Giter Site home page Giter Site logo

customer-segmentation's Introduction

Customer Segmentation: Can We Use Credit Card Habits to Create Effective Customer Segments?

➽ Divide Customers into Segments with a Clustering Model.

GitHub Workflow Status codecov.io contributions welcome contributions welcome

 

Table of Contents

  1. Problem Definition & Objectives
  2. Data Statistics & Exploratory Data Analysis (EDA)
  3. Model Development
  4. Cluster Analysis

 

Problem Definition & Objectives

What's the Problem & Why is it Important?

Understanding customers is essential for any business's success. Instead of using a cookie-cutter marketing strategy, segmenting customers into groups based on common characteristics enables marketers to gain in-depth insights, and develop personalized marketing such as campaigns, products, and offers that are relevant to each group. Rather than relying on assumptions about their preferences and needs.

 

What's the Goal of This Project?

This project aims to develop a clustering model that can divide customers into segments based on their regular patterns and behaviors associated with their credit card usage, while also providing potential marketing strategies for each segment.

 

Data

Data Statistics

  • Source: Credit Card Dataset for Clustering
  • Dataset Structure: 8950 samples (rows), 18 features (variables)
  • Data Types: 1 categorical, 17 numeric
  • Missing cells: 314
  • Missing cells (%): 0.2%
  • Total size in memory: 1.7 MiB
  • Average record size in memory: 199.0 B

 

Exploratory Data Analysis (EDA)

➤ Purchases vs One-off Purchases

The plotted data points is forming an upward pattern from left to right, which refers to a high correlation between the two features as indicated by the heatmap. This correlation could imply that most purchases were one-off purchases.
The distant datapoints may be due to the large number of outliers in the dataset or other factors.

 

➤ Credit Limit vs Balance

The scatter plot is somewhat forming an angeld triangle, where the hypotenuse of the triangle is hinting for a correlation between the two features.

 

➤ Balance vs Minimum Payments

Typically, the MINIMUM_PAYMENTS is often depends on the current BALANCE amount. But it appears that there's a very low correlation between them. Again, it may be due to outliers or other factors.

 

➤ Features That Changes Over Time

In general, all the plotted features against TENURE have an upward trend. Indicating that they are increasing over time.

  

Model Development

Since this is an unsupervised clustering problem , a K-means model was developed for the clustering.

K-means

Based on the above figure, the elbow is located at k=4 Therefore, the model will be created with 4 clusters.

 

The above figures illustrates the clusters distribution in the scatter plot, and the size of clusters in percentages. As indicated by both figures, The clusters are very close in size, with the exception of Cluster 4, which is the smallest cluster.

 

Cluster Analysis

After the examination of the clusters' statistical data and visualizations, the characteristics of each cluster were identified. The following section outlines each segment and potential marketing strategies for each one:

➤ Customers Who Prefer to Pay in Advance

  1. Offer attractive discounts, incentives, or early access to the product to encourage customers to pay in advance.
  2. Build trust by highlighting your secure payment methods and privacy protections.
  3. If applicable, emphasize the advantages of outright ownership that come with making advance payments. This can be particularly effective for big-ticket items.
  4. Reward loyal customers who consistently make upfront payments with exclusive benefits or discounts.

 

➤ Customers Who Prefer to Pay in Installments

  1. Highlight the convenience and flexibility of paying in installments.
  2. Provide clear and transparent information about the terms and conditions of installment plans, including the total cost, interest rates, and payment schedule.
  3. If applicable, offer competitive interest rates and terms.

 

➤ Customers Who Prefer to Make One-Off Payments & Have High Balances

  1. Highlight the exclusivity and quality of your product.
  2. Create a VIP loyalty program that offer them exclusive discounts, promotions, and early access to new products.
  3. If applicable, partner with luxury brands and retailers.
  4. Invest in high-quality branding and marketing materials to convey a sense of luxury and sophistication.
  5. If applicable, highlight how your product can be seen as an investment or a store of value. This is especially relevant for products like jewelry or art.

 

➤ Low Spenders Customers

  1. Offer discounts, promotions, and coupons to make your product even more affordable for low spenders.
  2. Low spenders are careful about where they spend their money, so make sure to clearly communicate the value that your product brings to them.
  3. If applicable, offer samples and free trials. This is a great way for low spenders to try your product before they buy it.
  4. Run social media contests and giveaways.
  5. Offer referral bonuses to customers who refer their friends and family.

 
Overall, considering the different types of payments across segments, make sure that customers are aware of the payment options available. Highlight these options in your marketing materials, website, and checkout process.

Additionally, it might be effective to promote limited-time offers or exclusive discounts for each payment method to create a sense of urgency and encourage customers to act quickly.

 

    ▶️ For more details, please check the notebook file here.

 

░░░░░ Thank you for your interest in this project! I hope you find it helpful😄 ░░░░░

customer-segmentation's People

Contributors

nadaaahh avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.