Best Practice: ROAS calculation: Combining Commercial Data & Media Data

stseebauer

Hi Guys,

I am struggling with the following use case: I have Ads data containing entity dimensions  Campaign Key, Media Buy Key, Creative Key. I can extract the advertised product id (SKU) from Media Buy Name. I have some Measurements incl. Media Cost.
I also have order data with product id (SKU), revenue, units, etc.
I mapped product id as product key in both data streams.

My task is to calculate ROAS on product and media buy level.
Calculating ROAS (Sales Revenue/Media Cost) works fine on Product Level.
BUT: I cannot slice the data on eg Media Buy Level.

Do I have a misconception?
Anyone of you having an idea/concept/best practice to achieve this?

Best
Stefan

Alex_Deliysky

Hi @stseebauer ,

Thank you for sharing your case. It appears that your question is too case specific, I recommend logging a ticket to our team at [email protected] or directly through the Datorama Platform.

Kind regads,
Aleks

Marie

Hi Stefan, Aleks, 

I think this case is not too specific, because it's been several times I've faced this issue without finding a solution. The "less specific" pattern is that we retrieve "standard" data from the adserver, and "conversion" data, either from the adserver floodlight report, or from the website directly. In the conversion data, we have a product dimension, which has a many-to-many relationship with the Media Buy Key. 

And the problem starts when the client wants to know what is the ROI/ROAS per product sold. Is it even possible to do it (and how ? ) or the only way to get closer to this is to use the taxonomy to calculate this based on the diffusion product ?

Thank you,

Marie