What is it?
Digital tracking methods have been profoundly impacted by GDPR, iOS14, and a general heightening of privacy regulations. This has left many marketers searching for alternative measurement methods that aren’t reliant on pixels or tracking users across the internet.
For many, Media Mix Modeling (MMM) has filled this gap. MMM is a statistical modeling technique that marketers use to determine which channels in their marketing mix are driving sales or acquisitions. MMM helps brands identify the incremental return of investment into every media channel and reallocate budgets to the highest-performing areas of their media mix.
MMM does this by using machine learning and statistical algorithms to find patterns in historical observational data. At a high level, models look for variation in marketing activity and attempt to line that up with variation in a business KPI.
MMM does this type of analysis over and over again for all marketing channels across all of the days they were active. It can then provide estimates of incrementality across every channel in a brand’s media mix.
Why is it important?
MMM can be traced back to the 1960’s when large CPG companies pioneered the use of models to evaluate the impact of their ad campaigns. This methodology provided empirical evidence and insights into their most profitable channels without needing to track the clicks or views of customers.
MMM has come a long way over the last 50+ years and modern MMM methods now address many of the limitations of these legacy approaches. Modern platforms focus on speed and verifiability, allowing marketing teams to trust their media mix models and use them for in-flight media optimizations.
MMM is designed to work with aggregated data, and because it does not rely on tracking pixels or cookies, it’s unaffected by an evolving regulatory landscape. This has made MMM an increasingly attractive measurement alternative for marketers who relied heavily on digital tracking methods.
What should you do about it?
Is MMM right for your brand? Typically, it’s best suited for teams making meaningful investments into a complex media mix that includes online and offline channels. For these brands, MMM can become an important layer in a larger attribution stack.
True incrementality is hard to find using any single measurement tool in isolation. Instead, marketers should combine their attribution methods together (including MMM) in a process called “triangulation”.
How does triangulation work in practice? Here’s an example for a brand using digital tracking (MTA), marketing mix modeling (MMM), and testing/conversion lift studies (CLS):
- Multi-touch attribution is used for daily, tactic-level optimizations like turning off under-performing ads or shifting budgets between prospecting campaigns.
- MMM is used to forecast different budget allocations between channels and determine how much can be spent into a channel before diminishing returns are observed.
- Conversion Lift Studies are used to calibrate the accuracy of MMM and help get closer to true incrementality.
TL;DR
Marketing Mix Modeling is a statistical modeling technique used to determine which channels in a marketing mix deserve credit for sales or acquisitions. It’s not reliant on digital tracking methods, but rather, uses aggregated data to measure marketing performance and the incrementality of investments into different media channels.
Brands investing in complex media mixes with hard-to-measure channels are often good candidates for MMM. For these brands, MMM can become a powerful addition to their media measurement stack and a point of triangulation for determining the true incrementality of their marketing investments.
