Ghost MMM: Modeling Your Competitor Without Their ‘Complete’ Data
For those of you who are F1 fans, you will recollect a cool analysis F1 came up with a year ago – Ghost Car Lap.
This comparison was designed to analyse qualifying performance of cars, enabling analysts to overlay a transparent car representing the fastest lap over the current driver’s run to show where time is gained or lost.
Couple of days ago, I was chatting with a very senior CMO and he asked me whether we model competition along with the brands’ regular MMM.
Yes, sometimes we do.
In domains like CPG, you can indeed build what I call a Ghost MMM.
A model where the dependent variable your competitor’s KPI and you have competitor’s media/marketing spends as independent variables.
📌 How is Ghost MMM possible?
In CPG, one often has access to:
– Third-party sales data (Nielsen, Kantar etc)
– Category-level trends
– Estimated media spends from trackers and panels
– Promotions, distribution, and pricing signals
Sometimes the brand themselves have a competition monitoring team (CMT) who collect all of the above data.
However the data collected are not as accurate as the brand’s own data.
But you have enough structure to work with.
📌 What purpose does Ghost MMM serve
Ghost MMM is not about getting perfect competitor numbers.
It is about understanding competitor behavior and response. More like a visualization of Game theory.
Ghost MMM in a way helps you answer some very important question like:
– What did my rival do when we increased Meta / Google spends that month?’
– When we raised our prices by $1.5, did the rival follow suit or they stayed the same?
and most importantly
– Do they even respond to our media / marketing initiatives or they follow a different trajectory.
Sometimes a rival responding to your moves is a good sign especially if they are a bigger player. It means that they take you seriously and are considering you as a valid threat !!
One can also use the ‘whitespace’ opportunities. That is a period where you find your rivals are not pushing enough, but you can make use of that market / timing gap.
📌 Good Data Enhances Ghost MMM
With accurate data one can even construct saturation curves of media spends of rivals.
I remember building a Ghost MMM nearly a decade ago. We figured that the rival was overspending on TV. The brand manager said, “Let’s make sure we never tell our rivals this. Never interrupt your opponent while he is middle of making a mistake 😅”
📌 Is Ghost MMM feasible only for CPG MMM?
No, not exactly. Just this year we started to experiment with Ghost MMM for D2C brands as well which are predominantly performance marketing heavy. Will cover about them in future posts.
📌 Ghost MMM is essentially a Ghost Lap for Marketing.
You don’t have the complete picture but yet you can see:
Where rivals are gaining/losing, how they are navigating the marketing ‘track’.
And often that is more than enough to give you a leading edge.
