Measure
Conversion rate calculator
Form completion and conversion rates from starts, views and submissions.
Forms
Form abandonment is when someone starts filling in a form and leaves before submitting it. Because these people already showed intent, they are the cheapest conversions to win back. This guide covers how to measure form abandonment, the usual causes, 10 fixes, and what Convertica has learned from rebuilding lead generation forms.
Form abandonment is when a visitor starts an online form, such as a sign-up, quote request, lead form or checkout, and leaves before submitting it. Reduce it by removing fields you do not need, asking easy questions first, using one column with labels above fields, validating inline and explaining why you ask for personal details.
Form abandonment is when a visitor starts an online form, such as a sign-up, quote request, lead form or checkout, and leaves before submitting it. It matters because these visitors already chose to act: something in the form stopped them. Each one is a lead, sale or sign-up you nearly had.
Web forms are where visitors turn into leads and customers. A newsletter sign-up starts the relationship, a lead generation form starts a sales conversation, and a checkout form takes the money. That is why fixing a form is often one of the quickest wins in conversion rate optimization: the people who abandon it were already close.
Form abandonment rate is the share of visitors who start a form but do not submit it: form starts minus form submissions, divided by form starts. Count a start when the visitor first interacts with a field, not when the page loads, or the rate will mix in people who never meant to fill it in.
| Metric | Formula | Example: 2,000 views, 400 starts, 100 submissions |
|---|---|---|
| Form start rate | Starts / form views | 400 / 2,000 = 20% |
| Form completion rate | Submissions / starts | 100 / 400 = 25% |
| Form abandonment rate | (Starts minus submissions) / starts | 300 / 400 = 75% |
| Form conversion rate | Submissions / form views | 100 / 2,000 = 5% |
The example numbers are only there to show the arithmetic. Compare your own form with itself over time, and by device, rather than with someone else's average. The conversion rate calculator works out the rates for you.
People abandon forms when the effort or the risk feels bigger than the reward. The usual causes are too many fields, personal questions asked without a reason, unclear value, confusing errors, privacy and spam worries, forms that are hard to use on a phone, and steps that appear without warning. Most of them can be fixed.
Track form abandonment by recording two events, a form start and a form submission, then adding field-level data that shows where people stop. Google Analytics 4 gives you the first part. A form analytics tool or session recordings give you the second, which is what tells you what to fix.
form_start, the first time a user interacts with a form in a session, and form_submit, when the user submits it. Check that both fire on your form: custom and multi-step forms often need events set up by hand.To reduce form abandonment, remove the fields you do not need, ask easy questions first, use one column with labels above fields, validate inline with clear errors, explain why you ask for personal details, support autofill and mobile keyboards, show progress, write a button that names the result, save progress, and test each change on real traffic.
For each field, ask what happens if you do not collect it today. If the answer is "nothing much", remove it or ask later, after the visitor has become a lead or a customer. Mark the few optional fields that remain as optional.
Start with simple choices that need no typing and no personal data, and leave contact details for the end. Visitors who have answered a few easy questions are more likely to finish. This was the idea behind Convertica's Empire Flippers form, below.
A single column gives one clear path down the form. Labels above fields stay visible while the visitor types, unlike placeholder text that disappears. Group related fields, such as an address, under a short heading.
Check each field as the visitor leaves it, mark the problem next to the field and say in plain words how to fix it. Never clear the form after an error. WCAG 2.2 success criteria 3.3.1 and 3.3.3 require errors to be identified in text and fixes to be suggested where they are known.
Next to a phone number, say what you will use it for and what you will not. Near the button, say how you handle personal data and link to your privacy policy. A short line of reassurance answers the worry before it stops the visitor.
Use the standard HTML autocomplete values, such as name, email, tel and postal-code, so browsers can fill in known details. Use type="email", type="tel" or inputmode="numeric" so phones show the right keyboard.
If a form has several steps, show how many and where the visitor is. Splitting a long form into short steps can make it feel lighter, as the guide to multi-step forms explains. Do not hide a surprise extra step at the end.
"Get my quote" or "Send my free guide" tells the visitor what they get. "Submit" tells them nothing. Put the button directly under the last field, make it the most visible element in the form and keep it within thumb reach on a phone.
On long forms, save what the visitor has entered in their own browser, so an interruption or an accidental page reload does not wipe it. If you want to keep partial entries on your server, check your consent and privacy obligations first.
Fix clear problems such as broken validation straight away. For changes you are unsure of, run an A/B test judged on completed submissions, and on lead quality where you can measure it. The A/B testing guide covers how.
Rebuilding a long, static form as a step-by-step flow can cut form abandonment sharply. Convertica did this for the Empire Flippers valuation tool, a lead capture form on a marketplace for buying and selling online businesses, and the original case study reports a 51.6% increase in conversions in 47 days.
The old form showed a page of empty fields for business and financial details. The new flow asked one easy question at a time: multiple choice, yes or no, and sliders for monthly figures, with a progress bar and no Next button. In the case study's words, "the user is 40% done before they have to enter any data." A later change moved the main controls within easy reach of a thumb on mobile.
