---|---|
| Views | Visits that opened the form | Counted |
| Starts | Visits that began answering | Counted |
| Completions | Visits that submitted | Counted |
| View-to-completion rate | Completions out of everyone who opened the form | Completions ÷ views |
| Starter-to-completion rate | Completions out of everyone who started | Completions ÷ starts |
| Abandonment rate | Starters who left without submitting | (Starts − completions) ÷ starts |
| Question drop-off | People who left at one question, out of those who reached it | Left at the question ÷ reached it |
The two completion rates are Zuko’s definitions (opens in a new tab). The per-question measure follows the field abandonment rate in Zuko’s white paper (opens in a new tab), which divides the people who abandon at a field by everyone who interacted with it.
Form abandonment benchmarks you can trust
The most useful public benchmark is Zuko’s: “34% of people who start a form do not successfully complete it.” Zuko sells form analytics, so this is a vendor’s own figure, but it says what it divides by.
- Where it comes from: Zuko’s benchmarking insights (opens in a new tab), last updated 16 December 2025, which calls it Zuko’s most up-to-date figure. The post doesn’t state a sample size or period, so quote the 34% as Zuko gives it.
- What it counts: people who started a form, not everyone who visited it.
- Baymard’s figure is for carts, not forms: the Baymard Institute (opens in a new tab) puts the average documented online shopping cart abandonment rate at 70.22%, across 50 studies. That counts shoppers who put something in a cart and didn’t buy, a different action from starting a form, so don’t compare a form against it.
Baymard’s survey of why US shoppers abandon checkouts still has lessons for forms. Among shoppers who left for a reason other than just browsing, 18% left because the site wanted them to create an account, 17% because the checkout was too long or complicated, and 17% because of website errors or crashes.
Be careful with form statistics repeated across blogs. Many come from vendor surveys, or divide by views in one place and by starts in another. If a figure doesn’t say what it divides by, don’t measure your form against it. Your best benchmark is your own form last month, which is what Zuko’s white paper recommends too.
Why people abandon forms
The causes with the best evidence are too many fields, errors people can’t find, asking for an account or more than the task needs, and forms that are hard to use in general. Layout seems to matter less than people think.
Too many fields
Baymard’s checkout research found that “the number of form fields in a checkout impacts overall usability far more than the number of steps.” Nielsen Norman Group’s 2016 article Website Forms Usability: Top 10 Recommendations (opens in a new tab) puts it more bluntly: “Every time you cut a field or question from a form, you increase its conversion rate.” Each question has to earn its place, in an AI draft too: see how to check an AI’s first draft before you publish.
Errors that show up too late
In Baymard’s inline validation testing (opens in a new tab), errors that only appeared after submitting “contributed to form or checkout abandonment” across several rounds, especially when the error was hard to find. Baymard also found that “31% of sites don’t have any inline validation at all.” Zuko’s white paper shows the same pattern in a worked example, where about 42% of abandoners left right after clicking submit, often because of an error elsewhere in the form.
Asking for too much, too early
Forced account creation was the reason 18% of those US shoppers gave for leaving a checkout. The form equivalent is a sign-in wall before the first question, or personal questions before people know why you need the answers.
Forms that are hard to use
In a controlled eye-tracking experiment with 65 people, Seckler and colleagues (opens in a new tab) improved real company forms using 20 web form guidelines. The improved forms led to “faster completion times, fewer form submission trials, and fewer eye movements”, and people were more satisfied with them. The worst forms gained the most.
Follow-up questions that keep coming
How you arrange branching questions changes how many people quit. In a web-survey experiment, Chen and colleagues (2022) (opens in a new tab) found that asking filter questions together led to fewer breakoffs at the start than interleaving each one with its follow-ups. The conditional logic guide covers what that means for branching forms.
What matters less: the layout
Survey experiments comparing one long page with several shorter pages find small or no differences in how many people finish. Changing the layout is rarely the fix on its own; see one question at a time forms.
