Hi @antoine.torri,

this is related to the different levels looking at the data. The eventtime is activity related, SO creation date is case related. You have to make sure to stick to the same levels. We frequently see this challenge when using a PU-function combined with a time dimension. Were you already able to solve the issue?

Best,

Justin

@antoine.torri You can try using ROUND_MONTH(PU_FIRST(VBAP,

"_CEL_O2C_ACTIVITIES".EVENTTIME, "_CEL_O2C_ACTIVITIES".ACTIVITY_EN IN (<%=changedFields%>))) to get the timestamp of the first change activity per case. This should solve the increasing numbers problem.

Best

Kevin

Hi Antoine,

Add as dimension ROUND_MONTH("_CEL_O2C_ACTIVITIES".EVENTTIME)and as KPI Count cases affected with changes

COUNT(DISTINCT

CASE WHEN "_CEL_O2C_ACTIVITIES".ACTIVITY_EN IN (<%=changedFields%>)

THEN "_CEL_O2C_ACTIVITIES".CASE_KEY ELSE NULL END

)

The KPI will count unique Sales order Items that had an associated changed activity per month of occurrence.

Best,

Gabriel