Marketing measurement in eCommerce and Omnichannel retail is in a really interesting turning point!
Over the years, we've seen an increasing frustration to the Legacy Marketing Mix Modeling (MMM) approach. Maintenance burden is high, insights lack granularity, and quality has been questionable.
At the same time, attribution tools have provided granular insights for tactical optimization, but they have strong, and nowadays broadly accepted, biases for favoring different media channels. Last-click is underestimating the ROI of social, and overestimating ROI of branded search. Ad platform -reported ROAS is inflated. And so on.
With this context, I'm so energized to be back from the summer vacations and continue bringing our vision for the Next Generation MMM solution to life at Sellforte 🙌 Next Generation MMM solutions have three advantages:
🤝 High quality results: Optimize and forecast with confidence
We combine the power of Bayesian Marketing Mix Modeling with model calibration. Model calibration is a concept for providing the model all available information on media performance, including info contained in incrementality tests (conversion lift tests, geo tests, hold-out tests) and attribution data (Google Anayltics 4, ad platform, MTA). This is a topic we're really passionate about and I will share more about our approach to model calibration in the coming weeks.
⚡ Fast start & Easy maintenance
Taking MMM into use and maintaining it has been way too laboursome. Using data connectors and automating many steps in the data-processing and modeling workflow specifically for eCommerce and retail businesses is delivering a huge step change. For a eComs with mostly digital advertising in use, the onboarding journey should be a matter of days - not months.
🔎 Ability to dive deep
Next Generation MMMs can provide campaign-level ROI measurement. This is made possible by the combination of MMM, attribution data and incrementality tests. Next Generation MMM solutions also encourage the use of media metrics to analyse drivers of media performance. For example, a +15% improvement in PMax ROI might be explained by an increase in Average Order Value.