I’ve previously written before about the attribution problem
currently facing the online advertising industry: with ad servers attributing 100% of credit to the last impression served, the current ‘last-touch’ attribution system completely misses the contribution effect that results from other ad impressions that occurred before the last impression was shown. Multi-touch attribution is recognized as one of the most important problems in digital advertising, especially when multiple media channels, such as search, display, social, mobile and video are involved.
The industry needs to move to a multi-touch attribution system that recognizes and assigns value to all the advertising content that has influenced each consumer. We, at Turn, believe that this is a problem that we need to solve together and we want to share our research results with the world. In that attempt, this week, I attended the ACM International Conference on Knowledge Discovery and Data Mining
in San Diego, where I presented a paper on Data-driven Multi-touch Attribution Models. I took a dive into the math behind the statistical framework and modeling approach that allows Turn to create a data-driven methodology which produces consistent and accurate data-driven multi-touch attribution metrics.
To learn more about data-driven attribution modeling, download the white paper here