• training@insanalytics.com
  • +91 90075 36541 | +91 90075 36560

Marketing Mix Modelling


A telecommunications services provider wants to identify media channels wherein its advertising campaigns yield the maximum utility in terms of revenue generation. Predictive analytics techniques are deployed to model the relationship between revenues and advertisement spending. Separate models are developed for the different media channels, viz, television, radio, newspapers and billboards. The lagged effect of advertising on revenues is accommodated, and variable transformation techniques are used to convert the inherent non-linear relationship between revenues and advertising into an approximately linear one. The resultant marketing mix modelling enables the company to identify the media channels that are most effective in generating revenues, together with the optimal levels of advertisement spending beyond which the beneficial effects of further advertising saturate with no further meaningful incremental revenue generation.

Business Backdrop

A telecommunications services provider wants to model its revenues as a function of the company’s spending on advertising across different media channels. The company advertises via several media channels, such as television, radio, newspapers, billboards, etc. The company wants to gauge the impact of advertising spend on the various media on its revenues. The company also wants to determine whether there are points of saturation for the various media channels beyond which further advertisement spend does not yield any meaningful incremental revenues.

Telecommunications service providers advertise their products & services over several media channels. The effect of advertising on the company’s revenues depends on a host of factors, such as the type of product/ service being advertised, the primary customer segment(s) that the product/ service targets, the reach of the media channel among the primary target customer segment(s), the frequency of advertising, the timeslots for which the ad spots have been bought (e.g., prime-time versus non-prime-time for television & radio advertisements), the placement of the advertisement and the circulation (e.g., cover-page, front-page, back-page, etc. for newspaper ads), the location of the advertisement (e.g., premium locations, residential areas, commercial areas, highways, etc. for billboard advertisements), and several other factors.

It is important for marketers to tie the benefits of advertising with the associated costs, so as to select the optimal advertising channel mix. Beneficial effects of repeated advertising on revenues rarely continue for prolonged periods of time, and the impact usually saturates after a certain duration of the advertising campaign beyond which no material effect is observed on the marketer’s top-line performance. Precise identification of the onset of such saturation enables marketers to effectively plan their advertisement spending across the various media channels.

Analytical Approach

Historical data on advertising spends across several media channels as well as revenues is analysed to explore the relationships between revenues and advertising spends. There might exist a time lag between the advertising spend and its effect on revenues. Various time-shifts are explored to zero-in on the precise length of the time lag. Advertising campaigns via different media channels are likely to influence revenues to varying degrees, and hence separate models are envisaged for each media channel. Four models are developed, one each for television, radio, newspapers and billboards.

For each media channel, the relationship between revenues and advertisement spending is found to be non-linear in nature. Accordingly, several variable transformation techniques are deployed to convert the relationship into an approximately linear one. This allows linear regression techniques to be utilized in arriving at predictive models that project the effect of advertisement spending via the various media channels on the company’s revenues.

The original non-linear relationship between revenues and advertisement spending is subjected to additional analyses to identify the optimal spending levels for each of the media channels. Beyond such optimal spending levels, saturation effects set in, and further advertisement spending do not result in material incremental revenues.

Solution Framework

  • Exploratory Analysis
  • Time Series Analysis
  • Linear Regression
  • Non-Linear Regression
  • Scenario Analysis

Business Benefits

With the results of the marketing mix modelling in hand, the telecommunications services provider is able to rank several media channels in the order of their effectiveness in generating revenues for the company. This allows the company to distribute its advertisement spending across the various media channels in a manner that maximizes revenue generation potential. In addition, the marketer is able to identify optimal spending levels for the different media channels, and can accordingly plan the duration of its advertising campaigns across the different media channels.

© 2017 All rights reserved
+ +