How to measure and calculate form abandonment rate
Count views, starts and completions, then look at drop-off for each question. The totals tell you how big the problem is; the per-question view tells you what to fix.
- Count views, starts and completionsTools define a start differently, so check what yours counts before you compare its numbers with anyone else’s.
- Pick a denominator and keep itUse starter-to-completion to judge the form itself, and view-to-completion to judge the whole journey, including the page or message that sends people to it.
- Find drop-off for each questionFor each question, count how many people reached it and how many left there. Drop-off is the number who left divided by the number who reached it.
- Don’t count people who are still typingGive a visit some idle time before calling it abandoned, or slow answerers will look like drop-offs.
- Read the partial answersIf your tool saves answers as people go, look at what people answered before they left. It often explains why they stopped.
- Compare with your own historyChange one thing at a time, then compare the same numbers over a similar period.
A worked example
These numbers are made up to show the arithmetic. Say a quote request form gets 500 views, 320 starts and 200 completions in a month.
- View-to-completion: 200 ÷ 500 = 40%.
- Starter-to-completion: 200 ÷ 320 = 62.5%, so the abandonment rate is 37.5%.
- Per question: 260 people reached “What’s your exact budget?” and 52 left there, a 20% drop-off and far more than any other question.
That one question is where to start. You might make it optional, or offer ranges with a “Not sure yet” choice, as Ribbit’s quote request form template does. Then compare next month’s drop-off on that question.
How Ribbit shows drop-off (Pro)
Ribbit Pro’s Analytics tab shows the funnel and the drop-off for every question. Tracking runs on every plan, so when you upgrade, the history is already there.
- Totals: Views, Starts, Submissions and Completion rate (submissions ÷ starts). A start is a visit that gets past the first screen, which on a form with a welcome screen means pressing Start, so compare Ribbit’s rate with its own history rather than with Zuko’s 34%.
- Visits over time: a chart for the last 7 days, 30 days or all time.
- Per question: how many people reached each question, and the drop-off percentage.
- Dropped or still going: a visit counts as dropped after more than 30 minutes idle on a question; anyone active more recently is shown as in progress.
- Partial answers: the Submissions tab filters by Completed, Partial or All, and the CSV export follows the filter. On an NPS survey, that keeps the scores of people who left before the follow-up; see how to run an NPS survey in Ribbit.
Drop-off analytics and partial answers come with Ribbit Pro: $29 a month, or $240 a year ($20 a month). As of October 2026, Typeform keeps drop-off rates for its Business plan, $129 a month or $91 a month billed yearly, while Tally and Youform include them on Pro at $29 a month. On Free you see every completed response. For the full comparisons, see Ribbit vs Typeform and Ribbit vs Tally.
See the question people quit on
Pro adds drop-off for every question and partial answers. Right now the first year is $150 with the early-bird price, applied automatically, then $240 a year.
How to reduce form abandonment: fixes ranked by evidence
Start with the fixes that have the strongest evidence behind them: cut fields, follow tested form guidelines and show errors as people go. Switching layout comes last.
| Rank | Fix | What the evidence says | Kind of evidence |
|---|---|---|---|
| 1 | Cut questions and fields you don’t need | Baymard: the number of fields affects checkout usability far more than the number of steps. NN/g: each field you cut raises conversion | Usability research on checkouts, and expert guidelines |
| 2 | Apply published web form guidelines | Seckler et al.: faster completion, fewer submission attempts, higher satisfaction | Controlled experiment with 65 people |
| 3 | Show errors next to the question as people go | Baymard: errors shown only after submitting contributed to abandonment | Usability testing, qualitative |
| 4 | Don’t make people create an account to answer | Baymard: 18% of US shoppers who left a checkout blamed forced account creation | Shopper survey, about checkouts |
| 5 | Group filter questions together | Chen et al.: grouped filter questions led to fewer early breakoffs | One survey experiment |
| 6 | Switch between one page and one question per screen | Paging and scrolling experiments found small or no differences in breakoff | Little effect either way |
Use these on the question where most people leave first. Your own per-question data is the most relevant evidence you have, and it tells you which of these fixes to try.
