Your model doesn't make sense: Facebook ads can't drive negative sales!? If we can 'tell' the model that result isn't possible with Bayesian priors, our model will be much more believable.
Based on Experience
Vetted industry professionals, that have experience working with these companies and others, consulted in an advisory capacity to ensure the content of this course was realistic.
This is a work of fiction. Unless otherwise indicated, all the names, characters, businesses, places, events and incidents in this course are either the product of the author's imagination or used in a fictitious manner. Any resemblance to actual persons, living or dead, or actual events is purely coincidental.
How do you answer the question "How can Facebook drive negative sales?" – learn how Bayesian MMM can help you find the answer!
Traditional marketing mix models can mistakenly categorize a channel as having a negative impact on sales, which isn't likely. With Bayesian Markov Chain Monte Carlo models (MCMC) you can rule out a negative coefficient in your model by setting Bayesian priors. Read More
This Course Includes:
1. Too many cooks spoil the broth
After initial wins on Facebook and a successful launch of TV ads, it was time to try TikTok. Performance looks good, but there's a problem: it has broken your measurement model!
2. Modeling with the bumpers up
One way we can protect against implausible model parameters is to use Bayesian MMM, where you can set informative priors that rule out unlikely values.
Frequently Asked Questions
Learn about Bayesian MMM today.
Discover how to solve this problem: How can Facebook drive negative sales?