Hi SK,
The issue was tied to the Cleanup Task Data schedule. We had it set to keep a maximum of 500,000 rows in the database. Due to a large number of running tasks from several enabled policies, it was truncating the data after a short period of time because we were hitting that maximum. We updated to 1,000,000 and have now gotten more than 10 days' worth of data. We've also disabled some long-running policies to better control the number of running tasks. So things look better now, thanks.