Average Is Not Good Enough

May 13, 2008 1:14 AM  By

Do you profile your customers and project values like age, income, household size and gender onto the prospect universe? You might be better off finding small segments that have slightly different characteristics.

For example, I once saw a telephone pole near a college campus that had held hundreds of concert and lost-pet posters. There were few staples near the ground. But they grew in density as you went up the poll, peaking at about 5 feet 10 inches. They thinned after 6 feet and ended entirely at 7 feet.

The staples approximated a normal distribution of the height of the average person’s eyes—and that of the local population. A clothing retailer in the area could use this information to set the sizing for its merchandise.

avoiding very small and very tall sizes as being limited proportions of the size range.

I drew the same conclusion after a dinner party I attended. My hosts belonged to a “beanstalk” club that requires members to be well above average height. That night, the average height in the room was 6 feet six inches. (I was below average.) Yet a sample of 1,000 men that included all the male club members I met would still average 5 feet 10 inches in height.

The tall segment would be unnoticed. A marketer relying on the average would miss an opportunity to find a great segment for hard-to-find sizes of clothing.

Ask your customers about their club memberships or the magazines they read. A single question might lead you to discover a marketing niche that would otherwise be lost in the height of the statistics.

Bill Singleton writes “Show Me The Data” each month for Lists and Data Strategies. He is a Manager of Analytics and Consulting Services at The Allant Group in Naperville, IL. He can be reached at: bsingleton@allantgroup.com and 630-579-3448.