How to calculate drop-off rate

The drop-off rate formula

A drop-off rate answers one question for a pair of adjacent funnel stages: what share of people did not make it forward? You get it from the step conversion rate, which is simply the later stage divided by the earlier one.

Step conversion = stage ÷ previous stage

Drop-off rate = 1 − step conversion

So if 30% of people move from one stage to the next, 70% dropped off. The two are the same fact seen from opposite ends, and they always add up to 100%. Run the calculation for every adjacent pair and you have a map of where the funnel leaks. There is a second number worth keeping beside it: the end-to-end (overall) conversion rate, which is the final stage divided by the first stage, not stage by stage. That one figure tells you how many people cleared the entire path.

A worked example

Take an ecommerce checkout funnel over a month: 12,000 product-page sessions, of which 3,000 add an item to the cart, 900 begin checkout, and 300 complete a purchase. Work each step in turn:

The biggest raw loss is at the top: 9,000 people leave the product page without adding to the cart. But notice that the highest drop-off rate, 75%, is also that first step, so reading the rates rather than the raw counts points you to the same weak spot for the right reason. The end-to-end rate is 300 ÷ 12,000 = 2.5%. To sanity-check it, multiply the three step conversions: 0.25 × 0.30 × 0.333 = 0.025, which is 2.5% again. To map your own funnel stage by stage, drop your counts into the funnel drop-off tool.

Drop-off rate vs bounce rate vs exit rate

These three get used as if they mean the same thing, and they do not. Drop-off is a funnel measure across two stages; bounce and exit rate are single-page, single-session measures from web analytics. Here is the clean distinction:

Bounce rate only counts a session as bounced when the entry page was the whole visit. Exit rate looks at any page and asks how often it was where the session ended, so a page can have a high exit rate without a single bounce. Drop-off ignores sessions entirely and tracks a population as it moves through steps you have defined. Use bounce and exit to diagnose one page; use drop-off to see where a multi-step journey loses people.

Why funnel steps multiply, not add

The end-to-end conversion rate is always lower than any single step because step rates multiply. A funnel does not lose 25% plus 30% plus 33%; each stage keeps only a fraction of the fraction the stage before it passed on. In the example above, 25% × 30% × 33.3% lands at 2.5%, well below even the weakest step. This is why a 2% overall conversion rate is normal for a several-stage funnel and does not, on its own, signal a problem.

The practical payoff is that a small gain at one step compounds through everything after it. Lift that first step from 25% to 30% and, holding the others steady, the overall rate rises to 0.30 × 0.30 × 0.333 = 3.0%: a fifth more customers from the same top of funnel. Because the rates compound, fixing the step with the worst step conversion usually returns more than chasing the stage with the largest raw drop. Read the step rates side by side, not the head counts. For any single stage, the conversion rate calculator turns that stage's counts into one conversion rate, so you can compare each step like for like.

What is a good (or high) drop-off rate?

There is no universal good or high drop-off rate, and any single benchmark number you see quoted is close to meaningless without context. A drop-off rate depends on two things: how many stages your funnel has, and how qualified people are at each one. A funnel with more steps naturally shows drop-off at every extra gate, and an early stage fed by cold, broad traffic will always drop more than a late stage of people who have already committed.

Judge each step two ways instead. First, against its own past: is this step's drop-off rising or falling versus last month? A step that used to hold 40% and now holds 25% is your signal, whatever the absolute number. Second, against its neighbours: the step with the worst step conversion in the current funnel is where the next fix belongs. For sales funnels, where stages run lead to opportunity to close, the same logic applies down the pipeline, and the lead-to-close calculator covers that final conversion. Treat "good" as relative to your own trend and your own stages, never as a figure copied from an industry chart.