group by and show a distinct value count

[amcneil32]

Hello, I need to group by and show a distinct value count. 

For the field "Cust_Subsegment" i would want as a result something like this

I've grouped by the distinct row count aggregate, but it looks like its just giving me the total row count for each distinct row. I need a count on the distinct values instead. thanks  

[amcneil32 · Accepted solution]

@Anonymous , i probably didn't  ask the question very well. I need to return all non-duplicate values. So from above I would need to return only 10199-MEM and  10199-NAS. I came up with a solution that appears to work. I first removed duplicates from the Merged field, which still returns one record for 1000269-TUL. I then duplicated the field, split by delimiter to get the number from the string. After that i just kept the duplicates on the number value.

[Anonymous]

are you trying to count the number of times Customer Segment appears? if so you could do something like this

distinctcount(name) then you would pull out the segment to the rows and add the distinct count to the values on a matrix viz

otherwise are you trying to count the uniqe combination of accountnum, name and segment?