Steps to Achieve Marketing Optimization

This is a guest column by Niren Sirohi, Vice President, Predictive Analytics at iKnowtion, as part of our Analytics Advantage series.

Niren-Big

Marketers are constantly faced with complex decisions about how to spend their limited marketing budgets to best meet their business goals and achieve maximum results. Interestingly, one of the more effective ways to do this involves the use of optimization techniques, which typically fall in the domain of Operations Researcher instead of Marketing.

Marketing optimization requires multiple steps and the use of advanced analytics. The example below will help to illustrate the overall process.

Consider the following typical marketing problem:

A marketing exec has a fixed budget of $20M, five products to offer; each through one of three potential channels, and four different incentives (or choose to use no incentive at all). With 3M customers, this marketer needs to determine at the customer level, which product to offer, through which channel, and which incentive (if any) to offer, so as to maximize the expected incremental profitability for the campaign.

Steps to Achieving Optimization and Maximum Results
The first step in solving a problem like the one above is to build predictive analytic models that will quantify, at the individual level, for each product, each channel, and each incentive offer, the expected incremental profitability. In order to do this, lift modeling techniques may be adopted.

The second step involves optimization. However, this is a hard problem to solve. In the example above, each customer could have 75 specific combinations of product, channel, and incentive offer to choose from – 5 products X 3 channels X 5 incentive offers. With 3M customers, that is 225M combinations to choose from. This translates into an integer programming problem of size 2 to the power 225M which is almost impossible to solve given the size. Fortunately, there is an approach to convert this mathematical problem into a more manageable linear programming problem that yields an approximately optimal solution.

Marketing optimization techniques can be easily applied in other areas of marketing, as well, e.g. in marketing mix, channel mix, pricing etc. These techniques can be implemented in a semi-automated fashion to enable repeated use with high degrees of efficiency, like in campaign settings where a campaign runs frequently.

##  ##  ##

About the Author:

Niren Sirohi is Vice President, Predictive Analytics at iKnowtion and responsible for leading the company’s predictive analytics practice. For more than 20 years, Dr. Sirohi has been developing and implementing strategic analytic solutions for global brands across a variety of industries including financial services, retail, consumer goods, hospitality, and telecommunications.

To learn more about how you can use marketing optimization, download our white paper titled Using Marketing Optimization to Drive Incremental Sales and ROI.

Your Comment