Decide on Your Segmentation Objectives Early

For the smaller retailer who wants to conduct a consumer attitude survey, a major challenge is obtaining an adequate sample size. I advise obtaining at least 100 responses for each theme on which you intend to base significant business decisions. If you obtain a response rate of, let’s say, 70%, this means surveying about 145 consumers in order to end up with the 100 responses.
     You also want to choose your sampling strategy and word your items based on how you intend to segment the survey results. After putting all that effort into getting your 100 responses, you won’t want to suddenly face the fact that you surveyed the wrong people or asked the wrong sorts of questions. Decide on your segmentation objectives early in the survey process.
     Among the possibilities:

• Lifestyle profiles. What beliefs, feelings, and intentions distinguish those who shop at your store frequently from those who do not?
• Store-specific profiles. What distinguishes people who shop at your store instead of, or in addition to, shopping at other specific stores?
• Offering-specific profiles. What characterizes the shoppers for the different sorts of products and services your store offers?
• Demographic profiles. What can you learn from the survey that will help you customize your marketing and selling? What are the differences between your female and male target audiences? The Millennials from the Baby Boomers? The bilinguals from the monolinguals?

     Segmentation keeps you from being suffocated by the data. Instead of looking at consumer survey results for all respondents together, you might pull out the results for all respondents who report a household income greater than $100,000 and say they are grandmothers interested in giving your product to a girl under the age of 5.
     To be a smart user of segmentation, you also can benefit from knowing about a few statistical techniques:

• Factor analysis and item reliability analyses group items into scales. You don’t want to make important business decisions based on the answers to one item. Instead, analyze groups of items that look at the same theme from different angles.
• Discriminant analysis identifies which of those groups of items separate the different slices of your consumer pie.
• Cluster analysis places the survey respondents into groups based on the prior statistical analyses. You could be surprised by what you discover. What you’d suspected to be differences between your male and female shoppers might not be that significant after all.

For your profitability: Sell Well: What Really Moves Your Shoppers

Click below for more:
Segment Your Data for Profitable Conclusions
Help Consumers Build Useful Attitudes