Join us for the next Community Q&A Coffee Break on Tuesday April 23, 2024 with Eric Matisoff, Principal Evangelist, Analytics & Data Science, who will join us to discuss all the big news and announcements from Summit 2024!
Attribution Model With Multi Dimesnion Breakdown
Hello Team, I am reaching out to you to get your thoughts and suggestions on how to achieve a reporting view.
Requirement : We would like to run an attribution model against Marketing channel to understand how much each channel is contributing towards revenue. We would further like to see the order ids which are part of this contribution and the associated SKUs / Product IDs for these orders.
Observation : If we run any attribution model with a look back window against marketing channel, it gives credit to each of the channel based on the model we choose. But if we breakdown the channel with Order ID and further by products, attribution models start adjusting the breakdown dimension (i.e. order id and products) too based on the model algorithm.
Problem Statement : While breaking down the channel with order id and products, since the attribution model logic is working on each of these dimensions, there might be cases where a particular product falls under an order id (based on attribution model) but actually the product was never part of that Order.
Screenshot1: Default reporting
Screenshot2: with same attribution model across dimensions(breakdown) [5th product is not part of the order]
Screenshot3 : with last touch model for breakdown dimensions
If we use last touch model at the breakdown levels, it gives ~1:1 result, however since lookback window is present, is it possible that it still shows few products which do not fall under this particluar order? If that's the case, what would the work around!
Appreciate your thoughts and suggestions.
