1. The MMM: The Gold Standard

Marketing Mix Modelling (MMM) is widely regarded as the most rigorous way to measure marketing effectiveness.

See how top 3 digital players pov on MMM: https://tommyacademy.io/2026/01/05/attribution-measure-update/

At its best, MMM uses econometric regression to decompose total sales into measurable contributions.

Things like: TV, digital, pricing, promotions, competitor activity, organic social, & seasonality.

The output tells marketers/founders exactly how every dollar spent drives business outcomes.

The narrative is compelling: move beyond last-click attribution, invest in long-term brand understanding, make budget decisions with real statistical confidence.

However, things may not be as easy as they seem.


2. MMM: The Untold

Here is what the industry rarely says out loud or highlight the untold.

2.1 MMM has a brutal data requirement

MMM requires a minimum of 2 years of clean, weekly data for sales, spend, pricing, promotions, & competitive intelligence.

On top of that, all data must be harmonized in one place.

Meta’s Robyn Analyst Guide recommends a minimum of two years of clean, weekly data.

You need at least 10x more data points than channels. 

And all of that data comes from sales, spend, pricing, promotions, competitive activity.

All of that needs to be integrated, cleaned, & aligned before a single line of model code gets written.

2.2 Data is fragmented or simply doesn’t exist in one place

Reality is that most brands cannot produce this.

  • OS data is fragmented.
  • eCommerce data is locked inside Shopee, Lazada, and TikTok Shop.
  • Pricing history is not tracked.
  • Nobody owns the integration.

Is it the organizational problem, an operational problem, or just a fundamental problem?

Tommy Nguyen (H.Thinh)

Reality is that:

  • Agencies control media data.
  • Clients own sales data.
  • Research firms hold retail data.
  • Data is everywhere & nowhere to store & organize.

These parties rarely sit in the same room with aligned incentives.

And even when data exists, the expertise to build MMM correctly is scarce and expensive.

It does require data cleaned up, organized, & well-structured at best.

2.3 There’s the expertise problem

Only 26% of in-house marketers run MMM themselves, according to Funnel & Ravn Research (April 2024).

Everyone else needs external help which costs time & money most mid-market brands don’t have.

3. MMM: The Reality

If you’re running an SME: “My sales data is split between my POS system, Amazon, Tiktokshop, Shopee, and an excel spreadsheet my sales team updates whenever they feel like it. Maybe, MMM sounds like it’s not for me. It is for the Fortune 500 “

If you’re at a startup: your world moves fast. A quarterly MMM report is ancient history by the time it lands. A history you know, for you once & only, why to build a solid model for looking backward?

The matter is to know if last week’s TikTok push actually moved the needle, not a model that takes three months to build.

What’s matter is being effective, fast & forward.

Fair point too!!!

4. POV

So yes. MMM is a gold standard. And for most marketers today, it’s also a distant dream.

But here’s the honest reframe: 

MMM isn’t a tool you use tomorrow. It’s a DESTINATION you prepare for today.

Tommy Nguyen (H.Thinh)

The industry owes you an answer. It is working on it.

3 things you can start right now & no data scientist needed:

  • Track weekly sales, spend, and pricing in one spreadsheet. Every week.
  • Document every promotion and campaign change, even informally.
  • Align with your agency on data format and ownership before the next campaign.

Do this consistently for 12–18 months, and MMM stops being a dream.

Remember that: The gap isn’t the model. It’s the data discipline that was never started.

In case, you need any advice, please email us: tommyacademy.vn@gmail.com. The team would handle & give you proper advice per your case.

TOMMY Nguyen (H.Thinh)

P/s: Opinions are my own. Please take consideration for your actions.

This is as for informational & educational purpose, No liability for actions taken.

Nothing in this article constitutes legal, compliance, or regulatory advice.

© 2026 TommyAcademy. All rights reserved.

Leave a Reply

Your email address will not be published. Required fields are marked *