Skip to content

Prajaktahz/Customer-Segmentation-Using-RFM-and-KMeans

Repository files navigation

📊 Customer Segmentation and RFM Analysis for Foodcorp

Project Overview

The Foodcorp Customer Segmentation and RFM Analysis project focuses on segmenting our customer base to enhance marketing efforts and improve customer satisfaction. By analyzing transactional data from 3,000 customers over six months, we identified six distinct customer segments using Recency, Frequency, and Monetary (RFM) metrics. The insights gained will allow Foodcorp to implement targeted marketing strategies that drive customer retention and engagement.


📁 Data Preparation

The dataset consists of four distinct files detailing the transactional behavior of our customers. Key highlights of the dataset include:

  • Total Customers: 3,000
  • Duration: 6 months of transactional data
  • Unique Categories: 20
  • Unique Products: 20,466
  • Data Quality: No null values present in all four files

🔍 Analysis Methodology

Data Exploration and Cleaning

  • Data Exploration: Analyzed data distributions and relationships among features to understand customer behavior.
  • Data Cleaning: Ensured data quality by removing duplicates and handling any inconsistencies.

Feature Significance and Clustering

  • Feature Significance Analysis: Evaluated the importance of various features to identify key drivers of customer behavior.
  • K-means Clustering: Applied K-means clustering to segment customers based on their transactional behaviors.

Dimensionality Reduction

  • Principal Component Analysis (PCA): Utilized PCA to reduce dimensionality and simplify the data complexity, making the clustering process more effective.

image

📈 RFM Analysis Results

The RFM analysis revealed the following six customer segments:

Customer Segment Recency (days) Frequency (visits) Monetary Value (£) Percentage of Customer Base Marketing Action
VVIP Customers 0.2 131.5 1399.8 18.7% Exclusive offers, VIP events, Personalized experiences
Platinum Customers 0.9 87.5 1011.1 10.2% Loyalty rewards, Upgrades, Cross-selling opportunities
Loyal Customers 2.6 64.8 800.8 24.7% Loyalty programs, Referral incentives, Surprise gifts
Need Attention 5.9 43.6 567.3 22.2% Re-engagement campaigns, Personalized recommendations
At Risk Customers 13.3 29.7 403.6 15.9% Win-back promotions, Targeted discounts, Feedback requests
Lost Customers 46.8 16 209.6 8.4% Reactivation campaigns, Customer surveys, Last chance offers

image

💡 Marketing Strategy Suggestions

To optimize engagement and enhance customer relationships, we recommend implementing the following strategies for key segments:

  • Segment 3 (Loyal Customers):

    • Personalized Loyalty Programs: Offer exclusive benefits and rewards tailored to high-spenders and frequent visitors.
    • Upselling and Cross-Selling: Recommend complementary products and highlight premium items in miscellaneous categories.
  • Segment 6 (Lost Customers):

    • Reactivation Campaigns: Develop strategies to win back lost customers through targeted promotions.
    • Appreciation and Recognition: Express gratitude through personalized gestures and discounts.

By focusing marketing efforts on these segments, Foodcorp can foster strong relationships, drive repeat purchases, and encourage brand advocacy.


image

🔄 Suggestions for Further Analysis

  • Periodic Evaluations: Conduct regular assessments of marketing campaigns to measure effectiveness and identify areas for improvement.
  • Customer Engagement Metrics: Analyze customer engagement data to refine marketing strategies continuously.
  • Behavioral Trends: Explore long-term trends in customer behavior to inform future product offerings and promotions.

📌 Conclusion

The Customer Segmentation and RFM Analysis for Foodcorp project provides valuable insights into consumer behaviors and preferences. By leveraging these insights, Foodcorp can implement targeted marketing strategies that enhance customer satisfaction and retention, ensuring sustainable growth and profitability in the competitive market.

About

Analyzed transactional database for 3000 customers in store using Python, identified key features & engineered new attributes to minimize correlations. Implemented PCA to reduce dimensionality & conducted clustering to segment consumers into 6 groups and proposed top segments for marketing.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors