Redefining Attribution: Strategies for Low-Quality Data

I am going to share one of the methods I have used to make a good attribution of sales data, leads, etc. when the data is not very reliable or of questionable quality.

The method consists of the following: first, I classify the channels from most reliable to least reliable, according to the history of the quality offered in the data of each channel, or whatever criterion is considered most appropriate. Then, I assign the results to each channel in the established order, starting with the most reliable channels, until I reach the last channel and assign the rest of the results that could not be attributed to the previous channels.

This method has worked quite well for me when I have an internal database on the website that tells me the total number of sales, leads, or whatever KPI I need to measure. The ideal scenario will always be to try to get as much accuracy as possible in the data we get from the tools, but there are cases where this is not possible.

The main limitation of this method is that it does not consider the possibility that multiple channels may have contributed to a single conversion.