Marketing Mix Modeling (MMM) – Optimize your Marketing ROI

MMM Model Update Presentation – The most tense period for everyone

MMM Model Update Presentation – The most tense period for everyone 😅 I have been part of 100’s of MMM model presentation. However every model update presentations still makes me a little bit anxious. And funnily the client side too feels the same anxiety. Every MMM model update carries with it a lot of suspense and questions. Such as: ● Are the key drivers of KPI still the same or are there any surprises? ● Did our brand equity increase or decrease? ● Did the increased spends work ? ● What happened to the competition effect? Did it wane or got stronger? ● What is the level of saturation of various media channels? How has it

Preventing MMM’s Death (Again)

Many had written off MMM (Marketing Mix Models) not so long ago. But like a phoenix, MMM keeps rising from the ashes.🔥 Thanks to: GDPR, Apple 14 privacy, Google’s cookie deprecation and ever growing pressure on marketing departments to justify their budget allocation, MMM is back to life again. 🚀 But how can we prevent MMM’s death (again)? In my opinion, the following caveats are important to keep in mind. ⚙ Don’t treat MMM like AutoML I have heard prospective clients wanting to fully automate the MMM pipeline. Automation is possible in parts but not fully. The variables that are selected automatically based on superficial correlation (or similar metric) are

How to get the maximum out of open source MMM libraries

How to get the maximum out of open source MMM libraries. (Hint: talk to MMM experts) Of late we are getting lot of calls from prospective clients for MMM adoption. Most of them have tried or are trying open-source MMM libraries. But here are some of the pestering questions clients have: · How do we choose the *best* model out of the many models? · How do we know if the model is correct? · Is the saturation curve correct? What happens if we keep deploying spends beyond the saturation point? · What should be the ideal budget allocation? · What are the right calibrations for the MMM model? Most

Why I am bullish on Marketing Mix Modeling MMM

In late 2018, We explored the option of VC funding. Some of the VCs asked us why we focused on Marketing Mix Modeling solutions. We told them at that time that MMM will be a sought after solution not only in FMCG/CPG space but in other domains as well. Call it a hunch or educated guess, we believed in our gut feel and bootstrapped our startup in early 2019. Fast forward 2023 🚀 We see many new companies mushrooming with only MMM focus. 📈 We see nearly 100% increase in demo requests and pitch meeting requests as compared to last quarter. 🎯 We see big firms like Meta, Amazon and

Mathematically Optimal Solution ≠ Optimal Marketing Solution

Mathematically Optimal Solution ≠ Optimal Marketing Solution One of the important use cases of Marketing Mix Modeling (MMM) is the Marketing Budget Optimization. The results of MMM are directly leveraged to provide various ‘What-if’ budget allocation scenarios. Such as: ▪️ Under the same budget, how could we have allocated the spends across media differently to yield a lift in KPI? ▪️ What should be the allocation across media if there is a 30% increase in budget? ▪️ What if we decide to cap media A’s allocation to 10% of budget, what would be the net effect on KPI? Most organizations are keen to know various ‘what-if’ budget optimization scenarios for

Why should you update your MMM model? and How often should you update them?

As MMM is gaining acceptance in newer domains, most prospective clients ask the following questions. ▪ Why should we update MMM models? ▪ How often should we update them? 📌 Why should you update MMM model? To explain this, I normally use the analogy of movie making. A movie is a series of snapshots put together. A MMM model or for that matter any model is a snapshot of reality. The problem is, a snapshot can tell you only so much. More specifically it can only tell you about the things that took place at the time of taking the snapshot. Market reality is very dynamic. Things evolve constantly.

Why MMM Needs Innovation

MMM is not new. CPG/FMCG domain has been adopting MMM for nearly 30 years. The statistical methodology behind it, is proven and time tested. Why then should one innovate ? Through this post, I would like explain why. 🎯 Reason 1: MMM is being adopted by new domains. There has been increasing adoption of MMM by new domains like digital first brands. The problems faced by established brands vs newer brands are different and so are the data collection processes. Most modern brands collect data at a much faster cadence and advertise at a faster cadence. Some of the old MMM methods do not work in capturing the right ROI at

Why Decomp RSSD is a Business Fit Metric

Decomp RSSD – The business fit calibration During the Meta panel discussion (link in Resources section), I had mentioned about Robyn’s Decomp RSSD metric. In our earlier posts too, both my co-founder Venkat and I extensively covered the topic of calibration and validation extensively. Calibration of MMM model can be achieved through goodness of fit. It is not necessary that statistically well calibrated models make business sense. Therefore one also needs to calibrate the model in terms of making business sense. During my agency days, once the MMM model was built, we always used to look at the contribution of the marketing channel vis a vis the money spent on

You can do MMM with small Marketing Spends

MMM is for enterprises of all sizes Of late I came across a lot of posts proclaiming that MMM is not ideal for SMBs. I want to dispel some myths on this. 📌Small Marketing Spends Small marketing spends is not a show stopper. Not statistically and not marketing wise either. There are various statistical techniques that can accurately compute the effect size no matter the small marketing spends. 📌The Learning Curve The other major reason cited by people to dissuade SMBs from adopting MMM is the steep learning curve. Granted there is some learning curve in MMM but it is not steep. Many ask us, “why we keep sharing posts

How to use AIC to select the best Marketing Mix Model (MMM).

How to use AIC to select the best Marketing Mix Model (MMM). Firstly, let’s look at what is AIC and the most common misunderstanding associated with it. The Akaike information criterion (AIC) is given by: AIC = 2k -2ln(L) where k is the number of parameters L is the likelihood The underlying principle behind usage of AIC is the ‘Information Theory’. In the AIC equation we have the likelihood. We try to maximize the likelihood. It turns out that, maximizing the likelihood is equivalent of minimizing the KL Divergence. But what is KL Divergence? From an information theory point of view, KL divergence tells us how much information we lost

Blogs