Data Modelling Challenge

Hi maruthi,

I would suggest that you use the query editor to get your fact_table to look like this

I did that with this code:

let
    Source = (enter Source Table here)
    #"Changed Type1" = Table.TransformColumnTypes(Source,{{"Date", type date}, {"Reach", Int64.Type}}),
    #"Unpivoted Other Columns" = Table.UnpivotOtherColumns(#"Changed Type1", {"Date", "Reach"}, "Attribute", "Value"),
    #"Extracted Text Range" = Table.TransformColumns(#"Unpivoted Other Columns", {{"Attribute", each Text.Middle(_, 7, 1), type text}}),
    #"Renamed Columns" = Table.RenameColumns(#"Extracted Text Range",{{"Attribute", "Speaker Number"}, {"Value", "SpeakerName"}}),
    #"Changed Type" = Table.TransformColumnTypes(#"Renamed Columns",{{"Speaker Number", Int64.Type}}),
    #"Filtered Rows" = Table.SelectRows(#"Changed Type", each ([SpeakerName] <> ""))
in
    #"Filtered 

(note I did not enter the email columns when copying your sample data as it was redundant so I would delete it, and I changed the date so that it would not produce errors on my computer when I changed the date column to Date type)

Then your data will be in a shape that Power BI can quickly and efficently evaluate over and compress, and your DAX formulas will be easier to write and the output you get will be what was expected.