The Credit Pros implemented Pecan AI to address client churn, reducing model development time from three months to just weeks and enabling proactive retention strategies that increased revenue.
The Challenge
The Credit Pros (TCP) is a financial technology firm that assists consumers in repairing their credit. Unfortunately, even though maintaining good credit is an important need for today’s consumers, TCP faced a customer churn challenge.
The ideal way to reduce customer churn is to catch it before it occurs so you can intervene. But the process of predicting which clients were likely to churn required training machine learning models, writing complex queries, and making predictions — a time-consuming and resource-intensive task.
TCP needed a solution that could streamline this process and provide actionable predictions quickly. Additionally, the company wanted to improve its ability to retain clients and increase revenue. To accomplish all these goals, they needed efficient predictive analytics tools.
Adopting Pecan AI was a strategic decision for The Credit Pros because it is user-friendly, intuitive, and integrates seamlessly into our data analytics workflow. It allows us to go from predictive questions to predictions efficiently.
Michael Mgbame, Data Analyst, The Credit Pros
The Solution
Pecan’s intuitive and powerful predictive analytics platform helped TCP address these challenges.
Michael Mgbame, a data analyst at TCP since October 2021, describes the onboarding process: “Getting started with Pecan AI was quite straightforward. The tool allows data analysts to query directly from data sources like Snowflake and connect to various data sources easily.” The platform’s user-friendly interface, combined with its AI chatbots, made it simple for TCP to move from their business needs to predictions seamlessly and quickly.
One of the key features that proved valuable was Pecan’s AI Assistant, which provides explanations and suggestions along the predictive modeling journey. “Pecan AI made this process easy by guiding us through querying, modeling, and predicting,” Michael says.
And, to ensure that TCP could quickly act on predictions, Pecan also facilitated writing predictions to other software like Salesforce.
The Pecan team’s support was crucial in overcoming challenges and maximizing the platform’s potential. “The Pecan team has been incredibly supportive, providing expert assistance whenever needed. For instance, they helped resolve issues with pushing predictions to Salesforce by collaborating with our Salesforce administrator and me,” Michael says.
The Results
The implementation of Pecan led to significant improvements in TCP’s ability to predict and mitigate client churn. Michael explains, “Currently, Pecan helps us predict the probability of client churn, focusing on whether a client will churn 30 days after a specific date. We write these probabilities to Salesforce, allowing our business team to devise treatments to mitigate churn.”
Getting proactive about churn with AI’s help has had a substantial impact on TCP’s business. “This approach has increased our revenue by retaining clients longer and reducing churn,” Michael says.
The efficiency gains were also remarkable: “Previously, it took us about three months to develop a model to predict churn. With Pecan, we achieved this within two to three weeks, enabling faster implementation and improvement,” Michael says.
Moreover, the adoption of Pecan AI has enhanced the skills and capabilities of TCP’s data analytics team. Michael reflects, “My knowledge and experience with machine learning have significantly improved due to the expertise and support provided by the Pecan team. The Pecan AI team also provides excellent documentation and resources, making it easier to understand and use the platform effectively.”
For the TCP team, Pecan is now an integral part of a comprehensive process from predictive modeling to business impact.
