After months drowning in data it’s such a relief to arrive at an accurate and plausible Marketing Mix Model, that sometimes I forget what I built it for in the first place: the Channel Multipliers!
MMM and attribution both have their own opinions on what channels are performing...More
You did a really good job with Rockerbox
I was impressed with the demo, but now I’m actually seeing our data, I have all sorts of questions about what’s driving performance for us
I’m also thinking about what other analyses this enables
For example I’ve been wanting to test out Marketing Mix Modeling, and I know Rockerbox have a template that works with their sheets sync
Can you try it out and show me what you find?
This course is a work of fiction. Unless otherwise indicated, all the names, characters, businesses, data, 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.
Marketing measurement vendors like Rockerbox provide a way to get all of your data in one place, which is traditionally used for testing out different attribution models, like last click or data-driven. However Marketing Mix Models can also benefit from this consolidation of data in one place. Normally MMMs take so long to build, and it’s so exciting when you finally arrive at a robust and accurate model, that it’s easy to forget what we did it all for in the first place! Comparing it to your existing attribution is the really actionable part of the model. Of course this is only good for making real world decisions, if you have a valid model (and that’s a bigger topic). However, assuming that you’re happy with your model, and it’s both accurate and plausible, you can use the multiplier tab to make decisions about channel optimizations. For example if we look at facebook - instagram, and the modelled revenue is 3.19x higher than what was being reported on a Last Click, we know our analytics is likely underreporting the value of the channel significantly. That might make sense given that Meta is no longer able to accurately track conversions since iOS14 (relying on Apple’s SKAN), and because last click tends to underweight upper funnel channels like Facebook and Instagram, where many sales may come from viewing an ad and then searching the brand term, instead of clicking. If the multiplier was negative, that would be the model telling us that the channel is claiming more credit than it deserves, and performance should be down weighted accordingly. These decisions make a big difference if you’re optimizing six or seven figure budgets across multiple channels: getting more accurate with your allocation can save you millions of dollars!
Complete all of the exercises first to receive your certificate!
Share This Course