What is Marketing Mix Modeling?

Marketing Mix Modeling (MMM) was introduced in the 1960s to match spikes and dips in sales to actions taken in marketing (and external factors). It can tell you the incremental impact of each part of your marketing strategy, so you can make data-driven decisions on where to allocate budget. Previously used mostly by Fortune 500 advertisers spending millions on SuperBowl ads, MMM is now the only future proof way to measure marketing ROI thanks to Apple, GDPR/CCPA and adblockers limiting marketers' ability to track users.

First plot your sales over time.

What makes conversions go up and down?

Find variables that can explain the data.

Match spikes and dips to actions and events.

Build a model that can predict sales.

Predict the past, and you can tell the future.

The 4 'P's of the Marketing Mix

MMM accounts for all factors of the 4 'P's and their impact on revenue.

Product.

features, brand, packaging, services.

Price.

discounts, credit, shipping, margin.

Place.

location, logistics, stock, distribution.

Promotion.

advertising, publicity, partners, sales.

1. Collect

Get your data in one place.

Pull data from ad platforms, analytics and your internal company databases.

2. Engineer

Clean and transform the data.

Get it into the right format, remove outliers & errors, and create new variables.

3. Model

Predict past performance.

Find patterns in the data and variables with explanatory power for what drives sales.

4. Optimize

Make strategic decisions.

Run simulations based on your model to optimize marketing spend for increased ROI.

78%

of marketers surveyed by the IAB were using MMM as of Nov. 2019

* 80% of respondents were Director, VP or C-level

63%

have been using MMM for more than 5 years

* from 116 advertisers, agencies, media/publishers, and marketing companies

“[MMM is] A Better Way to Measure Your Ad Campaign”

Wes Nichols

Harvard Business Review

“Marketing mix modeling (MMM) is a time-tested method for measuring the impact of your marketing.”

Kevin Hartman
Think with Google

“Media mix models are the future of mobile advertising”

Eric Seufert
Mobile Dev Memo

Learn Marketing Mix Modeling in 7-18 Weeks

LEARN MORE

Multi-Touch Attribution

Incrementality & A/B Testing

Marketing Mix Modeling

Benefits & Limitations

Real-Time Optimization

User Level Attribution

Limited Applications

Tactical Performance Insights

Large Effects Only

Strategic Budget Decisions

Unreliable

Accounts for Product Changes

When Planned Ahead

Incorporates External Factors

Can Exclude Effects

Works Across All Channels

Questions MMM Can Answer

Why doesn’t Facebook match Google Analytics?

There are a number of reasons why you can't trust Facebook's numbers. For example they claim 100% of all sales that occured within 24 hours of seeing an ad (even if the user scrolled past without noticing). Marketing Mix Modeling can give you the real incremental ROI of Facebook, measuring how many sales would have happened anyway had you not spent a penny on ads.

How do I measure the impact of TV?

There's no way to 'click' on a TV ad – so how do you measure performance? Marketing Mix Modeling has been tracking the ROI of TV campaigns since the beginning, and can even account for long-lasting effects on brand awareness. This technique also works on other non-direct channels, like billboards, print, radio and even more modern channels like podcasts and influencer marketing.

How do I measure performance when users don't want to be tracked anymore?

The ongoing privacy backlash against surveillance capitalism has resulted in a rapidly changing and unpredictable landscape. Whatever happens with public legislation (GDPR / CCPA), user behavior (ad blocking) or platform policy (Apple, Brave), we know we're not going back to the good old days of unlimited user tracking. Marketing Mix Modeling's great strength is that it doesn't need user level data to work – so it's the only future-proof way to measure marketing ROI.

If I doubled my marketing budget, what would be my ROI?

You might have noticed when you increase your budget your cost per acquisition (CPA) suffers. This effect is called Diminishing Marginal Returns and it's one of the key things Marketing Mix Modeling can help you forecast.

What was the impact of COVID-19 on my business?

As much as we'd love to be in full control of hitting our sales numbers, the recent pandemic has made it clear that's not the case. What's not clear is exactly how much impact macro trends have on our business – we can't ignore it, but we can't blame it for everything. Marketing Mix Modeling can help us predict what our sales would have been otherwise, so we can set realistic goals going forward.