Marketing Mix Modeling, Media Mix Modeling, Marketing Effectiveness, Experimentation, Causal Inference, Adstock, Marketing ROI, Statistics, Machine Learning, Marketing Attribution, Media Planning, Marketing Budget Optimization, Robyn, Multi touch Attribution, First Party Data, Privacy Proof Marketing Solutions.
Is there one true Marketing Mix Model (MMM) model ?

Is there one true Marketing Mix Model (MMM) model ?

Is there one true Marketing Mix Model (MMM) model ?
Is there one true Marketing Mix Model (MMM) model ?

Is there one true Marketing Mix Model (MMM) model ?

Short answer: Yes

“All models are wrong, some are useful”

I have seen various vendors use the above statement to tell client that “there could be many MMM models and no one model is absolutely right”. They thus push the decision to pick a model back to the client.

I think vendors saying this is a cop out.

First lets try to dispel the misunderstanding behind this statement “All models are wrong, some are useful.”

“All models are wrong, some are useful” is an aphorism (meaning it is a concise expression of general truth). But the aphorism in this case leads to misinterpretation.

Firstly, it is important to understand what modeling is.
The purpose of modeling is to provide an abstraction of real process. Basically, a good approximation of reality.

Anybody who mistakes the abstraction for the real, commits the Fallacy of Reification (yup, one more fallacy to add to the list of all fallacies which we data scientists/statisticians commit).

In an exact sense, a map is also wrong because it does not provide 1:1 mapping of the real world.

So, George Box’s phrase should be construed the same way as a map is considered wrong because it does not represent the real world.

Does a map provide 1:1 mapping to the real world?
No.
But is it mighty useful?
Hell yeah.

There is definitely a route that takes you to the destination as precisely as possible.

Coming back to MMM.

In MMM we deal with historical data, meaning the sales (or any KPI) has already been realized. Now the million dollar (literally) question is what lead to this sales? What combination of factors led to the sales.

Now because this is all in the past, there is only one set of combination which could have resulted in the sales. The job of a MMM vendor is to find out this combination of factors that led to the observed sales. So technically it is honing on the truth.

There are no multiple ways through which sales could have been gotten. This part is the scenario planning and budget planning exercise where once we have a ground truth understanding of what moved the needle of sales, we then try to see how we can tweak the combinations of marketing inputs to yield a better sales number.

In the latter, many different models are possible.

To summarize:

There is one true MMM model. Vendors should help clients find that model. We at Aryma Labs always do.

In scenario planning / Budget optimization, there could be multiple models.

 

Facebook
Twitter
LinkedIn

Recommended Posts

Chebyshev’s Inequality for Marketing Mix Model Diagnostics

Chebyshev’s Inequality for Marketing…

At Aryma Labs, we constantly endeavor to add as much science as possible…

How to use Robyn’s…

In my last post (ICYMI link in resources), I talked about the similarities…

Similarities between Decomp RSSD and Bayesian Priors in Marketing Mix Modeling (MMM)

Similarities between Decomp RSSD…

Open source Marketing Mix Modeling (MMM) tools are great for democratizing MMM. But…

Scroll to Top