Predactica

How Predactica's AI-Powered Customer Churn Prediction Model Helped Medical Device Supplier Save 35% Annual Revenue

Challenge

The medical device supplier for retailers faced high customer attrition and lost 35% of annual revenue due to churn. Predactica developed a customer churn prediction model to help identify which customers were likely to leave, allowing the company to allocate resources and retain customers.

Solution

Predactica used AI and machine learning to develop a customer churn prediction model for the company. The model was trained on historical data, identified key factors related to churn, and was highly accurate in predicting churn using real-time data. Predactica’s explainable AI (XAI) also helped identify reasons for customer churn.

Business Impact

  1. Identification of Churn Factors through Data Analysis
  2. Enhanced customer lifetime value
  3. Substantial Improvement in Customer Retention Rates

Company Overview

The large wholesale supplier of medical devices is a global company that specialises in providing medical devices to retailers, hospitals, and clinics. The company has a presence in multiple regions and serves customers across the healthcare industry. The company faced a significant challenge of customer churn, which was affecting their profitability and market share. The company had over 5,000 customers, and it was difficult to identify which ones were likely to leave.
They had been using traditional methods to analyse customer behaviour, but it was not yielding the desired results and it was too late by the time customer churn was detected. The daunting task appeared manageable after the company uncovered the potential of AI and machine learning to offer a solution.

Challenges

The company was facing extreme customer attrition. It costs hundreds of dollars to acquire a new customer. As they had a large customer base, it was difficult to identify which customers were likely to leave. This made it challenging for the company to allocate resources and develop strategies to retain customers. The company’s sales and marketing team were not able to keep up with the growing churn rate, and the company was losing an estimated 35% of annual revenue due to churn.
The company turned to Predactica, to develop a customer churn prediction model that could help them define the churn parameters and thus help in identifying customers who were likely to churn.

Key Steps Taken to Identify Parameters of Churn:

  • Data Collection: Gathered data from various sources, such as customer interactions, purchase history, feedback, and demographic information, to understand customer behaviour patterns.
  • Sample Data Analysis: Analysed a sample from the collected data to identify patterns and trends in customer behaviour. Sample data analysis helped to identify a few parameters of churn, such as product performance, customer service, pricing, and competition.
  • Process Set up: The company’s sales team developed a process for servicing the “at-risk” customers included in the recovery effort.

Solution

Predactica worked with the company to develop a customer churn prediction model using Predactica’s AI and machine learning SaaS platform to analyse customer data and predict churn. The model was trained using historical data, and it was able to identify key factors that were correlated with customer churn. The model was then tested using real-time data, and it was found to be highly accurate in predicting customer churn. In addition to identifying churn using accurate predictive models, Predactica’s explainable AI (XAI) capability helped customers identify reasons for why customers are churning.
  • Input Selection: Predactica’s AI model was trained on various inputs, including customer behaviour data, demographic information, purchase history, customer feedback, and customer interactions with the company’s products and services.
  • Feature Engineering: The AI model identified and analysed relevant features that are important in predicting customer churn, such as product quality, customer service, pricing, and competition.
  • Implementation: The company implemented the insights provided by Predactica to improve customer retention, such as adjusting pricing, improving product quality, and enhancing customer service. This allowed the sales and marketing team to proactively engage with at-risk customers and develop targeted strategies to retain them.

Key Results

Reduced Customer Churn Rate:
The company was facing a high customer churn rate of around 18%, which was causing a significant loss in revenue. After implementing Predactica’s AI-powered customer churn prediction model, the company was able to reduce the churn rate by 23% . This resulted in a saving of 35% of annual revenue.
Improved Customer Retention:
The AI-powered model helped the company to identify the factors that were causing customers to leave. By understanding the reasons behind customer churn, the company was able to take corrective measures to retain customers. This led to an improvement in customer retention rate by 30%.
Increased Sales:
The customer churn prediction model not only helped in reducing customer churn but also helped in identifying the customers who were likely to increase their purchases. This led to an increase in sales by 25%, which further helped the company to boost its revenue.

Conclusion

The implementation of Predactica’s AI-based customer churn prediction model allowed the wholesale supplier of medical devices to identify customers who were likely to churn and develop targeted strategies to retain them. The company was able to reduce customer churn rate, improve customer retention, and increase sales. This resulted in a saving of 35% of annual revenue and helped the company to strengthen its position in the market. The sales and marketing team were able to engage with customers more effectively, resulting in a higher retention rate and increased customer loyalty.