Customer Segmentation Reasoning Chain

validated business chain v1.0.0 cc-by-sa

Block ID: e9660a9d-2d8f-40e1-8900-20eab968db6b

Community-contributed block. PromptDNA makes no guarantee of output quality or fitness for purpose. User assumes all responsibility for use.

Template

Build a useful customer segmentation for {segmentation_context}, since a segmentation that doesn't change what you do is a decoration. (1) Start from the decision the segmentation must inform — who to target, how to price, what to build — because the right segmentation dimension depends on the decision, and segmenting by convenient demographics when behavior drives value produces neat charts that guide nothing. (2) Choose segmentation variables that actually predict differences in value, needs, or behavior, favoring what customers do and need over who they are, since the classic error is segmenting on easily-measured attributes that don't correlate with anything actionable. (3) Test that the resulting segments are genuinely different from each other on the dimension that matters, substantial enough to be worth serving distinctly, and reachable — a segment you can't identify or reach in practice is useless however real. (4) Size each segment and estimate its value and cost to serve, since the point is to focus on the segments worth winning, not to catalog all customers. (5) Identify which segments to prioritize, target, or deprioritize, accepting that a good segmentation implies serving some customers better and others less. (6) State the segmentation, the priority segment and why, and how it changes a specific decision — because the test of a segmentation is the actions it drives, not its elegance.

Build a useful customer segmentation for {segmentation_context}, since a segmentation that doesn't change what you do is a decoration. (1) Start from the decision the segmentation must inform — who to target, how to price, what to build — because the right segmentation dimension depends on the decision, and segmenting by convenient demographics when behavior drives value produces neat charts that guide nothing. (2) Choose segmentation variables that actually predict differences in value, needs, or behavior, favoring what customers do and need over who they are, since the classic error is segmenting on easily-measured attributes that don't correlate with anything actionable. (3) Test that the resulting segments are genuinely different from each other on the dimension that matters, substantial enough to be worth serving distinctly, and reachable — a segment you can't identify or reach in practice is useless however real. (4) Size each segment and estimate its value and cost to serve, since the point is to focus on the segments worth winning, not to catalog all customers. (5) Identify which segments to prioritize, target, or deprioritize, accepting that a good segmentation implies serving some customers better and others less. (6) State the segmentation, the priority segment and why, and how it changes a specific decision — because the test of a segmentation is the actions it drives, not its elegance.

Variables

Name	Type	Required	Trust level
segmentation_context		yes

segmentationcustomer-strategytargetingchain-of-thought

Ratings

0.0

Overall (0)

0.0

Accuracy

0.0

Consistency

0.0

Clarity

0.0

Efficiency

Benchmarks

Not yet self-validated against any benchmark. Automated, evaluative only - not a factor in whether this block was published.

Submitted by James P FounderMod via mcp · 2026-07-18