Bank Customer Churn Dataset is a collection of data related to customers of a bank who have either left (churned) or stayed with the bank. This dataset is typically used for predictive modeling to identify patterns and factors that lead to customer churn, enabling banks to take proactive measures to retain customers.

• id: Unique identifier for each customer.

• CustomerId: Unique identifier for the customer account.

• Surname: Last name of the customer.

• CreditScore: Numeric representation of the customer's creditworthiness.

• Geography:str, Gender:str:Country or region where the customer resides ,Gender of the customer (e.g., Male, Female).

• Age: Age of the customer.

• Tenure: Number of years the customer has been with the bank.

• Balance: Current balance in the customer's account.

• NumOfProducts: Number of bank products the customer uses.

• HasCrCard: Binary indicator (0 or 1) for whether the customer has a credit card.

• IsActiveMember: Binary indicator (0 or 1) for whether the customer is an active member.

• EstimatedSalary: Estimated salary of the customer.

• Exited: Binary indicator (0 or 1) for whether the customer has churned (the target).

Dive into Market Research Intellect's Customer Churn Analysis Software Market Report, valued at USD 2.1 billion in 2024, and forecast to reach USD 4.8 billion by 2033, growing at a CAGR of 10.2% from 2026 to 2033.

Churn prediction aims to detect customers intended to leave a service provider. Retaining one customer costs an organization from 5 to 10 times than gaining a new one. Predictive models can provide correct identification of possible churners in the near future in order to provide a retention solution. This paper presents a new prediction model based on Data Mining (DM) techniques. The proposed model is composed of six steps which are; identify problem domain, data selection, investigate data set, classification, clustering and knowledge usage. A data set with 23 attributes and 5000 instances is used. 4000 instances used for training the model and 1000 instances used as a testing set. The predicted churners are clustered into 3 categories in case of using in a retention strategy. The data mining techniques used in this paper are Decision Tree, Support Vector Machine and Neural Network throughout an open source software name WEKA.

1) Data Introduction • The Customer Churn Dataset is a dataset that collects various customer characteristics and service usage information to predict whether or not communication service customers will turn.

2) Data Utilization (1) Customer Churn Dataset has characteristics that: • The dataset consists of several categorical and numerical variables, including customer demographics, service types, contract information, charges, usage patterns, and Turn. (2) Customer Churn Dataset can be used to: • Development of customer churn prediction model : Machine learning and deep learning techniques can be used to develop classification models that predict churn based on customer characteristics and service use data. • Segmenting customers and developing marketing strategies : It can be used to analyze customer groups at high risk of departure and to design custom retention strategies or targeted marketing campaigns.

259

1) Data Introduction • The Telco Customer Churn Dataset includes carrier customer service usage, account information, demographics and churn, which can be used to predict and analyze customer churn.

2) Data Utilization (1) Telco Customer Churn Dataset has characteristics that: • This dataset includes a variety of customer and service characteristics, including gender, age group, partner and dependents, service subscription status (telephone, Internet, security, backup, device protection, technical support, streaming, etc.), contract type, payment method, monthly fee, total fee, and departure. (2) Telco Customer Churn Dataset can be used to: • Development of customer churn prediction model: Using customer service usage patterns and account information, we can build a machine learning-based churn prediction model to proactively identify customers at risk of churn.

If you found the dataset useful, your upvote will help others discover it. Thanks for your support!

This dataset simulates customer behavior for a fictional telecommunications company. It contains demographic information, account details, services subscribed to, and whether the customer ultimately churned (stopped using the service) or not. The data is synthetically generated but designed to reflect realistic patterns often found in telecom churn scenarios.

Purpose:

The primary goal of this dataset is to provide a clean and straightforward resource for beginners learning about:

• Exploratory Data Analysis (EDA): Understanding customer characteristics and identifying potential drivers of churn through visualization and statistical summaries.
• Data Preprocessing: Handling categorical features (like converting text to numbers) and scaling numerical features.
• Classification Modeling: Building and evaluating simple machine learning models (like Logistic Regression or Decision Trees) to predict customer churn.

