Counter reports should reflect usage for collection as it was on the dates selected (instead of as it was on the date run)
As it currently stands, if I run a report today covering 2018 usage for our consortially held collections, the usage totals only reflect usage for titles that are still in the collection (as well as newly added titles). It would be better if a report for 2018 usage included data for all the titles that were part of the collection in 2018. We have access to a couple large consortially held packages. The practice as it is currently in place results in a discrepancy of several hundred downloads in the BR2 count for Feb 2018 if we compare a report run in March of 2018 to a report run in September of 2019.