Marketing mix modeling or MTA: Which analysis is best for your campaign?

MMM and multi-touch attribution are both effective ways to analyze an advertising campaign. Here's how to choose between the two.

Wondering whether you should use marketing mix modeling (MMM) or multi-touch attribution (MTA) to evaluate your ad campaigns?

We’ve got answers (spoiler alert: it’s both!) 

MMM and MTA are both excellent tools for analyzing paid media campaigns, but they’re not interchangeable; they offer different perspectives. 

Here’s a quick rundown of each one and what it adds to your campaign reporting: 

What is marketing mix modeling (MMM)?

Marketing mix modeling (MMM) is a process of measuring the impact of multiple different marketing channels on sales, revenue, or market share simultaneously and using that data to optimize the allocation of ad dollars to each channel.

For media agencies like Exverus, we’re usually measuring the effectiveness of multiple digital media channels like paid search, social media advertising, retail media networks, and programmatic CTV alongside experiential and out-of-home campaigns. It’s a lot to digest! That’s where technology comes in. 

How does marketing mix modeling work?

There are more digital media channels than ever, so marketers need help to plan their ad spend and analyze campaign data from across many disparate sources. MMM tech providers like Keen Decision Systems use statistical regression models to provide marketers with standardization in a fragmented media world. 

MMM software combines historical sales data with inputs like 

• business goals

• economic conditions

• seasonality

• and changes over time 

to deliver a comprehensive picture of an ad campaign. These insights help marketers to better plan and optimize their ad budget allocation across paid media channels going forward. 

What is an example of MMM? 

Here’s an example output from a MMM analysis that shows which media channels drove the most revenue relative to investment for a previous client, Stella & Chewy’s: 

What is the difference between MMM and MTA?

Excellent question! MMM is more of a big picture guy, while MTA is more detail-oriented. 

MMM analyzes how various marketing elements collectively impact sales or other KPIs, while MTA identifies the specific contribution of each touchpoint in the customer journey.

Use MMM when you need a high-level view of your marketing mix and its impact on sales. It's particularly helpful for budget allocation and understanding how paid media interacts with other marketing efforts.

Additionally, MMM relies on historical, aggregated data (like campaign budgets or overall sales figures) while MTA needs more granular, real-time data on individual user interactions across channels.

Use MTA when you want to optimize individual paid media campaigns. It's ideal for identifying the most effective touchpoints and attributing conversions accurately.

The best approach depends on your specific needs and data availability.

“We need to balance these newer technologies with vetted data sources and measurement tools. Some agencies can be either too quick to jump on the latest un-tested fad or do not update their toolkits consistently.” - Exverus analytics director Charles Lai in Digiday

What is an example of MTA? 

Below is an example dashboard from an MTA project that shows the top combination of touchpoints leading to conversions and the average number of touchpoints in the path to conversion: 

MMM tips for brand marketers

• Carefully define marketing tactics at the level of granularity where actionable insights could be obtained, but not so granular that results become unstable.

• A good market model is guided not only by statistics but also domain expertise, or an understanding of the brand’s unique industry and consumer journey and habits, along with an understanding of media tactics. 

• The optimal model is often not the one with the strongest fit and accuracy statistics, but the one that has reasonably good statistics and produces intuitive and reliable outputs that converge with other data sources.

MTA tips for brand marketers

• Choose a MTA platform that can track most or all of the marketing channels that you use in your campaigns. Some platforms can only track click-throughs, while others can attribute based on impressions. Also look for data partnerships that could bypass walled gardens.

• As MTA relies on deterministic tracking, the coverage and maintenance of the MTA provider’s identity graph and their identity resolution technology are important considerations.

By understanding the strengths and limitations of each approach to campaign analysis, marketers and media planners can make better decisions together that optimize their paid media campaigns and drive stronger business outcomes. 

This article originally appeared in our weekly Paid Media Insights newsletter. For the most up-to-date news and tips to improve your media strategy, subscribe for free here.