Features:

The dataset includes the following columns:

• CustomerID: Unique identifier for each customer.
• Age: Customer's age in years.
• Gender: Customer's gender (Male/Female).
• Location: General location of the customer (e.g., New York, Los Angeles).
• SubscriptionDurationMonths: How many months the customer has been subscribed.
• MonthlyCharges: The amount the customer is charged each month.
• TotalCharges: The total amount the customer has been charged over their subscription period.
• ContractType: The type of contract the customer has (Month-to-month, One year, Two year).
• PaymentMethod: How the customer pays their bill (e.g., Electronic check, Credit card).
• OnlineSecurity: Whether the customer has online security service (Yes, No, No internet service).
• TechSupport: Whether the customer has tech support service (Yes, No, No internet service).
• StreamingTV: Whether the customer has TV streaming service (Yes, No, No internet service).
• StreamingMovies: Whether the customer has movie streaming service (Yes, No, No internet service).
• Churn: (Target Variable) Whether the customer churned (1 = Yes, 0 = No).

Data Quality:

This dataset is intentionally clean with no missing values, making it easy for beginners to focus on analysis and modeling concepts without complex data cleaning steps.

Inspiration:

Understanding customer churn is crucial for many businesses. This dataset provides a sandbox environment to practice the fundamental techniques used in churn analysis and prediction.

Dataset

This dataset was created by Subramanian Chandrasekaran

Contents

Explore the 'Bank Churn (test)' dataset, a comprehensive collection designed for evaluating predictive models and analyzing customer attrition in the banking sector. This test dataset, derived from real-world scenarios, offers a robust platform to assess the effectiveness of machine learning algorithms in predicting and understanding bank churn dynamics.

Analyzing customers’ characteristics and giving the early warning of customer churn based on machine learning algorithms, can help enterprises provide targeted marketing strategies and personalized services, and save a lot of operating costs. Data cleaning, oversampling, data standardization and other preprocessing operations are done on 900,000 telecom customer personal characteristics and historical behavior data set based on Python language. Appropriate model parameters were selected to build BPNN (Back Propagation Neural Network). Random Forest (RF) and Adaboost, the two classic ensemble learning models were introduced, and the Adaboost dual-ensemble learning model with RF as the base learner was put forward. The four models and the other four classical machine learning models-decision tree, naive Bayes, K-Nearest Neighbor (KNN), Support Vector Machine (SVM) were utilized respectively to analyze the customer churn data. The results show that the four models have better performance in terms of recall rate, precision rate, F1 score and other indicators, and the RF-Adaboost dual-ensemble model has the best performance. Among them, the recall rates of BPNN, RF, Adaboost and RF-Adaboost dual-ensemble model on positive samples are respectively 79%, 90%, 89%,93%, the precision rates are 97%, 99%, 98%, 99%, and the F1 scores are 87%, 95%, 94%, 96%. The RF-Adaboost dual-ensemble model has the best performance, and the three indicators are 10%, 1%, and 6% higher than the reference. The prediction results of customer churn provide strong data support for telecom companies to adopt appropriate retention strategies for pre-churn customers and reduce customer churn.

Analysis of ‘Churn Modelling’ provided by Analyst-2 (analyst-2.ai), based on source dataset retrieved from https://www.kaggle.com/shrutimechlearn/churn-modelling on 28 January 2022.

--- Dataset description provided by original source is as follows ---

Content

This data set contains details of a bank's customers and the target variable is a binary variable reflecting the fact whether the customer left the bank (closed his account) or he continues to be a customer.

Acknowledgements

Big thanks to https://www.superdatascience.com/pages/deep-learning Banner Photo by Sharon McCutcheon on Unsplash

--- Original source retains full ownership of the source dataset ---

Analysis of ‘Churn Modelling’ provided by Analyst-2 (analyst-2.ai), based on source dataset retrieved from https://www.kaggle.com/shubh0799/churn-modelling on 28 January 2022.

