Variance report using Metrics visual and static table

[JustinDoh1]

I am trying to create a variance report using a Metrics visual and a static table (that I am thinking about having a static data in SQL).

I created a metric visual as shown below:

These numbers were generated based on Count of outcome for each rows.

I have an illustration of data in a similar fashion (tabular) that I am trying to set up as a static table in SQL as shown below:

This data/table will eventually be used to do the variance report (find the difference on same category).

These data are not created in a table yet.

Now, I am trying to find the best approach to prepare data for the second table.

I don't think having a dummy data that repeats each row (just to create a Count for Aggregation) makes sense.

I guess the issue is how to compare data, one in visual (that have been aggregated in metrics) and other that have not been prepared yet.  

Thank you for help!

[v-rzhou-msft · Accepted solution]

Hi @JustinDoh1 

I think you want to build a static table in data model like second screenshot.

You can try to use Power Query to transform your data model. I build a sample like yours to have a test.

Sample:

Matrix:

Firstly we need to use Group By function to get Count column as below.

Then we pivote "Matrix Column" as below to get the result.

Best Regards,

Rico Zhou

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

[v-rzhou-msft · Accepted solution]

Hi @JustinDoh1 

I think you want to build a static table in data model like second screenshot.

You can try to use Power Query to transform your data model. I build a sample like yours to have a test.

Sample:

Matrix:

Firstly we need to use Group By function to get Count column as below.

Then we pivote "Matrix Column" as below to get the result.

Best Regards,

Rico Zhou

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

[JustinDoh1 · Accepted solution]

@v-rzhou-msft 

Thank you so much for your feedback.

I ended up doing this process:

1) Created a static table (Table2) for the second data.

2) Created a joined key (called 'JoinKey') for the two tables (Table1 and Table2).

3) I joined Many to Many relationship for the two tables (using the joined key 'JoinKey').

4) Tried with Quick Measure (for the first time 🙂 ).

It generated a measure like this:

Count of JoinKey minus Count of JoinKey = COUNTA('table1'[JoinKey]) -

COUNTA('table2'[JoinKey])

It worked somehow.  🙂

I have a question. What is risk of having Many to Many relationship?