Moving values to a new column based on another column

[park]

Hi.

Consider the situation. I have column CompanyName that are associated with rows of another coumn called Type. These are "Chain" ,"NF", "FF", and "DF"

CompanyName	Type
Walmart	Chain
Costsco	NF
Publix	NF
Bar	FF

Consider that Costco and Publix are both Chains, but read as NF. Which is wrong.

How do I create a calculated column or expression that places these names into the right Type, "Chain." This is going to be on a large scale of data with multiple other company names as well.

Thank you! 

[tackytechtom]

Hi @park ,

Maybe something like this?

You could create a nested if statement with all your conditions. Here an example

Here the code:

if [CompanyName] = "Publix" or [CompanyName] = "Costco" then "Chain" else [Type]

Alternatively, you could create an own mapping table according to your definition and then you merge the mapping table with your large table on CompanyName.

Let me know if this helps! 🙂

/Tom
https://www.tackytech.blog/
https://www.instagram.com/tackytechtom/

[park]

Thanks! This look like it might work. I will try in a second 🙂

Will this work for approximate matches too? Like lets say I have Publix #1 - Publix #100

thanks! 

[tackytechtom · Accepted solution]

Hi @park ,

The above code won't work with approximate matches. Try this one instead:

Here the code:

if Text.Contains([CompanyName], "Publix") or Text.Contains([CompanyName], "Costco") then "Chain" else [Type]

Hope this helps 🙂

/Tom
https://www.tackytech.blog/
https://www.instagram.com/tackytechtom/

[park]

Thank you! And if I have lets say 20 different company names, I can continue the OR statement as many times as needed? Or does it only work for two arguments 

[tackytechtom]

Hi @park ,

This should work for as many arguments. Please keep in mind that someone needs to update that code in case new companies are added. A better way would probably be to fix that data issue in the source itself (just saying 🙂 )

/Tom
https://www.tackytech.blog/
https://www.instagram.com/tackytechtom/

[park]

Thank you for the solution. This works great ! 🙂

Is this argument case sensitive as well? If so, is there a way to get around?