Form abandonment is when a visitor starts a form and leaves before submitting it. The abandonment rate is the share of form starts that never become submissions. It is one of the most useful numbers on a lead generation site, because the people who abandon a form have already shown interest. They clicked, they typed, and something stopped them.
Most sites cannot see it, because they track only the submit. To measure abandonment you need at least two events per form, a start and a successful submission, and ideally one per step or field. Once you can see where people leave, the causes are usually plain: a question asked too early, a field people do not want to fill, a gate before the payoff, or a technical error.
This guide covers how to measure abandonment, how to read it, the fixes that address each cause, and what you may and may not do to recover people who left.
Why form abandonment matters more than form views
A visitor who never sees your form tells you about placement. A visitor who starts and leaves tells you about the form itself. That second group is closer to becoming a lead than anyone else on your site, and the reasons they left are usually fixable.
The numbers can be large. In a calculator funnel we audited for a client, 59 to 63% of visitors left at a form placed before the result. Those visitors had done the work of answering the calculator questions and were then asked for contact details before seeing their number. That pattern points to one fix before any redesign: move the form after the result.
Our own sites taught us the same lesson. Ten Foxyapps tools asked for free text and an email before showing anything, and over about two years they were loaded about 8,500 times and produced 21 leads, about 0.25% (ViralSales inventory, November 2026). We replaced them with checks that show the answer first.
How to measure form abandonment
The minimum: start and submit
Track two events per form:
- Start: the first interaction, such as focus on the first field or a click on the first answer.
- Submit: a submission your server accepted. Fire this from the success response, not the button click, or validation errors count as submissions.
Abandonment rate = (starts minus accepted submissions) divided by starts.
GA4's enhanced measurement can collect form_start and form_submit automatically, but it can fire on forms you do not care about, such as search boxes, and can miss forms that submit through JavaScript without a page load. For lead forms, push your own events into the dataLayer and trigger GA4 tags from Google Tag Manager. The form tracking guide walks through that setup.
Better: one event per step or field
For multi-step forms, push an event each time a step is reached, with the step number. For long single-page forms, push an event when a visitor leaves each field (on blur), with the field name only:
js
window.dataLayer.push({
event: 'form_field_complete',
form_id: 'quote_request',
field_name: 'company_size'
});
Never send the typed value. Field names and step numbers are enough to find the problem, and typed values can contain personal data the visitor did not agree to share with an analytics tool.
In GA4, build a funnel exploration with one step per field or screen. The step with the largest drop is your first fix.
A ViralSales check counts views, starts, completions and leads separately, so each step's drop is visible without building the funnel yourself. It also pushes start, complete and lead events into the site's own GA4 or dataLayer after the site's consent banner.
Common causes and their fixes
| Where people leave | Likely cause | Fix to test first |
|---|---|---|
| Right after starting | The form looks longer than expected | Split into steps, or cut fields |
| At the email or phone field | Asked before the visitor got anything | Move contact details after the result or summary |
| At one specific question | Unclear, intrusive or hard to answer | Turn it into multiple choice, make it optional, or remove it |
| At the submit button | Validation errors, captcha, or an unclear outcome | Show errors inline, simplify the captcha, rename the button |
| Mostly on phones | Fields or buttons too small, keyboard covers content | Test on a real phone, use the right input types |
| After a page reload | Answers lost on error | Keep entered answers on validation failure |
Asking too early
The most common cause on lead generation sites. Visitors will answer questions about their situation, then stop when asked for an email before they see anything back. Show the score, estimate or recommendation first. The email then becomes a way to keep it.
Free text where a choice would do
Typing a description of your problem is hard work on a phone. Most qualifying questions can be multiple choice with four or five options and an "Other".
Hidden technical failures
Some abandonment is not abandonment. The submit request fails and the visitor gives up. Ad blockers are one cause: they match words in request paths, and counted on 16 November 2026 across EasyList and related lists, 1,410 rules match /track, 318 match /events and 292 match /collect. A form plugin that posts to a path with one of those words can fail for visitors with a blocker, which looks exactly like a person changing their mind. Submit a test entry from a browser with a blocker on, and check your server logs for failed requests.
On WordPress, plugin updates can break forms or tracking without warning. A scheduled check from WP Commit can confirm forms and tags still work after updates.
Winning back people who abandoned
There are two kinds of recovery, and only one is safe by default.
On-page recovery: always fine
- Keep answers in the browser so a visitor who leaves and returns finds the form where they left it.
- Show a short reminder on their next visit: "Want to finish your site check? You were on question 4 of 6."
- Offer a shorter path for those who stall: a calendar link instead of the full quote form.
Follow-up by email: only with consent
Capturing an email that was typed but never submitted, and then emailing that person, is a privacy problem in many jurisdictions and an unwelcome surprise everywhere. Only follow up on partial entries when the visitor explicitly agreed to be contacted at an earlier step, and your privacy policy says so. If you are not sure, do not do it.
Abandonment benchmarks you should set yourself
You will find plenty of published averages, and most of them mix contact forms, checkout forms and signup forms into one number that fits no one. A more useful benchmark is your own: measure each form for two weeks, record the abandonment rate per step, and then judge every change against that baseline. Change one thing at a time so you know which change moved the number.
The multi-step form guide covers how to order steps so the drop stays small, and the CRO audit guide puts form abandonment in the context of the full page.
Where to start
- Add a start event and a server-confirmed submit event to your main lead form, and calculate its abandonment rate over two weeks.
- Add step or field events (names only, never values), build a GA4 funnel, and find the single step with the largest drop.
- Fix that one step: move contact details after the result, turn a free text field into multiple choice, or remove the question, then compare against your baseline.