The purpose of personas (or Personae in the sarcastic English) is to impart empathy in design.
The purpose of market segments vary depending on who you’re talking to. If you’re talking to a marketing science analyst, they should tell you that the purpose of a market segment is to use variations in self-referential communities to adjust the marketing message (positioning) so that it is more relevant to a particular audience, ultimately resulting in higher profits for the business.
Put another way, a market segment should, ideally, impart empathy in quantitative marketing design.
Personas are traditionally the product of qualitative insights.
Market segments are traditionally the product of quantitative insights.
Institutionally, neither side really wants to listen to each other. (In fact, it’s hard enough to get Data Miners to even come out to Web Analytics Wednesdays – and Web Analysts are quants!). Culture and jargon matters.
I don’t believe that Market Segments, as they’re being done today by most of the major companies, are really imparting any empathy in marketing design. They produce 120 cells of variations and walk away – what is a copywriter supposed to do with that? But they could be upgraded and improved to do so.
I believe that Personas are sometimes effective at producing empathy in design. I’ve sat in enough Creative meetings to know that it’s effective. Humans being humans though, Personas don’t always work: I’ve read case studies where the ultimate design had nothing to do with the personas.
What’s ultimately needed is a third type of product: a hybrid of personas and market segments. Of course, that’s blasphemy to both sides of the community. Quals and IA’s would rightly freak out. Quants and Data Miners would rightly freak out. Sure, there are differences in the techniques and purposes of both – but I see more overlap between the two having sat on both sides of the fence.