After a business & IT framing, data are collected & analyzed, feeding the first set of models, but also providing valuable learnings to later streamline and automate their collection as much as possible.
Similarly, the initial modelling provides both
Reliable measurements of the marketing performance & associated learnings (accounting for complex but important effects such as long-term impact, halo, cannibalization, saturation, evolution in dynamics, anticipation & hangover of promotion...)
Learnings for MMM in other business areas & industrialization (zones of confidence, useful data, feature engineering, extrapolation rules, deep-dive feasibility...)
Implementing MMM on new markets, leveraging the learnings & technical assets from the initial modelling, providing similar measurement & learnings to optimize a larger proportion of the marketing investments, while assessing the possible convergence & economies of scale across the MMM community
Exploring & developing AI-powered modules lowering MMM-related effort and cost, across its lifecycle (initial modelling, updates, retraining) and all steps of its value chain (from data collection to modelling to output analysis & consumption)
Building an MMM tool tailored to our client's business & IT specifics, owned and operated by our client, allowing for autonomous operations (automated and/or streamlined data collection, high-frequency models update, 24-7 results consumption, simulation & optimization features)
Deploying the tool across geographies & business units, embedded in a transformative program with a significant change management component, maximizing business adoption & marketing optimizations
