BLOOM regularly helps clients with segmentation and clustering analyses. Reasons for doing a proper segmentation of your customers differ, but usually include (a combination of) the following:

• Identify the most and least valuable customers
• Predict which new customers are likely to become most valuable
• Focus your marketing efforts
• Get an understanding of the needs of different customer segments
• Develop products that appeal to certain segments

Two common ways to divide your customer base into homogeneous groups are rule-based segmentation and clustering. The former method involves segmenting by previously determined variables (geographic, demographic, psychographic, and behavioural variables). Clustering – e.g. k-means, hierarchical or other more specialised techniques – is not based on any predefined set of rules. Rather, (machine learning) algorithms are used to identify certain patterns from the data itself.

BLOOM supports clients in selecting and executing the segmentation approach that is right for them. And it doesn’t stop there: the insights from these analyses are translated into an actionable end-product – e.g. a RFM matrix, a data science solution, or a step-by-step strategy.