Typing Tools are classification algorithms used to put consumers into the segments of a segmenation scheme. Another way of putting it is that they are predictive models for limited dependent variables.
It's not unusual for an enterprise to want to put new customers into segments that have been previously defined. For example, a firm might have developed a post hoc descriptive benefits segmentation scheme based on a sample of customer data, and then need to classify other customers using their segmentation scheme in order to target them.
Now if the same data, i.e., measures on the segmentation basis variables, were to be available for the new customers, developing a typing tool (classifier) for the firm to use should in principle be straightforward. But what if it is impractical or infeasible to collect the data on the basis variables for new customers? For example, what if the firm can't collect survey responses about benefits from all new customers? What can be done to develop an effective typing tool, then?