It could be said that there are some that are already outdated even though they are still practical. However, the ease of use can often mask flaws in conversion data and mislead your company's analyses.
Attribution models like 'First click', 'Last click', 'Time decay' and 'Position-based' distribute conversion credit in ways that may not make sense for the data your business needs to evaluate.
According to writer, professor and one of the world’s leading authorities on web analytics, Avinash malaysia phone number example Kaushik, traditional models are not recommended and do not have as many positive points as models that involve Data Driven. “But if you are a genius, you can customize your own model,” he explains.
In comparison, while Data Driven attribution models provide customer centricity, economic KPIs for marketing actions and operational efficiency, the traditional models mentioned above are lacking in these aspects, with the only visible advantage being implementation time.
Media Mix Modeling
Following this philosophy, we developed a customized attribution model that gives credit to all campaigns , providing real indicators. This is the MMM, or Media Mix Model, which gathers information within a single database and performs a data science project, allowing you to know the conversion percentage of each media.
Using data from a real case with Media Mix Modeling, in the first month of operation it was possible to see that 20% of media investments were made in vain, and using it we obtained a 15% reduction in CAC.
It is important to note that the content of this article was taken from our participation in the CMO Summit, where Bruno Dihl Prolo , Consulting Executive Director at MATH, was one of the speakers. Bruno gave a talk that highlighted the need to let go of old attribution models.