Hi,
Could you help me come up with a solution for the following scenario?
In general I want to display if there is change in both the previous and current values for data records and then count.
For example, I need to count all customers whose values have changed between the selected dates.
My usecase is to compare the data for each combination of customer_id, segment, and category. If there is a difference in value between the selected dates, calculate the distinct count of (customer_id & category).
Ex 1: Selected dates: 2/1/2012 (new) and 1/30/2012 (old),
As the value changed for customer_id : CS-2490 and Segment = Furniture, from "Binders A" to "Binders" then count as 1 i.e. countd (customer_id & category) and ignore Consumer/Art records as there is no change in the values.
If there no change, then do not count this custid.
Ex: 2
Dates: 2/1/2012(new) & 1/27/2012(old) .
for custid CS-2490 As there is change in the value(Null to Binders for Home Office and no record for consumer in old ) then count should be 2. i.e. countd(cust id&category)
I want the calculation in such a away that the logic should be part of calculation instead of having it as a filter pane.
Output.
- Total number of unique custid's i.e. countd(cust id&category)
- Total number of unique custid's by Segments
- Display the records that are changed only using action filter.
Let me know if you need further clarification or examples.
Hi, @Prakash Desai
Sorry for the late reply. I may not have understood it correctly yet, but what do you think of the attached content?
This is processed using the LOD FIXED function.
<sample>
*If you get the best results from this exchange, I would appreciate it if you could choose the best answer or upvote.
