Filter out customers who don't buy certain products

[BrianJ]

@robH,

OK, I get it. I am sure this is doable, but I think it’s going to be a relatively heavy lift because counting blank rows is a fundamentally different calculation than the other iterations over your Income Groups table.

The approach that comes to mind is based on the solution @Nick_M developed to address the “Top N and Others” question:

Top N and an Others bucket

i’ve worked on something like this before. It can be done, but it’s not as easy as you would think. Here’s the outline of what I have done. There are notes in the code, but hopefully will be able to follow along: /*These measure will take the TopN of a dimension ( City in this case) and return the top N individual valuesl //along with a new row called "Remaining" with the remaing total //Then a rankx measure will sort the cities from highest to lowest with the Remaining row always at the botto…

I think the outline of the general approach would be:

1). Using your current measure, replicate the info in your matrix visual virtually as a table variable

2. Develop a measure to count your blank rows, probably using COUNTBLANK, or COUNTROWS(ALL) minus COUNTROWS(ALLNOBLANKROW)

3. Using ROW, build a row incorporating the info from 2) and matching the structure of your virtual table in 1).

4. As another variable in the same measure as 1), append the row created in 3) to the table in 1) using UNION

5. write the DAX to output the relevant info from your complete virtual table in 4) to your report visuals as the RETURN statement in the measure created in 1) and 4)

As I said, doable but not easy… (unless someone has figured out a way to iterate over blank rows in a parameter table that I’m unaware of).

@Nick_M’s solution above is well-commented, so should provide a nice adaptable framework if you decide this is a road you want to go down. And the forum is always here to help…

• Brian