It seems like the 30th day doesn’t exist in the database or doesn’t load?? Which causes it to look like my game has gotten 700%+ players when it clearly isn’t true (it’s actually been declining 
)
These are the filters I put:
This is just an acknowledgment announcement!
We’ve filed a ticket into our internal database for this issue, and we will update you when we have further information!
Thanks for the report!
1 Like
Thanks again for reporting these issues and following up to clarify in private messages. For these metrics, we’re displaying the full limits of our retention, and that does mean two things you noticed:
- The first data point shown will be incomplete as we don’t have data from its full time interval.
- We don’t have any way to calculate meaningful period-over-period comparison chips, so the ones that we display are quite wrong.
Since these are definitely misleading, we’ll work on fixing them.
The data dropout you saw for data between 6/13 through 6/17 seems like it is fixed now, or more accurately that period is now out of retention so I can’t check on it.
If anything like that recurs, though, please do comment here or open a new report as that might indicate an unexpected data loss that we need to address urgently.
One more suggestion is that if you are doing analysis over longer periods of time, you can use our Engagement data sets, which have longer retention than the Performance page. You could look at the Daily Active Users metric over the past year as a replacement for Concurrent Users. And the Average Session Time also has a year of data retention in the Engagement page.
They’re aggregated differently so that we can provide more demographic information and longer retention than in the performance page, but hopefully they can serve your needs in this case.
1 Like