There’s an insightful article called Marketing Malpractise: The Cause and the Cure in December’s Harvard Business Review. It discusses where market segmentation is failing and how purposeful products offer a solution. It’s thesis is that many segmentation strategies lead marketeers to solve the wrong problems and improve products in ways that are irrelevant to customers needs.
“The structure of a market, seen from the customers’ point of view, is very simple: They need to get things done … When people find themselves needing to get a job done, they essentially hire products to do that job for them. The marketer’s task is therefore to understand what jobs periodically arise in customers’ lives for which they might hire products the company could make.”
Having done this and decided what product to develop, they continue:
“With few exceptions, every job people need or want to do has a social, a functional, and an emotional dimension. If marketers understand each of these dimensions, then they can design a product that’s precisely targeted to the job. In other words, the job, not the customer, is the fundamental unit of analysis”
For technology products I don’t think you can totally exclude the customer’s capabilities from how the job will be accomplished. If the job is reading-internet-email, there’s going to be a big difference between how someone who receives one e-mail a day and someone who receives a thousand a day expects to solve the job. You could more tightly define the product but then you’d land up with a hundred reading-internet-email products.
A post by Havoc Pennington Redhats desktop technical lead ruminates on what this means for complex technology development. He applies this to the development of the GNOME Linux desktop, and to some degree the wider commercial Redhat products. I’ve a feeling that he hits the nail on the head when he talks about flexibility: the jobs that people use technology products for are often sufficently complex that a number of features will be required to satisfy them. This in turn means that the feature set will meet the needs of those trying to accomplish a number of different jobs. Jotspot is an example of this, where they have multiple mini-applications on top of their base product.
Returning to the articles criticims of common segmentation practises. In my experience small businesses’ often segment by types of customer because that’s the data that is accessible, cheap and easy to use for metrics. It’s measurable that a product has 5% penetration of the 0-250 employee sized companies in the UK mainland: good luck saying that a product has an unknown level of penetration in the reading-internet-email job and could you have more money next year?!.
The article also covers building purpose brands and how brands should be extended. There are alternative summaries of the paper at Corante, Marketing Bytes and Noise Filter. If you want some further reading then Manyworlds has an interesting system where you can find related articles.
Is Christensen right, are jobs the fundamental unit of analysis