Survey Analysis

Client Context

A market research firm had collected data from 3,200 consumers across the UK for a major retail client. The survey included 45 Likert-scale items measuring brand perception, purchase intention, and customer satisfaction, alongside demographic variables. The client needed statistically rigorous segmentation and driver analysis to inform their rebranding strategy.

The Challenge

The dataset presented several challenges: 12% of responses had missing data patterns suggesting non-random attrition, several items showed high multicollinearity, and the client needed both a segmentation of their customer base and a key driver analysis identifying which brand attributes most strongly predicted purchase intention. The timeline was four weeks.

Our Approach

We began with data cleaning and multiple imputation using chained equations (MICE) to handle missing data. Exploratory factor analysis (EFA) with promax rotation reduced the 45 items to eight interpretable factors. We then conducted a two-step cluster analysis (hierarchical followed by k-means) on factor scores, identifying four distinct customer segments. Key driver analysis used dominance analysis on a logistic regression model predicting high purchase intention, quantifying each driver's relative importance. All analyses were performed in SPSS and R, with reproducible syntax files provided to the client.

Results

The four segments — Brand Loyalists, Value Seekers, Convenience Buyers, and Experience Chasers — each showed distinct demographic profiles and driver priorities. For Brand Loyalists, perceived quality was the strongest predictor of purchase intention (dominance weight = 0.34), while for Value Seekers it was price perception (dominance weight = 0.41). The client used these insights to redesign their loyalty programme and target advertising spend, reporting a 15% increase in repeat purchases within the following quarter.

Client Testimonial

"The segmentation and driver analysis gave us clarity we simply did not have before. We could finally see which brand attributes moved the needle for each customer group."

— Head of Insights, RetailMetrics Group

[Explore More Case Studies → /case-studies/]

[Discuss Your Survey Analysis → /contact/]