Ways to avoid circular dependency

[KK111]

Hello! I am trying to create a calculated column by dividing one column from table 1 by another column from table 2. I use the following: 

=sumx(naturaljoin(table1,table2),column 1/ column2)

however I get circular dependency error. What can I add to my syntax to create a workaround?

the issue is that I cannot use a measure because I need to slice and dice my data and these two columns do not have direct relationship. 

Usually I use calculate and then all or removefilter function. But I can't use it because it this case I'm dividing kg by % and with calculate I have to use an aggregate function and I can't use an aggregate function with % 

thank you 

[V-lianl-msft · Accepted solution]

Hi @KK111 ,

You can try to use LOOKUPVALUE function to create a calculated column in Table 1, then use SUMX function to calculate.

Column 2 = LOOKUPVALUE(Table2[Sales],Table2[Product],Table1[Product])
Column 3 = SUMX(Table1,Table1[Column 2]/Table1[Quantities])

Best Regards,
Liang
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.