I hate market segmentation. At best, it’s boring. At worst, it’s misleading. I don’t trust it. I especially hate using market segments for small studies. How much do you really have in common with someone just because they are your same age and race? And why should I care? I’m not talking about cultural identity, that’s a different story. I’m not talking about large studies. Market segments can be OK when we’re looking at 1000s of people. The US census is a good thing – how much are people making, what is their education level, is this skewed by race or income etc. Great!
I’m talking about making and selling products to me and some other folks just because we’re the same on paper. I do research studies of small sample sizes for companies to give them information about their customers’ lives and daily experiences in order to help make good products and services for them. I’ll interview 6, 20, maybe 40 people to get a sense of their daily lives, their struggles, their goals and whatever else they feel like is worth sharing on a particular topic – anything from driving sedans to eating organic foods to using office furniture…whatever the company wants to know more about.
Why market segments suck ass for small sample sizes is that they’re not representative of the larger population. Let’s say my project was to study writers who publish articles online. I might interview a handful of writers with a wide range of backgrounds – range of years of experience, age, ethnicity, gender, location, topics they write about, ways they publish – I want to catch a wide variety. I want to look at a big range because similarities will be all that much more interesting if we compare people that initially appear to be different. If ten writers of different backgrounds tell me, “it takes 2 years of publishing for free before anyone will pay you,” now, wow, that’s surprising. The juicy part is looking across market segments and arbitrary categories to get to the meat of what people are saying and doing. I couldn’t take three of those writers who happen to be black and who happen to write poetry and say, “black writers write poetry.” It’s not true, it’s not representative, it’s misleading.
And that’s why small, qualitative studies aren’t good at answering every type of question. If you want to know how big the market for self-inflating soccer balls is, I can’t help you with small samples. If you want to know what some barriers to exercising are in suburban American, then, Hello, My name is Ethnography. But don’t ask me for barriers to exercise and then ask how many people I talked to own soccer balls, and oh, were they Hispanic? Sigh.