--- Dataset description provided by original source is as follows ---

Context

The dataset is the details of the customers in a company.

Content

The column are about it's estimated salary, age, sex, etc. Aiming to provide all details about an employee.

--- Original source retains full ownership of the source dataset ---

• Main objective: Find out customers who will churn and who will not.
• Methodology: It is a classification problem. We will use decision tree and random forest to predict the outcome.
• Steps Involved
• Read the data
• Check for data types https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F10868729%2F1ffb600d8a4b4b36bc25e957524a3524%2FPicture1.png?generation=1688638600831386&alt=media" alt="">

1. Change character vector to factor vector as this is as classification problem
2. Drop the variable which is not significant for the analysis. We drop "customerID".
3. Check for missing values. None are found.
4. Split the data into train and test so we can use the train data for building the model and use test data for prediction. We split this into 80-20 ratio (train/test) using the sample function.
5. Install and run libraries (rpart, rpart.plot, rattle, RColorBrewer, caret)
6. Run decision tree using rpart function. The dependent variable is Churn and 19 other independent variables

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F10868729%2F8d3442e6c82d8026c6a448e4780ab38c%2FPicture2.png?generation=1688638685268853&alt=media" alt=""> 9. Plot the decision tree

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F10868729%2F9ab0591e323dc30fe116c79f6d014d06%2FPicture3.png?generation=1688638747644320&alt=media" alt="">

Average customer churn is 27%. The churn can take place if the tenure is more than >=7.5 and there is no internet service

1. Tuning the model
2. Define the search grid using the expand.grid function
3. Set up the control parameters through 5 fold cross validation
4. When we print the model we get the best CP = 0.01 and an accuracy of 79.00%

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F10868729%2F16080ac04d3743ec238227e1ef2c8269%2FPicture4.png?generation=1688639197455166&alt=media" alt="">

1. Predict the model
2. Find out the variables which are most and least significant. https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F10868729%2F61beb4224e9351cfc772147c43800502%2FPicture5.png?generation=1688639468638950&alt=media" alt="">

Significant variables are Internet Service, Tenure and the least significant are Streaming Movies, Tech Support.

USE RANDOM FOREST

1. Run library(randomForest). Here we are using the default ntree (500) and mtry (p/3) where p is the number of independent variables. https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F10868729%2Fc27fe7e83f0b53b7e067371b69c7f4a7%2FPicture6.png?generation=1688640478682685&alt=media" alt="">

   Through confusion matrix, accuracy is coming 79.27%. The accuracy is marginally higher than that of decision tree i.e 79.00%. The error rate is pretty low when predicting "No" and much higher when predicting "Yes".

2. Plot the model showing which variables reduce the gini impunity the most and least. Total charges and tenure reduce the gini impunity the most while phone service has the least impact.

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F10868729%2Fec25fc3ba74ab9cef1a81188209512b1%2FPicture7.png?generation=1688640726235724&alt=media" alt="">

1. Predict the model and create a new data frame showing the actuals vs predicted values

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F10868729%2F50aa40e5dd676c8285020fd2fe627bf1%2FPicture8.png?generation=1688640896763066&alt=media" alt="">

1. Plot the model so as to find out where the OOB (out of bag ) error stops decreasing or becoming constant. As we can see that the error stops decreasing between 100 to 200 trees. So we decide to take ntree = 200 when we tune the model.

