Marketing Mix Modeling using Robyn open-source package • Data Science Festival

Marketing Mix Modeling using Robyn open-source package by Igor Skokan & Leonel Sentana

Open Source Automated Marketing Mix Modeling – Project Robyn We believe in opening the black box, transparent and customizable models that democratize econometrics & MMM and solve for analyst bias and subjectivity, so we’d love to share our work on “Robyn” = an open source automated Marketing Mix Modeling (MMM) code developed by Meta (Facebook) Marketing Science. It aims to reduce human bias by means of ridge regression and evolutionary algorithms and allows ground-truth calibration to account for causality. And it’s all open source! We have technical data science details as well as business case studies from companies who adopted the technique to share.

The code is available via our microsite: https://facebookexperimental.github.io/Robyn/ and in this GitHub repo: https://github.com/facebookexperimental/robyn

Supported by