Negative average time to convert in funnel performance report. #453
Conversation
commcare_connect/program/helpers.py
Outdated
| @@ -50,7 +52,7 @@ def get_annotated_managed_opportunity(program: Program): | |||
| .annotate( | |||
| workers_invited=Count("opportunityaccess", distinct=True), | |||
| workers_passing_assessment=Count( | |||
| "opportunityaccess__assessment", | |||
| "opportunityaccess__assessment__opportunity_access", | |||
There was a problem hiding this comment.
this looks like it traverses a foreign key in two directions, can we not just skip the join entirely and count opportunity accesses that meet the filter below
There was a problem hiding this comment.
Fixed - 8b40e2d
commcare_connect/program/helpers.py
Outdated
| Avg( | ||
| ExpressionWrapper( | ||
| Greatest( | ||
| Subquery(earliest_visits) - F("opportunityaccess__invited_date"), |
There was a problem hiding this comment.
how does this end up negative? I see that we floor it at zero, but why is that necessary? Zero seems equally wrong if we trigger that condition
There was a problem hiding this comment.
This issue occurred because we recently added the invited_date field with auto_now_add=True. As a result, invited_date was set to now() during the migration for earlier records.
There was a problem hiding this comment.
that means the average will still be wrong right? since it includes a bunch of incorrect zeros mixed in with real data?
There was a problem hiding this comment.
Fixed and added test also -944785e
CCCT-570
CCCT-585