This idea is related to a question I posted : Incorrect attribution of conversions to Target activities in Analytics(?)
It would be great to be able to choose the attribution period. Currently there is no expiration, which means that conversions are attributed to a past activity even long after this activity was live. So, doing your Target activity evaluation one week after your test or one month after your test or 2 months after your test might result in ending up with a wrong conclusion related to that test (because of being impacted by many other external factors, campaigns & tests).
Some people might now think "but you can do this in the Segment builder by creating a segment on visit level and selecting Activity (Analytics for Target) exists THEN conversion exists" (cfr next screenshot).
But no.. the result will still be that conversions are attributed to activities that were active even long ago. Reason: the IF condition 'Activity (Analytics for Target) exists' does not expire. It does not refer to the actual (one and only) impression.
Ideally we would be able to select the actual impressions only and then select/define the evaluation period that is required for a certain test. This can be next hit, same visit, within a week,... since the type of conversion (a click or a sale) probably defines the length of this evaluation period.