Understanding the logic in a DAX Statement, and which table to create the column - DAX Wizards Help!

[irnm8dn]

A few weeks ago someone in the forum was gracious enough to help me use and apply a "Related" function allowing me to compare two columns, from two different tables.  What I don't understand is why DAX statement works in a specific sequence. 

For instance:

Table A = A table of names that I know to be correct

Table B = A table of names that I know some to be correct, others not.

I want to compare Table A with Table B and used a DAX statement like this:

Match = If(Table B[Name] = Related(Table A[Name]), TRUE(), FALSE() )

What I don't understand is why PBI would not accept:

Match = If(Table A[Name] = Related(Table B[Name]), TRUE(), FALSE() )

It seemed more logical to me to use the latter because I know Table A to be the correct and the table to which I want to compare the information.  Perhaps I am understanding the operation incorrectly?

Also, PBI seems sensitive to which of the two tables I create the column in, and would like to undertsand why and what controls this.

[LivioLanzo]

Hi @irnm8dn

What does the relationship linking TableA and TableB look like?

[irnm8dn]

The relationship is based on a Campaign Name.  Table A had a unique instance of the Campaign Name and the customer it belongs to, Table B has the Campaign Name and Customer Name has repeating rows because it includes "daily" performance information.

The goal of the comparison is to basically identify Customer Names in Table B that do not match Customer Names in Table A where the Campaign Name is the same.

[irnm8dn]

Just looked, thanks.

Can you help me understand it a little more?  For instance, help me with the steps that would address this logic in a DAX statement.

If my table relationships were joined by Campaingn Name.

The customer name in Table A is the same as the Customer Name in Table B, AND the Campaign Name is the Same, True, else False.

Keeping in mind that Table A has unique rows of Campaign Name and Customer Name, Table B has uno-unique rows of this information.

Thanks!

[LivioLanzo]

@irnm8dn

Would you be table to post a sample data? And also, I am assuming you want to add a calculated column? 

[irnm8dn]

Links to sample data below.

Table A - represent the correct information what we regard as "True"

Table B - represents correct AND incorrect information non unique information

The goal is to relate the two tables by Campaign Name, and identify where the customers names represented in Table B are not consistant with what is in Table A.  The end result would  allow us to update the Customer Name in the "system of record" where the informaiton in Table B is found.

A PBI table would show 4 columns.  Every instance of the Campaign Name (from Table B) would be displayed in a PBI table. 

Campaign Name (from Table B) | Customer Name (from Table A) | Customer Name from (Table B ) | True/False Customer Name in Table A Matches Customer Name in Table B

Thanks for your help.

[v-frfei-msft]

Hi @irnm8dn,

As the picture, If you create a related column in Table A, there are more than one values will be returned. It is not possbile obviously.

Regards,

Frank

[irnm8dn]

Not sure I understand your response.  The table on the right is kind of what I am looking for, so two questions:

1.  Did Campaign Name come from Table A or Table B in the graphic on the right?

2.  What is a reasonable DAX expression to compare the column vaule called Customer Name which appears in both tables.

Thanks.

[LivioLanzo]

RELATED does not flow from the one side table to the many side table so you can only use it in Table B.

You should take a look at RELATEDTABLE