If GA4 records generate_lead when a required field is empty, the trigger is measuring an attempt rather than a successful lead. Move the measurement to a confirmed form outcome. Renaming the click event does not make its underlying signal more accurate.
Define what counts as generated
Write one sentence with the form owner: for example, a lead is generated when the server accepts a valid inquiry and the configured intake system stores it. A later qualification or sale is a different business stage.
Google's recommended-event reference defines generate_lead around the lead being generated. It does not make every submit-button interaction a lead. Avoid assigning a made-up monetary value just to populate a parameter.
Run a negative test before changing tags
On an authorized staging or test path, submit an empty required field, then an invalid email, then one valid synthetic inquiry. Use the form response, stored entry where available, and tag-debugging timeline together.
| Action |
Form outcome to observe |
Lead expectation |
| Click with missing required value |
Rejected |
No success event |
| Server rejects spam or validation |
Rejected |
No success event |
| Valid accepted inquiry |
Accepted |
One intended event, subject to consent |
| Reopen a confirmation view |
No new inquiry |
No new lead solely from reopening |
A form_submit observation is not automatically equivalent to the business definition. Confirm what the installed plugin and measurement integration actually emit.
Invalid field: No success event. Valid inquiry: One intended event. Revisit: No new lead by itself. Consent: No forced grant. Explanatory checklist, not a customer test result.
Connect to the plugin's real success contract
Inspect the published GTM trigger or native integration. A broad click selector, generic DOM submit listener or confirmation-text visibility trigger can fire too early or repeatedly. Use the form plugin's documented success integration for the installed version.
Record whether the event means the browser received a success response, the entry was saved, or the CRM accepted a record. These can occur at different times. Do not describe an email notification as delivered solely because the form accepted the submission.
Send only the non-sensitive parameters needed for measurement. Do not include names, email addresses, message text or access-bearing URLs in analytics events. The existing Meta lead-event guide covers the separate destination if both platforms use this form.
Check repeats and consent
Test a double click, a failed attempt followed by success, and a back-button revisit. One successful action should not be counted by both an old click trigger and the replacement success trigger. Retire only the redundant measurement path once the new one is verified.
Repeat under the site's approved consent states. Do not force consent to granted to produce a debug event. Document expected suppression or consent-mode behavior separately from form acceptance.
Keep the before-and-after event definition with the reporting notes, because a corrected count may drop without a drop in real inquiries. HandL WP can audit form conversion triggers against the actual saved outcome.
References reviewed October 13, 2026. Examples are explanatory, not customer test results.