Drop-off Analysis
Drop-off analysis identifies the specific steps in a funnel where users abandon the process, quantifying how many leave at each stage and helping diagnose why.
In depth
Drop-off analysis works by computing the percentage of users who proceed from one step to the next, then highlighting the stages with the steepest declines. A single high-drop step is often worth more attention than several small leaks, because fixing it compounds gains for every step downstream. Pairing the numbers with session recordings or exit surveys turns a "what" into a "why," which is what makes the analysis actionable.
The classic mistake is treating all drop-off as bad: some abandonment is healthy disqualification, especially in lead generation where you want unfit prospects to self-select out. In a quiz funnel, drop-off analysis tells you whether people quit because a question is unclear, the quiz feels too long, or the lead-capture ask comes too early, so you can separate harmful friction from useful filtering.
Example in practice
Frequently asked questions
Is all drop-off a problem?
No. Some abandonment is healthy when unqualified visitors self-select out of a lead funnel. The goal is to reduce drop-off caused by friction or confusion, not the drop-off that filters out poor-fit prospects.
Which step should I fix first?
Start with the step that has the steepest decline and the highest downstream value, since fixing it compounds across every later stage. Use session recordings or surveys on that step to understand the cause before changing anything.
How does drop-off analysis relate to funnel visualization?
Funnel visualization is the chart that displays each step and its conversion rate, while drop-off analysis is the interpretation of where and why the biggest losses occur. You typically read the visualization first, then drill into the worst drop.