“Create a customer segmentation dendrogram with Newsletter, Webinar and Trial in a Prospects cluster; Annual, Enterprise and Monthly in a Customers cluster; a customer-distance scale; and a cut at 2.5.”
看懂背後的決策。
Use names that distinguish prospects from paying customers.
Define the behavioural inputs behind customer distance before sharing the tree.
Use 2.5 to create an initial set of workable campaign groups.
Use this diagram to explain a small segmentation model when the order of cluster merges matters as much as the final groups.
常見問題
What is a customer segmentation dendrogram?
It is a hierarchy of customer groups built from their similarity on selected behaviours or attributes. Branches show the order in which segments merge.
What should customer distance represent?
It should represent a documented measure of difference, such as engagement, purchase cadence or product usage. The inputs should be consistent across every customer.
How do I choose a cluster cut?
Choose a threshold that produces a number of groups the team can act on, then check that each group has a meaningful business interpretation.