Market Segmentation Simulator

Segmentation Criteria

Segments (K)

Number of segments 4

Segments

Clustering State

Iteration

Customers

150

Inertia (J)

–

Converged

How it works

Each customer is a point in a 3-D feature space built from the active criteria — for example Demographics plots Age, Income and Household size; Behavior plots Purchase frequency, Average basket and Loyalty; Needs plots Price sensitivity, Quality focus and Service need. Segmentation is performed with k-means clustering, the standard algorithm real marketing teams use to turn a raw customer base into addressable segments:

Assignment:  cluster(x_i) = argmin_k ||x_i − c_k||²
Update:      c_k = mean of all x_i assigned to cluster k
Inertia:     J = Σ_i ||x_i − c_{cluster(i)}||²   (lower = tighter segments)

Run k-means step — performs one assignment + centroid-update pass (Lloyd's algorithm); repeat until inertia stops falling.
Auto-run — steps automatically until the centroids stop moving (convergence), the same stopping rule production clustering pipelines use.
Reseed centroids — picks K new random starting points, useful to see that k-means can converge to different local optima.
Segmentation criteria — swapping Demographics/Behavior/Needs re-projects the same customers onto a different set of marketing-relevant axes, showing that "the market" looks different depending on what you segment by.

Real-world relevance: this is the core technique behind CRM segment builders and ad-platform audience clustering — a marketer picks K target segments, feeds in customer attributes, and the algorithm groups similar customers so each segment can get its own messaging and budget.

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.

Market Segmentation Simulation