Marketing-Mix Modeling

Marketing accountability means that marketers must more precisely estimate the effects of different marketing investments.

Marketing-mix models analyze data from a variety of sources, such as retailer scanner data, company shipment data, pricing, media, and promotion spending data, to understand the effects of specific marketing activities.

To deepen understanding, marketers can conduct multivariate analyses, such as regression analysis, to sort through how each marketing element influences marketing outcomes such as brand sales or market share.

Especially popular with packaged-goods marketers such as Procter & Gamble and Colgate, the findings from marketing-mix modeling help allocate or reallocate expenditures.

Analyses explore which part of ad budgets are wasted, what optimal spending levels are, and what minimum investment levels should be.

Although marketing-mix modeling helps to isolate effects, it is less effective at assessing how different marketing elements work in combination. Wharton’s Dave Reibstein also notes three other shortcomings:

1. it focuses on incremental growth instead of baseline sales or long-term effects;
2. it has limited integration of metrics related to customer satisfaction, awareness, and brand equity; and
3. it generally fails to incorporate metrics related to competitors, the trade, or the sales force.