The Prestige-ification of Marketing Measurement – Are you watching closely?

Few weeks ago I read a post titled “why is Bayesian MMM so popular”.

There are so many articles like that which hype Bayesian methods as though they are a silver bullet.

But do you really want to know why Bayesian methods are popular in Marketing Measurement and perhaps in other fields?

It is not for the right reasons.

One particular quote from the famous movie ‘Prestige’ sums up the reason

“Now you’re looking for the secret, but you won’t find it, because of course you’re not really looking. You don’t really want to work it out. You want to be fooled”

Every time a Bayesian MMM vendor presents their models, you wonder how did they do it? what priors were set? what is the basis for those priors?

But at the same time you are also not actively looking to probe them because you want to believe that the methodology works. Even though it doesn’t !!

So you convince yourself by looking at ‘scientific looking charts’ that are nothing but simple sampling from a distribution charts. A couple of
normal distribution curves thrown in to give you familiarity. A few charts that scream ‘we quantify uncertainty’.

You want to believe in all of the above not because they are valid but because it is convenient.

📌 The DAG story

Another field where Prestige-ification has started to happen is Causality.

Again just a few days back a causal expert and I shared notes on a article published by open source marketing measurement library creator.

The article had numerous DAGs. All looking beautiful and scientific. But on deeper probing we found :

– Incorrect arrows.
– Missing confounders.
– Conceptual errors in temporal ordering.

But did the audience care? 100’s of likes on that post just because ‘it just looked causal’

Marketing Measurement buyers are supposed to make life a little bit difficult for us vendors. But unfortunately it doesn’t happen.

In marketing measurement today, aesthetics are often rewarded more than accurate attribution.

Solutions that ‘look scientific’ are bought rather than ones that truly are scientific.

Sometimes it is such a sinking feeling to see a method enjoy scrutiny free popularity.

I see Bayesian methods adoption as hamster running fast on a wheel. Yes the wheels are turning faster but the hamster is going nowhere. I believe scientific progress too will be limited to ‘what is already known’ if more Bayesian methods are adopted. We will be stuck in our own priors.

Can change be brought in Marketing Measurement and other domains to scrutinize Bayesian methods more?

I think the starting point would be with the question – “Are you watching closely”?