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F10868729%2F87211e1b218c595911fbe6ea2806e27a%2FPicture9.png?generation=1688641103367564&alt=media" alt="">

Tune the model mtry=2 has the lowest OOB error rate

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F10868729%2F6057af5bb0719b16f1a97a58c3d4aa1d%2FPicture10.png?generation=1688641391027971&alt=media" alt="">

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F10868729%2Fc7045eba4ee298c58f1bd0230c24c00d%2FPicture11.png?generation=1688641605829830&alt=media" alt="">

Use random forest with mtry = 2 and ntree = 200

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F10868729%2F01541eff1f9c6303591aa50dd707b5f5%2FPicture12.png?generation=1688641634979403&alt=media" alt="">

Through confusion matrix, accuracy is coming 79.71%. The accuracy is marginally higher than that of default (when ntree was 500 and mtry was 4) i.e 79.27% and of decision tree i.e 79.00%. The error rate is pretty low when predicting "No" and m...

The Customer Churn Analysis Software market is experiencing robust growth, driven by the increasing need for businesses to understand and mitigate customer attrition. The market's expansion is fueled by several factors, including the rising adoption of cloud-based solutions, the proliferation of big data analytics, and the growing demand for predictive analytics capabilities to proactively identify at-risk customers. Businesses across diverse sectors, including SaaS, e-commerce, and telecommunications, are increasingly leveraging these sophisticated tools to gain actionable insights into customer behavior, personalize their offerings, and improve customer retention strategies. This market is characterized by a competitive landscape with both established players like Adobe and Google, and specialized niche providers such as Infer and Churnly Technologies Limited. The integration of AI and machine learning capabilities within these platforms is a prominent trend, enabling more accurate prediction models and automated interventions to reduce churn. While the initial investment in such software can be a restraint for some smaller businesses, the long-term return on investment, in terms of improved customer retention and reduced acquisition costs, is a compelling driver for market growth. The forecast period (2025-2033) is expected to witness significant expansion, building upon the historical growth from 2019-2024. Assuming a conservative CAGR (let's estimate it at 15% based on industry trends), and a 2025 market size of $5 billion (a reasonable estimate given the presence of major players and the importance of the sector), the market is projected to reach approximately $17 billion by 2033. This expansion will be propelled by continuous technological advancements, the growing adoption of subscription-based business models, and a heightened focus on customer experience management across industries. Regional variations will likely exist, with North America and Europe leading the market initially due to higher adoption rates and technological infrastructure, but emerging markets in Asia-Pacific are expected to show significant growth in the later years of the forecast period. The competitive landscape will remain dynamic, with mergers, acquisitions, and the emergence of innovative solutions shaping the future of customer churn analysis software.

In recent years, with the continuous improvement of the financial system and the rapid development of the banking industry, the competition of the banking industry itself has intensified. At the same time, with the rapid development of information technology and Internet technology, customers’ choice of financial products is becoming more and more diversified, and customers’ dependence and loyalty to banking institutions is becoming less and less, and the problem of customer churn in commercial banks is becoming more and more prominent. How to predict customer behavior and retain existing customers has become a major challenge for banks to solve. Therefore, this study takes a bank’s business data on Kaggle platform as the research object, uses multiple sampling methods to compare the data for balancing, constructs a bank customer churn prediction model for churn identification by GA-XGBoost, and conducts interpretability analysis on the GA-XGBoost model to provide decision support and suggestions for the banking industry to prevent customer churn. The results show that: (1) The applied SMOTEENN is more effective than SMOTE and ADASYN in dealing with the imbalance of banking data. (2) The F1 and AUC values of the model improved and optimized by XGBoost using genetic algorithm can reach 90% and 99%, respectively, which are optimal compared to other six machine learning models. The GA-XGBoost classifier was identified as the best solution for the customer churn problem. (3) Using Shapley values, we explain how each feature affects the model results, and analyze the features that have a high impact on the model prediction, such as the total number of transactions in the past year, the amount of transactions in the past year, the number of products owned by customers, and the total sales balance. The contribution of this paper is mainly in two aspects: (1) this study can provide useful information from the black box model based on the accurate identification of churned customers, which can provide reference for commercial banks to improve their service quality and retain customers; (2) it can provide reference for customer churn early warning models of other related industries, which can help the banking industry to maintain customer stability, maintain market position and reduce corporate losses.

