Of Personas and Market Segments: A reply to Hamel

David Hamel wrote:

The issue behind it all is that the web isn’t static, it is constently changing.  Ever heard of AOL?  Of course you have.  Know anyone still using it? Probably not.  What about MySpace?  Also there is the inevitable the march of time.  Your persona for Bob has his age at 52.   In five years time, will Bob still be useful?  Probably the difference between 52 and 57 isn’t that large.  But what if your target demographic is 22?  There is a much larger difference between the interestes of a 22 year old and a 27 year old…In conclusion, use personas but don’t let them get stagnate, your personas represent people and people change.

Hamel rightly points out that personas come into being and are sometimes printed out on huge boards and laid out in the office. And then they’re taken down after the redesign and we forget about them. They don’t evolve. They don’t breathe.

I’d argue that the typical market segment, “male, 56-65, rural” is incredibly bland and not adequate at all. I think somewhere along the line, perhaps over the decades, we forgot why market segments are supposed to be powerful: like people talk and they are self-referential. This was demonstrated with the original Word of Mouth marketing studies focused on seed distribution. Word of mouth is critical. No company can possibly afford to pay for every conversion.

The segment should describe, instead, where rural older males are talking and how do you get them to refer you along. A person is so much more than just an age, gender, income bracket, number of kids and general DMA. Quantitative marketing science is capable of real contributions and advancement in that field.

If personas impart empathy in design, then segments should impart empathy in marketing.

And directly to Hamel’s point: they should be periodically revisited.

That might not sit well with the common “rip’em down and reinvent them from scratch” mentality that dominates the world – but I’m very comfortable with the notion of optimizing segments and personas over time as a program of ‘learning’.

I’d like to see more of that.