What happens if a user who is lost according to this forty day time frame, I guess that can change for each customer, but what happens if they come back in after that time period? Are they still going to be considered churned and does that change the reports and the counts that I might see in my reports retroactively?
Yeah so I think that's a question that comes up pretty frequently and basically the idea here is that if a user has been marked as churned, we put them into this lost user bucket so these are all the users, the total count of users, that have been churned for the past 90 days. Now if a user has been marked as lost again we don't see any activity for 40 days for example, if that user say that they were totally inactive for the month of July but then they come back and they need to use the application for a new project at that point they will no longer be considered a lost user at that point we'll take that lost user and we'll mark them as an active user instead.
So say for example if you're doing some analysis on data from back in maybe April and you see that this number of lost users gets reduced, maybe you checked it last week and then you check it this week and this number is lower than it was. That's kind of to be expected again because some users might be re-engaging with the product and at that point they're moved from that lost bucket into the active account of users. Now if for example you see that there's a big swing so this count of lost users for a specific date range is changing very frequently what might be happening is that lost user threshold is either too high or too low. So again if that values changing frequently what you might want to think about doing is increase that user threshold because again what might be happening is a typical user is just infrequently used in the application so you might not want to necessarily declare them as a lost user because they might come back at a later point and use the application again maybe for a new project a few months later.
Now this also kind of ties into offline usage as well because say for example a machine goes offline for a period of time they're engaged with your product but because they're not connected to the internet, we're not collecting any data from them within the Usage Intelligence world. Now depending on how that machine is configured we're still logging data to that local cache file about that offline usage data but because we stopped collecting data from that machine what might happen is if they're offline for a very long period of time we might declare them as a lost user, despite the fact that again they're engaged at the product they're just using it an offline state. When that machine comes back online the later point in time what's going to happen is all that offline data is going to sync up with our hosted infrastructure and then at that point we're going to take that machine we're not going to mark that user as a lost user we're going to sync up all of that offline usage data that's going to be marked as an active user and you'll be able to see all the data associated with that installation in its offline activity within the dashboard. So that's also something to keep in mind if you see these values kind of fluctuate a little bit you know the count of lost users it's probably not cause for concern.
However again if you see those numbers fluctuate more significantly then you might want to think about configuring that lost usage threshold.
