CUSTOMER SEGMENTATION ANALYSIS OF CANNABIS RETAIL DATA: A MACHINE LEARNING APPROACH
AuthorPapetti, Ryan Henry
AdvisorThompson, Richard H.
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractAs the legal cannabis industry emerges from its nascent stages, there is increasing motivation for retailers to look for data or strategies that can help them segment or describe their customers in a succinct, but informative manner. While many cannabis operators view the state-mandated traceability as a necessary burden, it provides a goldmine for internal customer analysis. Traditionally, segmentation analysis focuses on demographic or RFM (recency-frequency-monetary) segmentation. Yet, neither of these methods has the capacity to provide insight into a customer’s purchasing behavior. With the help of 4Front Ventures, a battle-tested multinational cannabis operator, this report focuses on segmenting customers using cannabis-speciﬁc data (such as ﬂower and concentrate consumption) and machine learning methods (K-Means and Agglomerative Hierarchical Clustering) to generate newfound ways to explore a dispensary’s consumer base. The ﬁndings are that there are roughly ﬁve or six clusters of customers with each cluster having unique purchasing traits that deﬁne them. Although the results are meaningful, this report could beneﬁt with exploring more clustering algorithms, comparing results across dispensaries within the same state, or investigating segmentations in other state markets.