I have a single-page application, and I am tracking events on the page, in Google Analytics, passing our internal user ID and timestamp as attributes on the event.
I would like to know the best way to find the top "bounce" (UPDATE: "exit" is a better term for this) events - i.e. events that are more than averagely likely to be the last event that a user fires.
I think I can do this by extracting the raw event data from the GA API, and simply running my own analyses on it, using the timestamps. Is this correct? (UPDATE: It seems not, because GA doesn't allow the export of raw event data.)
Is there any way to do this within the GA interface? I think I'm right in saying that the Event Flow won't show me the overall exit rates for a particular event.
I'm looking to create information like this:
- create new document: last event 10% of the time it was fired
- edit document: last event 5% of the time it was fired
- turn on track changes: last event 80% of the time it was fired
So that I can spot which events look problematic, like the last one in this example.