Relationship many-to-many issue. Is technically a many-to-one

[Anonymous]

Hi all,

So I'm perplexed and wondering if the good folk in the community can assist! I've attached some photos, with (hopefully not a too bad) explanation of what I'm attempting to do, and my output.

So, I'm wondering why after 'removing duplicates' in power query, it's keeping the badly formatted names with the good one 
(ie: It's keeping Joe McDonald, Joe Mcdonald, and Joe McDonald [with a space after McDonald]) which is exactly what I want, but then relating my fact tables, it's producing many to many?

Fact tables (5 of them) have manually entered names, some badly written.

Dimension table = append all 5 tables, keep only request name and team, remove duplicates.

In my pbix, I try and relate the fact tables to the dimension and receive many-to-many warning. It theoretically is many-to-one as they are not identical.

Below pictures attempting to explain. Thank you in advance! 🙂

[mussaenda]

Hi @Anonymous , 

Do you have null values on you filed that you are using for the relationship? if yes, that is what causing the many to many detection

[Anonymous]

Only one, because I'm looking to report on those that haven't filled in their name too.

I did to filter out blanks in power query, as there was a blank and a null value in power query, so I just left null and filtered out blank. Do you think that's what causing it? Or would a singular null value cause this too?

[mussaenda]

Hi @Anonymous ,

Yes. Assign anything to your null and do  the relationship again