1. Project Overview

• Objective:

  Analyze customer purchasing behavior to identify meaningful customer segments and support data-driven targeting strategies.

• Business Question:

  What customer segments exist based on purchasing behavior, and how can they be targeted differently?

• Context:

  Understanding customer behavior is critical for optimizing marketing efforts, improving retention strategies, and maximizing revenue through personalized actions.

2. Dataset

• Source:

  Kaggle – Customer Personality Analysis.

• Size:

  ~2,200 customers | multiple behavioral and demographic variables.

• Main variables:

  • Customer ID.
  • Total Spending.
  • Purchase Frequency.
  • Campaign Acceptance.
  • Demographics (Age, Education, Marital Status).

• Data issues identified:

  • Date fields stored as text.
  • Extreme age values.
  • Multiple campaign acceptance fields.
  • No predefined customer segmentation.

3. Data Cleaning & Preparation

• Issues found:
  • Incorrect data types (dates, numeric fields).
  • Outliers in age.
  • Missing derived behavioral metrics.
• Cleaning steps:
  • Converted date fields to Date format.
  • Created calculated columns (Age, Total Spending, Purchase Frequency).
  • Removed or flagged extreme values.
  • Standardized metrics for analysis.
• Decisions taken:
  • Age capped to realistic ranges.
  • Behavioral metrics prioritized over demographics for segmentation.

4. Analysis

• KPIs:
  • Total Customers.
  • Total Revenue.
  • High Value Customers (%).
• Key metrics:
  • Total Spending per customer.
  • Purchase Frequency.
  • Campaign Acceptance Rate.
• Findings:
  • Small customer segments generate a disproportionate share of revenue.
  • Medium-value customers show strong growth potential.
  • Low-value customers represent volume but low contribution.

5. Visualization (Power BI)

• Dashboard overview:

  4-page interactive dashboard covering overview, segmentation, customer behavior, and business actions.

• Key visuals:

  • KPI cards
  • Customer segmentation bar charts
  • Spending vs Frequency scatter plot
  • Distribution histograms
  • Insight-driven summary visuals