The common practice is to segment markets along demographics, psychographics, geographies, and behaviors. From my experience, the first three segmentation techniques are not actionable, as they fail to discover homogeneous, mutually exclusive, and collectively exhaustive segments. The last method, behavioral segmentation, has descriptive (see Amplitude) yet no predictive value. Our intention is to fix the market variable and to then design a product accordingly. Therefore, we require predictive capabilities. Toward that end, we can leverage customer needs. Customer needs drive customer behavior. Therefore, customer needs segmentation is the other side of the coin to behavioral segmentation.

Previously, we uncovered customer needs. For example, in the context of quality assurance, while monitoring tests (process step), product teams need to minimize the likelihood that reports are non-descriptive (customer need). Thereafter, we surveyed customers to quantify their needs according to importance and satisfaction. That gave us an opportunity score for each customer need: Importance + (Importance –Satisfaction). As a next step, we can employ factor analysis to identify those customer needs that survey participants disagree on—some say those are important and unsatisfied, while others say the opposite. Thereupon, we can employ cluster analysis to arrive at homogeneous, mutually exclusive, and collectively exhaustive segments.

As part of the customer survey, we also collected contextual data. That data, in tandem with additional conversations with customers, allows you to personify each segment.

That concludes the series on Shipping What Matters. It is now up to you to serve your market accordingly (product development).