0

I have two separate data sources with effectively the same columns

  • Date
  • Vendor
  • Campaign
  • Ad group
  • Keyword
  • Impressions
  • Spend

What I am trying to do is stitch the data together in a fact table in order to get a combined view of Impressions and Spend. However, as I am dealing with 2.5 million rows, as soon as I make a concatenated key to join the data =[Date]&[Vendor]&[Campaign]&[Ad Group]&[Keyword], my file size jumps from 11Mb to 100Mb, I assume due to the high level of the cardinality in my new column.

What I am wondering is, how can I get join this data together that would remove the need for this concatenated key? I have a date table in place if that helps, but would a method that would utilize a measure instead of a calculated column would be ideal.

Many thanks, Chris

Improve this question

1 Answer 1

Reset to default
1

I would add a Power Query layer using the Append command to combine the 2 fact tables into one. The combined table would be what is presented in Power Pivot.

Impressions rows will have blank Spend values, and vice versa. Your chosen aggregation for the 2 measures should take care of that.

Improve this answer
1
  • Simplicity is the best. Can't believe I didn't think of that. Thanks so much for the insight!
    – Chris
    Commented Mar 7, 2016 at 19:54

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .