This repository hosts the code for analyzing customer churn based on transaction data for Ödeal. The analysis provides valuable insights into customer retention and loss patterns over time.

Churn analysis is the evaluation of customer attrition in a business. By tracking when customers discontinue their transactions or interactions, businesses can identify patterns and reasons behind the loss of clients.
- Customer Retention Insight: By understanding the churn rate, Ödeal can measure how well it retains customers over time.
- Identifying At-Risk Customers: Early identification of customers who may churn allows Ödeal to take preemptive action to retain them.
- Service Improvement: Insights from churn patterns can guide Ödeal in refining their services to meet customer needs more effectively.
- Targeted Marketing: Analyzing churn helps in tailoring marketing efforts towards the right customer segments, enhancing ROI.
- Revenue Growth: Ultimately, reducing churn rates can lead to increased revenue and growth for Ödeal.
The function analyze_customer_churn
takes transaction data and calculates churn based on a specified threshold of inactive days. By default, if a customer has no transactions for 90 days, they are considered to have churned.
df
: A pandas DataFrame containing the transaction data.id_col
: The name of the customer ID column.transaction_date_col
: The column containing the transaction dates.transaction_id_col
: The column containing the transaction IDs.churn_days_threshold
: The threshold in days to determine if a customer has churned.
A DataFrame with churn analysis by customer, including churn status and average transaction frequency per month.
This project is licensed under the Apache License 2.0 - see the LICENSE file for details.