+51.6%
Conversions (valuation tool submissions), 47 days
Source: Empire Flippers, valuation tool submissions, 47 days, split-URL test then rollout
Lead generation forms convert best when they ask only for what the next step needs, start with easy questions, sit next to a clear reason to fill them in, and say what happens after submission. Every qualifying question should earn its place: ask it only if your sales team actually uses the answer to decide what to do.
Shorter lead forms usually bring more leads, and more of them unqualified. Longer forms bring fewer, better-qualified leads. Decide which your business needs, then measure the result in your CRM, not just in form submissions. Following CRM best practices keeps that data clean enough to trust. Progressive profiling helps: ask for the basics now and collect the rest in later conversations or forms.
Put the form where the decision is made: next to the offer and the proof, near the top of a short page, or repeated after the key sections of a long one. In a Convertica test for a home services lead generation site, a quote form that opened from call-to-action buttons was tested against the same form shown in the page, so visitors could take in the message first. The popup version increased successful bookings in that test.
Forms that ask one question at a time, like a conversation, can work well for quotes and valuations, where the visitor needs to give several details. They work best when every step is quick to answer and progress is visible. For a two-field newsletter sign-up, they add steps for no reason.
| Area | Do | Avoid |
|---|---|---|
| Length | Only the fields the next step needs | Collecting everything "while we have them" |
| Order | Easy questions first, contact details last | Phone number as the first field |
| Layout | One column, labels above fields | Side-by-side fields and labels inside the boxes |
| Required fields | Mark the few optional fields as optional | A star on every field and no explanation |
| Errors | Inline, next to the field, saying how to fix it | A red banner at the top after submitting |
| Button | Names the result: "Get my quote" | "Submit" or "Send" |
| Reassurance | Privacy line and what happens next, by the button | No word on how the details will be used |
| Mobile | Large fields, the right keyboard, button in thumb reach | Tiny fields and date pickers that are hard to tap |
Follow up with people who abandon a form only where you have their consent or another lawful basis to contact them. Capturing what someone typed before they pressed submit can count as collecting personal data. Fixing the form usually wins back more people than chasing them afterwards, and it carries no privacy risk.
AI agents can fill in a form when it is built from standard HTML fields with visible labels and autocomplete values, and when nothing requires a step only a person with a mouse can pass. The same choices make forms easier for people using assistive technology, so they are worth making regardless of how many agents visit.
WCAG 2.2 success criterion 1.3.5, Identify Input Purpose, asks for exactly this kind of machine-readable field purpose, and 3.3.8, Accessible Authentication (Minimum), asks that sign-in not depend on a cognitive test such as remembering a password or solving a puzzle unless an alternative is offered. Both apply directly to a login form. Agent readiness is new, and nobody has years of data on it yet, including Convertica. It is one of the eight checks in the free CRO audit.
Use this form abandonment checklist on any form: sign-up, lead, quote or checkout. Fill it in yourself on a phone, then answer each line yes or no. Every "no" is a fix to make or a change to test.
Work out your form rates and plan tests with Convertica's free calculators.
Measure
Form completion and conversion rates from starts, views and submissions.
Plan a test
How many visitors each form version needs before a test starts.
Read a result
Whether a finished form test is a real result.
Form abandonment is when a visitor starts filling in an online form, such as a sign-up, quote request, lead form or checkout, and leaves before submitting it. It is measured as the share of form starts that do not end in a submission.
Form abandonment rate is the number of visitors who started the form but did not submit it, divided by the number who started it. For example, if 400 visitors start a form and 100 submit it, 300 abandoned it, and the form abandonment rate is 300 divided by 400, or 75%.
The most common causes of form abandonment are too many fields, requests for personal details such as a phone number without a reason, unclear value, confusing error messages, worries about privacy or spam, forms that are awkward on a phone, and surprise extra steps.
Reduce form abandonment by removing fields you do not need, asking easy questions first, using one column with labels above the fields, validating inline with clear error messages, explaining why you ask for personal details, supporting autofill, showing progress on longer forms and testing each change.
Often, for longer forms. A multi-step form that starts with easy questions and shows progress feels like less work and commits visitors step by step, as in Convertica's Empire Flippers case study. For a short form of two or three fields, one step is usually simpler. Test it on your own form.
A good lead generation form asks only for what the next step needs, starts with easy questions, says what happens after submission and how the details will be used, uses a button that names the result, and sits next to the reason to fill it in. Ask qualifying questions only if your sales team uses the answers.
Google Analytics 4 enhanced measurement can record form_start, the first time a user interacts with a form in a session, and form_submit, when the user submits it. Comparing the two gives a basic form abandonment rate. A form analytics tool shows which field people leave at.
Only where you have a lawful basis and the visitor's consent where it is required. Saving what someone typed before they press submit may count as collecting personal data, so check the rules where you and your visitors are. Saving progress in the visitor's own browser, for them to finish later, avoids that problem.