There is an outstanding post by Nico Peruzzi on reverse segmentation over at Research Access from last week.
I have to admit, I had never heard the term "reverse segmentation" and I'm a fool for segmentation. Can it be that I've missed some HUGE development in market research? Nope. There is a good chance you've already been doing and struggling "reverse segmentation" and just didn't know that this was what we were calling it these days.
A Definition of the Reverse Segmentation Process
Let's start with the basics -- segmentation is grouping. Marketing managers and small business owners alike intuitively know and understand that smaller, more targeted groups are more desirable because we can develop a message that appeals to them and makes it more likely for them to choose us.
There have typically been two basic kinds of segments; demographic (more concrete and observable segments such as gender, income, age, etc) and psychographic (emotional, aspirational, attitudinal attributes such as independent, organized, practical or cautious).
There is a common criticism of segmentation in that we typically collect either demographic or psychographic info, but not really both. And this is where the "reverse segmentation" comes in.
What's The Trigger for Reverse Segmentation - How Do You Know You Should Do it?
Say you already have an existing customer segment of college-educated women from Ohio between the ages of 25 and 35 who purchase your product. And now you wonder how many of them are procrastinators or how many of them go skydiving? Because if you had more than 50% that were procrastinators, you could develop an offering just for them. This sounds like a job for reverse segmentation.
Richard Hamer from Deft Research puts in another way in his article titled "What the Heck is Reverse Segmentation Research?" He explained that all you do is take the demographic data you already have within an existing customer database and do more research to assign new (perhaps psychographic) data.
Use QuestionPro's and Survey Analytics' Custom Variables to Do Reverse Segmentation
After I read that- the first thing I thought of was the hundreds of survey's I've done using the custom variables tools in QuestionPro. THIS is where I put the data I already knew about my customers. For example, I wanted to know if customers that ran frequent and multiple transactions had different satisfaction levels than those that were less expert and ran few transactions.
We took our existing database and created 4 segments based on the number of transactions that these customer ran. Then we ran the survey and attached their satisfaction results to those segments. After that we ran some analysis to see if there was a difference.
As it turns out reverse segmentation might be a new term -- but we've had the tools to do it all along.
What's been your experience with "reverse segmentation"?