Templates to start from
- TemplateJob application form templateHire without the inbox chaos: every applicant’s contact details, role, CV link and start date land as one tidy row.
- TemplateVolunteer application form templateMatch volunteers to the right jobs, with each person’s availability, interests and experience, plus an emergency contact.
- TemplateSign-up form templateGrow your mailing list with a short sign-up that asks what people want to hear about and records a clear yes or no to emails.
- TemplateLead capture form templateQualify leads as they come in: role, company size, need and timeline, so your sales team calls the best-fit people first.
Questions people ask
What is a good form abandonment rate?
There’s no universal number, because tools divide by different things. Zuko’s data is a fair reference: 34% of people who start a form don’t finish it, so about two thirds do. Your own form’s history is the better benchmark.
How do you calculate form abandonment rate?
Take the people who started the form, subtract those who submitted it, and divide by the people who started: (starts − completions) ÷ starts. With 320 starts and 200 completions, the rate is 37.5%.
What is the difference between form abandonment and cart abandonment?
Cart abandonment counts shoppers who put something in a cart and don’t buy; Baymard’s average across 50 studies is 70.22%. Form abandonment counts people who start a form and don’t submit it. The two figures aren’t comparable.
How do I find where people drop off in my form?
You need tracking for each question: how many people reached it and how many left there. Ribbit Pro shows this on the Analytics tab; in other tools, look for drop-off or partial submissions, usually on a paid plan.
What are partial form submissions?
Answers saved as someone moves through a form, before they submit. They show what people answered before they left. In Ribbit, partial answers are part of Pro.
Does a shorter form always get more completions?
Cutting fields is the best-supported fix, but a question you really need is still worth asking. Cut what you won’t use, then check the per-question numbers after the change.
Sources
- Zuko, “8 surprising insights from Zuko’s benchmarking data” (opens in a new tab)zuko.io · checked
- Zuko, “Form performance” definitions (opens in a new tab)zuko.io · checked
- Zuko, “Using Data to Optimize your Forms and Checkouts” white paper (opens in a new tab)zuko.io · checked
- Baymard Institute, “Cart abandonment rate statistics” (opens in a new tab)baymard.com · checked
- Baymard Institute, “Usability Testing of Inline Form Validation” (opens in a new tab)baymard.com · checked
- Baymard Institute, “Checkout flow: average form fields” (opens in a new tab)baymard.com · checked
- Seckler et al. (2014), “Designing Usable Web Forms: Empirical Evaluation of Web Form Improvement Guidelines”, CHI 2014 (opens in a new tab)research.google · checked
- Google Research blog, “Simple is better: making your web forms easy to use pays off” (opens in a new tab)research.google · checked
- Nielsen Norman Group, “Website Forms Usability: Top 10 Recommendations” (2016) (opens in a new tab)nngroup.com · checked
- Chen, Cernat, Shlomo and Eckman (2022), “Impact of question topics and filter question formats on web survey breakoffs” (opens in a new tab)rti.org · checked
- Tally pricing (opens in a new tab)tally.so · checked
- Youform pricing (opens in a new tab)youform.com · checked
- Typeform pricing (opens in a new tab)typeform.com · checked
Related
- GuideOne question per screen formsWhat a one question at a time form is, what research says about completion, when a single page is better, and how to design one well.
- GuideConditional logic in formsConditional logic in forms explained: skip, branch and show/hide rules, six worked examples, the pitfalls that make people quit, and how to test each path.
- GuideNPS survey questionsThe official NPS survey question and 0–10 scale, how to score it with a worked example, 12 follow-up questions by score, and the mistakes to avoid.
- PagePricing
- TemplateQuote request form templatePrice jobs on the first reply, not the third: get the scope, budget, deadline and example links before you write a quote.
- TemplateContact form templateGive people one simple way to reach you. Four short questions, and every message comes with its topic, so the right person replies.