Marketing Mix Modeling & Big Data
In the wake of the Big Data revolution, data mining and machine learning methods have taken centre stage in unstructured data analysis. However, when it comes to marketing ROI, simulation and causal based inference, marketing mix modeling continues to play a key role.
This article looks at various econometric techniques used by mix modelers to help deal with modern big datasets.
Advanced Methods in Marketing Econometrics
This article outlines three broad innovations in marketing mix modeling (MMM) techniques ranging from competitive structure and the measurement of long-term effects through to incorporation of digital and social media.
All such innovations lead to increased flexibility and realism with more accurate client deliverables, helping to maintain the relevance of MMM in the modern digital economy.
Dynamic MMM and Digital Attribution
P.M Cain (2014), Admap.
In order to maintain its relevance in the modern digital economy, the traditional marketing mix model needs to be re-structured to incorporate digital media. This article presents a general dynamic approach that explicitly models the offline-online consumer purchase journey.
Brand management and the marketing mix model
P.M Cain (2014), Journal of Marketing Analytics.
Successful brand management requires a simultaneous holistic view of all players in the market. Conventional marketing mix analysis, however, focuses solely on individual brands in isolation. This paper argues that a dynamic version of the discrete choice attraction model is a preferable framework.
Marketing mix modelling and return on investment
P.M Cain (2010), Integrated Brand Marketing and Measuring Returns, Palgrave Macmillan.
This paper provides a detailed outline of the marketing mix modelling process, dealing with state-space econometric techniques applied to marketing analytics and models for evaluating the long-term effects of marketing investments.
Limitations of Conventional Marketing Mix Modelling
P.M Cain (2008), Admap.
This article argues that the conventional marketing mix model ignores the long-term effects of marketing investments by construction and needs to be re-structured to accommodate both short-run and long-run variation in the data.
Modelling and forecasting brand share: a dynamic demand system approach
P.M Cain (2005), International Journal of Research in Marketing.
This paper proposes a dynamic state-space AIDS model of brand share, providing a general framework to examine the time series properties of the data. The techniques provide an accurate assessment of the short and long-run effects of marketing mechanics.