Readers of the blog here might be interested in checking out a recent article in Advertising Age on the "death of consumer segmentation." If you haven't read it already, here is the link:
The Death of Consumer Segmentation?
While this article isn't directly related to market research online communities, it does have some implications for researchers and MROCs. Here are my thoughts on the article...
Reasons why consumer segmentation doesn't always work
In the article, the author (Michael Fassnacht) argues that society is volatile, segments are changing faster than ever before and that people now belong to a multitude of segments. In this sense I think he's right on, especially with the latter point...
For example, we'll often recruit participants for MROCs based on targeted segmentation profiles (typically derived from comprehensive segmentation studies run prior to the community). While participants are recruited to meet these specs, we often notice overlaps as we get to know participants over time. Someone who may have fit a certain segment perfectly at first may end up blending into other segments as we learn more about them.
Overreliance on this segmentation criteria (particularly while conducting research) can be misleading. It's important to leave some room for the "gray area" between segments, and identify those in the analysis. While analyzing the feedback by segment can be a useful exercise, it should be approached with some caution due to the overlaps that often exist.
Reasons why the "self selection" method isn't always accurate
The author also suggests that enabling self-segmentation can be a way to move away from brand-controlled segmentation to a more accurate representation of a segment. However, that might not be the "holy grail" of segmentation either. Consider, for example, the 90-9-1 rule of participation on the web. If you're relying on consumers to self-segment, you are inherently relying on the 9% of interent users who actually self-segment themselves in any meaningful way by customizing their experience in on a social site (e.g., Facebook). In what ways does that inherent participation bias impact the way segments are identified through self segmentation?
To me this is akin to the arguments of "crowdsourcing" back in the day... The buzz around it became so powerful that companies thought consumers would end up creating entire campaigns for them like magic, as long as they provided the means to do that. Of course we all know that isn't true (as we've blogged about before with regards to participation in online communities)... Relying on consumers to self-segment so you can create a more meaningful segmentation profile sounds wonderful in theory, but will still need to be supplemented by brand-controlled segmentation to have an impact in any "actionable" way.
Takeaways and thoughts...
Is consumer segmentation as we know it "dead," only to be replaced by self-segmentation (or a blend of both)? Probably not likely any time soon... To me this holds the same credence as the age-old "the focus group is dead" argument - a lot of huffing and puffing, but little substance.
What do you think? What implications does this have for market research? Comments are welcome...
Matt
(If you're interested in reading more analysis of this article, check out Simon Kendrick's post on the Curiously Persistent blog for a very thorough breakdown of the article...)