Starting December 3, join live sessions with database experts and the Microsoft product team to learn just how easy it is to get started
Learn moreShape the future of the Fabric Community! Your insights matter. That’s why we created a quick survey to learn about your experience finding answers to technical questions. Take survey.
Dear Users,
I have a table of payments and a table of cases. although there exists a one to many relationship, RELATED does not seem to be working.
There are multiple payments for the same case with different payment dates.
what I am trying to achieve is to to extract the first payment date for any case. for this I used the following formula:
Payment Date = LOOKUPVALUE(Paid[Payment Date],Paid[Case-WE Month],Case[Case-WE Month], min(Paid[Payment Date]))
I am receiving the following results. as you can see that it is giving me dates which precede the date of the data itself i.e. the dataset starts from October 2022, whilst I am getting the invalid date of 2 October 2020. as a result, I am getting negative payment days as well.
what can I do to fix the issue?
This is a calculated column? or a measure?
Please provide sample data and show the desired result
I have cracked the code, thank you for your interest and responses.
This is a caluclated column and not a measure, please see the example with dummy data below, the desired result is in RED.
This is a calculated column?
LOOKUPVALUE needs an odd number of parameters (there are 4 in the provided code).
It should be
(return value, search column1, search value1, search column2, search value2........
the last expression in the code is for alternate result, that is why it is even number of parameters. considering the issue explained, you could advise me a better solution if the one I used is not.
User | Count |
---|---|
25 | |
13 | |
12 | |
11 | |
8 |
User | Count |
---|---|
44 | |
27 | |
19 | |
16 | |
12 |