Table of contents

• How do you calculate drop-off rate?
• Where does drop-off rate mislead?
• How do you reduce it?

How do you calculate drop-off rate?

drop-off rate = 1 - (visitors reaching step N+1 / visitors reaching step N)

Measured between adjacent steps, it is the complement of the step conversion rate. Measured across a whole funnel, it is the complement of the end-to-end conversion rate.

Two variants get confused and should be labeled explicitly:

Step drop-off compares against the previous step. Localizes friction.

Cumulative drop-off compares against the funnel entry. Shows total attrition to that point.

A funnel with four steps each losing 30% has a 30% step drop-off throughout and a 76% cumulative drop-off by the end. Reporting one while a stakeholder assumes the other is a routine source of confusion.

Where does drop-off rate mislead?

It assumes the sequence is real. If visitors can reach step three without step two, the calculated rate compares populations rather than measuring loss.

It ignores return visits. A B2B buyer who leaves at pricing and returns four days later to convert counts as a drop-off in a session-scoped funnel. On long consideration cycles this overstates loss substantially.

It treats all drops as failures. Some are correct disqualification. A pricing page that loses people who cannot afford the product is filtering, which is doing its job.

It is sensitive to how steps were defined. Splitting one step into two creates a new drop-off point without changing anything about the site.

Key takeaway: drop-off rate is only as meaningful as the funnel definition beneath it, and funnel definitions are choices rather than facts.

How do you reduce it?

Start by confirming the drop is real. A step whose event stopped firing looks identical to a step everyone abandons, and the difference between those two situations is the difference between a code fix and a redesign. Complete the flow yourself on the published site and watch the event land before you interpret anything.

Once the drop is confirmed, the diagnosis is the standard chain: segment by device and source, localize inside the step with form analytics or a scroll map, then watch recordings filtered to sessions that dropped. Funnel analysis sets that sequence out in full.

What a drop-off rate never supplies is evidence that your fix works. It locates the problem. A test decides the solution.

Related terms

Conversion funnel · Funnel analysis · Exit rate · Form abandonment · Bounce rate

Deeper reading: Diagnosing why users drop off using session recordings. Service: Conversion Rate Optimization.

FAQ

How is drop-off rate different from bounce rate and exit rate?

Drop-off rate is a property of a transition between two steps. Bounce rate and exit rate are properties of a single page. A process that never changes URL still has a drop-off rate, and a page that belongs to no funnel still has a bounce rate.

What is a normal drop-off rate?

The question has no answer that transfers between sites, because splitting one step into two creates a new drop-off point without changing anything a visitor experiences. The only stable comparison is one funnel definition measured over time, which means settling the definition before you start reporting on it.

Is drop-off rate the same as abandonment rate?

Close, and the difference is scope. Abandonment names a specific process the visitor left, as in cart abandonment or form abandonment, and is usually measured across that whole process. Drop-off rate measures one transition between two steps and applies to any sequence.