How we made the most of 150 data points

When it comes to marketing effectiveness, analysts often rely on digital advertising. But the reality is a mix of digital plus offline, which is provided by monthly aggregation flights and 150 rows of data over three years. In such conditions, classical ML models give negative weights to working channels, and time-series models answer the question "what will happen?" rather than "what were the sales due to." In a real-world case report, I'll show you how to use Bayesian regression in PyMC Marketing to build a causal, interpretable model of the contribution of TV, radio, OOH, and digital to sales. I'll figure out how to validate the model when the data is "classically insufficient" and why we believe in this model.