Forum Discussion
Finding / Displaying Mismatch Between 2 Columns in Different Tables
Hello, was looking to recieve some advice for this scenario.
I'm looking at 2 text columns from 2 different tables and need to compare them and print the ones that do not match. It needs to display the differences from the 2 filtered visuals. Visual 1 uses a column in Table 1 and Visual 2 uses a column in table 2. Both tables are recording the similar information but is recorded in 2 different systems, so the objective is finding the deltas between the 2 systems and displaying it.
| Visual 1 | Visual 2 | Visual 3 (Deltas) | |
| NAME | NAME | NAME | |
| A | A | D | |
| B | B | F | |
| C | C | ||
| ABC | D | ||
| ABC | |||
| F |
Thanks in advance.
6 Replies
- Zubair_MuhammadCommunity Champion
- AnonymousNot applicable
Thank you for the response, but this will not work for me. The 2 tables do not have the same amount of columns. The objective is, in Visual A it is showing a filtered data set from Table 1, while in Visual B it is also using the same filters, but using a data set from Table 2. And from those 2 filtered data sets, i need to display the fields that mismatch.
I believe I need to use a measure to properly do this, but where i get stuck is that these are text fields, therefore i cannot aggregate them to be used in a measure. Using the MAX of each text field in order to be able to use a measure will not work either since it is only taking the field with the largest length.
- v-lili6-msftCommunity Support
hi, Anonymous
Based on my test, you may try to use NOT and IN Function to add a measure as below:
Step1:
Add a name fact table
(You could use this formula add a new name table
Name = DISTINCT(UNION(VALUES(Table1[Name]),VALUES(Table2[Name] ))))
Step2:
Create a measure like this:
Measure = var _table=CALCULATETABLE(VALUES('Table1'[Name])) return CALCULATE(MAX('Name'[Name]),FILTER('Name',NOT('Name'[Name]) in _table))Step3:
Drag the name field from name table
and then drag the measure into visual level filter and set filter is not blank
Best Regards,
Lin