Blogs

How Train – Test Validation Quietly Changes the Philosophy of MMM Yesterday, I interviewed a candidate for an MMM role from a rival MMM company. Initially, I was genuinely impressed. They spoke about a fairly rigorous frequentist setup and mentioned several checks beyond just R squared to evaluate model quality. That was refreshing to hear because MMM quality cannot be judged by R squared alone. But then the candidate mentioned that they perform the exact same checks on the test dataset as well !! I was a bit taken aback. I further asked what they would do if the prediction accuracy went down in the test data (which almost always

Marketing Measurement Has Its Own Schrödinger’s Cat Problem According to Quantum Mechanics: Before measurement, a particle exists in multiple possible states (a superposition). But the moment you observe it, reality “collapses” into one observed outcome. Schrödinger illustrated this paradox with his famous cat experiment. A cat is placed inside a sealed box with a radioactive trigger mechanism. Until the box is opened and observed, quantum mechanics implies the cat exists in a superposition: Both alive and dead. Only observation resolves the uncertainty into one realized state. To explain this, Many Worlds Theory (MWT) proposes: Every possible outcome actually happens, but in separate branching universes !! So in one branch, the

Meta’s MCP + CLI and why accurate Marketing Measurement matters even more. Last week, Meta opened the door to Agentic Advertising. And I am sure Google, TikTok, Amazon, Reddit and even OpenAI won’t be far behind. With the launch of Meta Ads MCP + CLI, AI agents can now directly interact with Meta Ads infrastructure. All through natural language !! This is a pretty significant shift. Until now, AI mostly sat outside the execution layer. It generated ad copy, summarized reports or suggested ideas. Now the AI can actually touch the media buying system itself !! But there is an even bigger opportunity here. 📌 What if MMM becomes the

Move 37 Won’t Come From the Core of MMM but the Periphery In 2016, AlphaGo vs Lee Sedol match, AlphaGo played Move 37. A move so unconventional that even the best human players thought it was a mistake. But it wasn’t. It was a move which eventually proved decisive for the victory against Sedol. For the fist time, it was a glimpse of intelligence beyond human intuition. 📌 So why we haven’t seen Move 37 in Marketing Measurement yet? Well I would say we are looking at the wrong place firstly. The Move 37 will not happen at the core of MMM building. This nucleus is about: – Variable selection

Retargeting Campaigns Turn Your Estimand from ATE to ATT In various platforms, one has the option to retarget audiences. For example Meta’s ASC Retargeting campaigns or Google’s Pmax. Marketers / analysts however apply the same measurment philosophy for retargeting as they do for other non-retargeting campaigns. This in my opinion is not the right approach. In case of Retargeting, most marketers think they are measuring overall impact. They are not. They are measuring impact on a very specific, pre-selected group. Let me briefly explain the causal estimands of ATE and ATT. ATE (Average Treatment Effect) -> Effect of treatment on the entire population ATT (Average Treatment Effect on the Treated)

Causality is the cheat code for accurate predictions In a recent client project, we were pleasantly surprised to see that our model had predicted net new customer revenue with 95% accuracy for the future month of March 2026. Generally Marketing Mix Modeling (MMM) don’t achieve this much high accuracy in the future periods. But Media Mix Modeling (MeMM) can. If you are a D2C brand operating only with Meta ads or Google Ads or TikTok or just any of the two, it is easier to predict KPI accurately. 📌 How is this possible? Well firstly MMM is implicitly causal in nature. But if we control for all variables and OVB

Why you should not include Macroeconomic variables in Base of MMM? Yesterday, I wrote about “Baseline or Base in MMM is NOT “Unexplained Sales”. The post drew lot of interesting comments. One interesting comment was “Why don’t we include Macroeconomic variables as part of Base in MMM?” It is a good question. I believe they should not be included as part of the Base for the following reasons: Firstly, the media and marketing events are assumed to have continuous positive effect on the KPI (every day, week, month). But macro events have intermittent and unpredictable effect (sometimes negative too) on the KPI. In some weeks or months they may have

Baseline or Base in MMM is NOT “Unexplained Sales” Recently I came across a interesting post in which the OP said the following “Baseline is revenue + whatever revenue the model was simply incapable of attributing to ads, which can happen for many reasons, such as complex interactions and behaviors not included in the model.” Other misconceptions is “the baseline is a residual of everything the model’s media variables couldn’t explain”. Technically the above are not accurate. The base / intercept is the mean value of Y (your KPI) when you turn all your independent variables to zero. The main point is that the variables are all included and accounted for. So

Is your media actually driving growth or just piggybacking on it? A recent MMM project left both us and the client quite baffled. Despite millions in media spend, the model kept assigning very low contribution to media across multiple iterations. Naturally, our first instinct was: “Are we getting the model wrong?” But then we decided to go back to the EDA charts. The EDA charts pointed to a very good correlation between media spends vs sales in the latest years. At face value, it seemed that the media is working. But we quickly reminded ourselves what we teach in our causality course – correlation is a low bar. Remember for

Why only few organizations successfully convert MMM insights to Decisions Many reports state – only 25-30% of the organizations are able to convert MMM insights into any meaningful decision. But Why? We have been in MMM space for more than a decade. From our experience, the breakdown happens in three places: 📌 Organizational Memory Fade MMM is not a one-time exercise. It is a cumulative learning system. But in reality: Insights sit in decks and that too only a select few access to them. Teams change, vendors change and Context gets lost. Every new MMM cycle starts with – “What did we conclude last time?” “What were the actions taken?”.