Market Segmentation Using Cluster Analysis Based On Survey Data

Project Description

We've conducted a large consumer survey (5,000 respondents) that will yield approximately 100 variables about individual survey takers and are looking to use this data set to inform a market segmentation using cluster analysis (Two-Step, Latent Class and/or K-means), followed by discriminate analysis to identify the most important predictors of segment membership. Variables include relatively standard consumer market segmentation qualities:

• Demographics
• Psychographics
• Technographics
• Attitudinal Statements
• Shopping/Buying Habits & Behaviors

We have a set of hypotheses that can be used as a starting point for this analysis and are looking to test these and formalize them through statistical analysis of the data, which currently resides in Qualtrics and can be easily exported in the desired format.

Key deliverables include:

• Receive briefing on the data and project
• Collaborate on hypothesis establishment for segment
• Formulate recommended clustering approach and methodology for approval
• Run clustering analysis (3 rounds, with refinements between)
• Descriptive analysis of clusters and identifications of intereting similarities/differences
• Final report document on segments and key characteristics