Conversion funnel drop off: find the handoff

A funnel report counts visitors who reached each named step. When the count falls sharply between two steps, the loss is usually in the gap, where the page changes. A button can open a new tab, a redirect can drop the parameter that tells your analytics where the visitor came from, a consent banner can cover the next button, and a confirmation email can leave the site altogether. Each of these looks like a visitor who gave up. Some of them are a page that never loaded for that visitor, and the funnel report has no way to show the difference.

• Each pair of steps: clicks on the first step set against views of the page it opens.
• Redirects: does the page after a redirect keep the campaign tag and the signed-in state?
• New tabs: does a step open in a new tab, which some tools count as a new visit?
• Confirmation emails: does the link arrive, open, and land on a page that works?
• Banners on a phone: does a consent box sit on top of the next button?

Write the steps in order: the button, the page it opens, the form, the submit, the confirmation, the email, the link in the email, and the first screen of the product. Each neighboring pair is a handoff, and each handoff has a count on each side: the clicks on the first step and the views of the next. A large gap between them means the page the button opens did not load, opened somewhere else, or did not register as a view. If the form-to-submit gap is small but the submit-to-confirmation gap is large, the submit sends visitors somewhere that never shows a confirmation.

A redirect is a handoff with no page of its own. If the redirect drops the tag that says where the visitor came from, the next page is counted as direct traffic, and the funnel loses the link to the ad or search that brought them. A new tab does something similar: the step opens somewhere else, and the visit on the first tab ends. To check, click through from the button and look at the traffic source your analytics records on the arriving page. If the source changes to direct, or to your own domain, the handoff broke the count, the visitor, or both.

A confirmation email is the hardest handoff to see, because it leaves the site. The visitor may open it on another device, the mail may land in spam, the link may expire before anyone clicks it, or the link may ask for a sign-in the visitor never set up. Your funnel stops counting at the send and starts again only when the link lands on a page your tags can see. Mimiq's simulated visitor has a working inbox, so a run can open the confirmation link, follow it and show whether the first screen loads. The report gives the step where the email handoff stopped, which is the one place the funnel cannot show you.

Onboarding flow testing

Redirects, sign-in changes and email senders are often added late, after the funnel was last checked. Rerun the full path from the first button to the confirmed account after each change to a redirect, a tag or an email template, because the gap can move even when no page changed. Check the test inbox each time, since a confirmation that lands in spam is a drop no page can show.