Data Analytics: Customer Segmentation

Use clustering techniques to segment customers for a retail brand based on purchasing behavior.

# Aplly.xyz Case Study Submission

## Title
Data Analytics: Customer Segmentation

## Type
Data Analytics

## Difficulty
Intermediate

## Estimated Time
60 minutes

## Overview
Use clustering techniques to segment customers for a retail brand based on purchasing behavior.

## Case Details

Function Focus: Data Analytics — manual reasoning, decomposition, and critical judgment (no spreadsheet or AI tool permitted in Phase 1)

Scenario:
You are the analyst at a fictional consultancy ("Praxis Advisors") tasked with answering: Data Analytics: Customer Segmentation. You have a clean, synthetic dataset described below. You must produce a defensible answer using structured reasoning — no tool-assisted shortcut on the first pass.

Dataset Structure:
- customer_id
- age_group
- annual_income_lakhs
- annual_spend_lakhs
- purchase_frequency
- avg_ticket_size
- product_category
- channel
- tenure_months
- return_rate_pct
- loyalty_score

Tasks:
1. Define the business question in one sentence and the metrics that would answer it.
2. Split the dataset into meaningful segments/cohorts by hand and compare their profiles.
3. Drill from the headline number down to the component that drives it (decomposition), showing each step.
4. State what the data does NOT tell you, and what additional data you would request.
5. Only after the manual analysis, reproduce the drill-down in a spreadsheet or BI tool and reconcile.

Expected Output:
A one-page analysis memo: business question, segment profiles, decomposition tree, data-gap list.

Evaluation Criteria:
Clarity of the business question, meaningful segmentation, correct decomposition logic, quality of data-gap analysis.

## Data Sources

| customer_id | age_group | annual_income_lakhs | annual_spend_lakhs | purchase_frequency | avg_ticket_size | product_category | channel | tenure_months | return_rate_pct | loyalty_score |
|---|---|---|---|---|---|---|---|---|---|---|
| C0001 | 18-25 | 3.1 | 3.7 | 16 | 1358 | Electronics | Both | 70 | 2.6 | 4.8 |
| C0002 | 18-25 | 4.6 | 3.2 | 40 | 420 | Fashion | Both | 70 | 12.6 | 5.0 |
| C0003 | 36-45 | 20.7 | 0.6 | 12 | 5995 | Groceries | Offline | 20 | 6.5 | 7.9 |
| C0004 | 18-25 | 4.6 | 1.6 | 24 | 5211 | Electronics | Both | 59 | 16.1 | 9.8 |
| C0005 | 46-60 | 4.3 | 3.9 | 42 | 5334 | Groceries | Both | 25 | 21.1 | 1.4 |
| C0006 | 26-35 | 19.9 | 11.8 | 16 | 7392 | Home & Kitchen | Offline | 59 | 19.1 | 4.3 |
| C0007 | 36-45 | 10.5 | 8.2 | 46 | 7974 | Beauty | Online | 22 | 16.0 | 3.2 |
| C0008 | 46-60 | 11.0 | 11.9 | 42 | 5912 | Fashion | Both | 42 | 25.3 | 8.0 |
| C0009 | 26-35 | 21.0 | 9.8 | 27 | 2422 | Fashion | Both | 41 | 6.4 | 5.5 |
| C0010 | 46-60 | 5.7 | 2.1 | 37 | 4674 | Beauty | Both | 55 | 26.9 | 4.6 |


Full dataset: https://github.com/arora200/aplly_case_db/datasets/041-data-analytics-customer-segmentation (synthetic, 50 records)

## Solution Frameworks
Segmentation & cohort analysis, descriptive→diagnostic drill-down, KPI decomposition, visualization critique

## Solver Guidance & Tutorials
_Solver guidance added by the pipeline (tutorial links) — see `solver_guidance` field._

## What You'll Learn
- Translate business questions into metrics
- Segment and profile data by hand
- Decompose headline numbers to root causes

## Tags
Data Science, Analytics, Machine Learning

## Registration Links
Register as Solver / Register as Evaluator