Comparison of the proposed algorithms with other ensemble models.

This dataset provides detailed call log records linked to customer churn events, including call metadata, customer demographics, churn reasons, and resolution outcomes. It enables comprehensive analysis of why customers leave, how call center interactions influence churn, and supports the development of targeted retention strategies. The dataset is ideal for churn prediction modeling, root cause analysis, and customer experience optimization.

Dataset

This dataset was created by Vamshi

Contents

Context and Methodology

The dataset originates from the research domain of Customer Churn Prediction in the Telecom Industry. It was created as part of the project "Data-Driven Churn Prediction: ML Solutions for the Telecom Industry," completed within the Data Stewardship course (Master programme Data Science, TU Wien).

The primary purpose of this dataset is to support machine learning model development for predicting customer churn based on customer demographics, service usage, and account information.
The dataset enables the training, testing, and evaluation of classification algorithms, allowing researchers and practitioners to explore techniques for customer retention optimization.

The dataset was originally obtained from the IBM Accelerator Catalog and adapted for academic use. It was uploaded to TU Wien’s DBRepo test system and accessed via SQLAlchemy connections to the MariaDB environment.

Technical Details

The dataset has a tabular structure and was initially stored in CSV format. It contains:

• Rows: 7,043 customer records

• Columns: 21 features including customer attributes (gender, senior citizen status, partner status), account information (tenure, contract type, payment method), service usage (internet service, streaming TV, tech support), and the target variable (Churn: Yes/No).

Naming Convention:

• The table in the database is named telco_customer_churn_data.

Software Requirements:

• To open and work with the dataset, any standard database client or programming language supporting MariaDB connections can be used (e.g., Python etc).

• For machine learning applications, libraries such as pandas, scikit-learn, and joblib are typically used.

Additional Resources:

• Source code for data loading, preprocessing, model training, and evaluation is available at the associated GitHub repository: https://github.com/nazerum/fair-ml-customer-churn

Further Details

When reusing the dataset, users should be aware:

• Licensing: The dataset is shared under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

• Use Case Suitability: The dataset is best suited for classification tasks, particularly binary classification (churn vs. no churn).

• Metadata Standards: Metadata describing the dataset adheres to FAIR principles and is supplemented by CodeMeta and Croissant standards for improved interoperability.

Analysis of ‘Customer Churn’ provided by Analyst-2 (analyst-2.ai), based on source dataset retrieved from https://www.kaggle.com/hassanamin/customer-churn on 14 February 2022.

--- Dataset description provided by original source is as follows ---

Binary Customer Churn

A marketing agency has many customers that use their service to produce ads for the client/customer websites. They've noticed that they have quite a bit of churn in clients. They basically randomly assign account managers right now, but want you to create a machine learning model that will help predict which customers will churn (stop buying their service) so that they can correctly assign the customers most at risk to churn an account manager. Luckily they have some historical data, can you help them out? Create a classification algorithm that will help classify whether or not a customer churned. Then the company can test this against incoming data for future customers to predict which customers will churn and assign them an account manager.

Content

The data is saved as customer_churn.csv. Here are the fields and their definitions:

Name : Name of the latest contact at Company

Age: Customer Age

Total_Purchase: Total Ads Purchased

Account_Manager: Binary 0=No manager, 1= Account manager assigned

Years: Totaly Years as a customer

Num_sites: Number of websites that use the service.

Onboard_date: Date that the name of the latest contact was onboarded

Location: Client HQ Address

Company: Name of Client Company

Once you've created the model and evaluated it, test out the model on some new data (you can think of this almost like a hold-out set) that your client has provided, saved under new_customers.csv. The client wants to know which customers are most likely to churn given this data (they don't have the label yet).

Acknowledgements

We wouldn't be here without the help of others. If you owe any attributions or thanks, include them here along with any citations of past research.

Inspiration

Your data will be in front of the world's largest data science community. What questions do you want to see answered?

--- Original source retains full ownership of the source dataset ---