Meerkats AI
← All posts
Measurement·13 Jun 2026·8 min read

Media Mix Modeling for ₹1–50 Cr brands: when it helps, when it's theatre

MMM is having a revival — privacy killed the pixel, so statistics are back. But below a certain scale, a mix model mostly redecorates your assumptions. An honest sizing guide.

TL;DR
  • MMM regresses outcomes against spend history to estimate each channel's contribution — no pixels, no cookies, privacy-proof by construction.
  • It needs what small brands don't have: 18–24+ months of history, real spend variation, and multiple meaningful channels. Feed it two channels of steadily-rising spend and it returns your inputs with confidence intervals.
  • The honest ladder: reconciliation → incrementality tests → MMM. Most Indian D2C brands under ₹20–30L/month multi-channel spend should stop at step two.

Media Mix Modeling is the oldest measurement idea in the room — P&G was doing it before the internet — and it's back in fashion because iOS broke the pixel and MMM never needed one. The revival is real, the open-source tooling (Meta's Robyn, Google's Meridian) is genuinely good, and the vendor pitches have gotten far ahead of what the math can deliver at D2C scale. Here's the sizing truth.

What MMM actually does

Feed it weekly history — spend per channel, revenue, plus controls (seasonality, promos, price changes) — and it fits a model estimating how much each channel contributed, including saturation curves (diminishing returns) and adstock (delayed effects). Output: contribution shares, marginal ROAS per channel, and budget-reallocation suggestions. No user-level tracking anywhere.

What it needs to not hallucinate

RequirementWhyWhere brands fall short
18–24+ months of weekly dataThe model learns from variation over timeBrand is 14 months old; tracking changed twice
Spend variationIf spend never moved independently, channel effects can't be separatedSpend only ever went up, on all channels together
3+ material channelsTwo channels = two knobs; the model mostly restates their correlation85% Meta, 15% Google
Stable definitionsRevenue must mean one thing across the whole windowGross in the old sheet, net in the new dashboard

The second row is the silent killer. MMM learns from differences — weeks where Meta rose while Google fell. If your history is "everything up and to the right together," the model cannot mathematically separate the channels, so it leans on its priors. And its priors were set by whoever configured it. That's how a mix model becomes a very expensive mirror.

The scale test, in one table

Monthly ad spendChannelsWhat measurement stack fits
< ₹5L1–2Reconciliation + weekly MER vs break-even. Anything fancier measures noise.
₹5–20L2–3Add brand-pause and one geo holdout per quarter. iROAS deflators do MMM's practical job.
₹20L–1Cr3+MMM starts to add signal — if the history and variation exist. Calibrate it with your geo tests.
> ₹1Cr4+MMM quarterly, incrementality to calibrate, reconciliation always-on. All three, each doing its own job.

If you do run one: the three sanity checks

  • Does it reconcile to the bank? Channel contributions must sum to something consistent with actual revenue. A model crediting channels with 130% of reality has already told you what it's worth.
  • Does it disagree with anything? A model that confirms every existing belief wasn't identified — it was decorated. Real models produce at least one uncomfortable number.
  • Was it calibrated against an experiment? Modern practice anchors MMM with incrementality results (Robyn and Meridian both support this). A model with no experimental anchor is a hypothesis, not a measurement.
The uncomfortable summary
MMM is real statistics with a real appetite. Feed it properly and it's the only privacy-proof macro view you can get. Feed it 12 months of two correlated channels and it will return your own assumptions, beautifully typeset, for ₹3–15L a year. Know which purchase you're making.

FAQ

Is MMM more accurate than attribution?
Different altitude. Attribution is order-level and always-on but blind to causation; MMM is causal-ish at the macro level but blurry per-campaign. Mature stacks run both plus experiments — each auditing the others.
Robyn or Meridian or a vendor?
The open-source tools are free but need someone who understands Bayesian regression to configure honestly. Vendors package the same math with better UX; the risk isn't the tool, it's uncalibrated priors nobody challenges.
Can MMM see retention or LTV?
Standard MMM models topline (usually new + returning blended) against spend. It won't tell you cohort quality — pair it with the cohort math from your own data spine.
We're at ₹8L/month and a vendor is pitching MMM. What do we ask?
Three questions: what spend variation exists in our history for the model to learn from? Which experiment will calibrate it? And what decision will it change that reconciliation + a geo test wouldn't — at a tenth of the price?

See your real numbers, not the platform's.

Connect Shopify, Meta and Google. Reconciled truth in 20 minutes.