Strategies for messy data and building relationships

[condale777]

I have a bigger model going from multiple systems where I am trying to filter by "company". The problem is that company is different in each system.For example, one will be "Jim's Sporting Goods" and on another table it will be "J Sporting Goods".  I would like to create a dimension table to be able to slice the multiple fact tables without creating virtual relationships. I guess the functionality I am looking for would be like a lookup formula. 

Does anyone have any strategies for creating this dimension table in the model or do I have to use virtual relationships? 

[MFelix]

Hi @condale777 ,

Believe that the best option in this case is to have a table with the ID of each company, and add that field on your tables then you can make a relationship by the ID instead of the name or creating virtual relationships.

If you compile all the names into a excel file, a sharepoint list,  a table somewhere then you can do the merge by name and get that data into each table directly, only maintaning a single spot for the renames.