Like so many buzz words in business & marketing, “customer segmentation” is one of those terms that is interpreted by folks to mean many different types of things. If the word “segmentation” were blurted out in a room of 20 business people, chances are it would conger up 20 different images.
So what is customer segmentation, and how can it be used to propel one’s business?
Segmentation defined
Customer segmentation is a method for grouping customers based upon similarities they share with respect to any dimensions you deem relevant to your business – whether it be customer needs, channel preferences, interest in certain product features, customer profitability, etc.
The key is for you, the marketer, to first decide on what basis you wish to segment your customers (or prospects for that matter). And, the only way to answer this question is to first determine what your objective is for the segmentation, and thus what you want the segmentation to “do for you”.
Common segmentation objectives
Common objectives for segmentation include but are not limited to: the development of new products, the creation of differentiated marketing communications, the development of differential customer servicing & retention efforts, channel strategy, and the maximization of profit/ROI for existing products.
Once you have decided what your objective is for the segmentation, you can answer the question, “what do I want the segmentation to do for me?”
A brief example: segmenting for customer winbacks
Let’s say you worked for a subscription-based magazine such as Time Out New York (TONY). Your boss has asked you to optimize TONY’s retention strategy utilizing the current save tactic of sending people who have recently canceled their subscriptions (aka “attritors”) 1 of 3 “win-back” mailers. This existing save tactic has been employed by TONY for the past 2 years, and the method for determining which attritor receives which mailer has been based largely on “intuition” (aka random selection).
Your first step in undertaking this project would be to clearly state your objective. Your objective, as per your boss, is to optimize TONY’s retention strategy for recent attritors. This is shorthand for saying, “I want you to maximize your return on your retention-dollars invested”.
Without getting into the nitty gritty of the approach, what you essentially want to do is determine the relative ROIs for each of the 3 mailers at the individual attritor level. For each mailer, you then want to identify those attritors with high ROIs (i.e., those attritors who re-instated their TONY subscriptions after receiving the mailer and provided you with future profits that well-exceeded the cost of the mailer).
Next, for each win-back mailer you want to identify those attributes which the high-ROI attritors have in common, essentially creating a profile for “high-ROI attritors” for each mailer.
The final step is to operationalize the three profiles you’ve created so you can use them to determine which of the 3 mailers, if any, to send to future attritors. This essentially entails implementing a process in which new attritors are matched up against the 3 profiles to determine which, if any, best describe them.
A more sophisticated approach would be to build predictive models that would calculate the expected ROI for each mailer for each attritor, and then send out the mailer with the highest expected ROI to the attritor. And, for those attritors in which all 3 mailers have negative expected ROIs you might choose not to send any win-back mailers.
Closing thoughts
In closing, segmentation can be tricky and complex, and no doubt requires a great deal of expertise & experience. Putting in place flawed segmentation strategies can be far more detrimental to a business than not having them at all. However, when designed the right way, segmentation strategies can provide tremendous returns relative to one-size-fits-all approaches.