Skip to main content

I’m confused how this funnel works. With “Any Order’, I would assume that the order of A and B are irrelevant, because it is in “any order”. However, when I swap the two events in the chart setup, the beginning and the ending are completely different numbers. I would understand if the beginning numbers are different, but the ending numbers turn out to the be the same amount. 

Am I misunderstanding something here? Or is it some kind of bug? 

 

Excuse me, just to clarify. 

The ending numbers are NOT the same. If they were the same, I would understand it. But the fact that they are different makes it seem that “Any Order” is not true. 


Hello @liukaichi !
According to Amplitude’s article “Build a funnel analysis”:

Any order is exactly what it sounds like: as long as a user completes all the steps listed, Amplitude will consider them converted. Users who fire the later events without firing the first event are not considered to be part of the funnel; they have to fire the first event for that to happen.

I tried it on my own with different steps that I swapped, and the final conversion amount is almost the same, but there is indeed a small difference each time (regardless of if I’m counting in Unique Users or Totals). 
It does not impact my analysis as the differences are relatively small, but I agree that there is something we are missing here 😂


Hi @Matrak @liukaichi In an 'Any Order' Funnel Analysis chart the first and last step matter. Users must do the first event to be considered on the chart. If the user has never done the first step they will not enter the Funnel Analysis. The number on that event reflects the number of users who have performed the first step. Users must do all the events in the Funnel to be considered in the last step since no step can be skipped. So although users can complete the events in any order they must have completed all the other steps and the last step in the funnel to count as converted for the last step. 

 

This might be why you’re seeing small differences in the numbers.


Reply