They capture the interactions on the web page and "replay" them later in a deep analytic-esque session. The console is also available.
To me it looks like that the software is storing the whole document as state (including the mouse location) and uses some API to make the console available as well.
The stored data is in time-series format.
Later this stored data can be used to:-
I was wondering if anyone is already aware of how this is possible?
Divyendu Singh
I get the job done!
j
stuff ;)
Well in theory they just track all events of the user and if you have starting even and the last tracked event you have a timeline.
you then stream these events via a listener/websocket to your endpoint and record it there.
this would be 1 solution :) the interesting part would be do they store the data and just apply it? or do they actually create the videos instantly?
both is possible but 1 is less storage.