Traditional customer segmentation still works...sometimes

In my last job, I spent a lot of time thinking about how traditional customer segmentation was an unsophisticated, dated way to think about markets. Just for clarity, I am defining a "traditional" customer segmentation as a way of dividing up customers by certain (usually) observable characteristics. Think: ages, gender, income level, political party affiliation, dietary restrictions, etc.

The "correct" way I learned to view market segmentation was through needs. That is, you should figure out what needs drive customer purchasing behaviors, and then determine how to cater to / reach these different need groups. As a simple example, say you had 3 customers: an old man, a young woman, and an old woman. If you had to form 2 groups with these characteristics, you might decide either of the 2 options:

• Old man and old woman | young woman
• Old woman and young woman | old man

Imagine that you're a hotel company: how useful would it be to define a brand based on age? Wouldn't it be more useful to know what the customers wanted from a hotel stay?

Under the alternative approach, I would try to understand the needs of these different customers first. In this example, the old man and the young woman are both planning a tropical beach vacation while the old woman is planning a cultural trip to a large city. Now, it's much easier for a hotel company to target a specific group of consumers: those with a specific need. But, without understanding the market needs, you may not have naturally grouped some of these customers together.

This approach makes a ton of sense in practice: think about the application to air travel (business trip versus leisure trip regardless of personal income), restaurants (night out with the kids versus night out with friends), etc. And, while it's still helpful to characterize the customers within / between these need groups (can help with marketing if certain attributes stand out), it becomes a dependent variable.

So how does this relate to wine?

Terry's presentation opened my eyes to the fact that wine marketing is effectively based on traditional segmentation. Yes, certain, distinct needs exist in the wine market, but given the unbelievable variety of products that exist to fulfill almost every need (think about the abundance of wine at different price points / different taste profiles / different "prestige" levels), traditional segmentation is perhaps a significantly more effective exercise than a needs-based approach. 

Terry segmented the market based on observable characteristics: cowboys, middle sisters, loners... and then made the product appeal to the emotions of those segmented groups. The dependent variable here was the needs of the different groups (e.g. what taste profile the wine was designed to reach).

So maybe, traditional segmentation is still alive and well, under certain market conditions. It's something I'll consider more in the future.