How a finanacial firm leveraged machine learning (ML) for targeted marketing initiatives and increased account opening rates by 15% compared to an untargeted approach.
INTRO AND CLIENT BACKGROUND
In the dynamic landscape of financial services, our client, a tech-forward industry leader providing simple, personalized payment, lending and saving solutions.
Their portfolio includes private label credit cards for multiple brands, loyalty programs, and direct marketing, derived from the capture and analysis of transaction-rich data.
The organization focuses on harnessing customer insights to build meaningful relationships between brands and their customers making them a leader in their industry.
BUSINESS GOALS
The core business goal is to develop a Propensity to Open (PTO) predictive ML model to enable targeted marketing.
This model aimed to pinpoint brand customers with a high likelihood of opening a new credit card account.
By integrating this predictive model into acquisition campaigns, the goal was to enhance targeting precision and improve the acquisition of new cardholders.
KEY CHALLENGES
1. Data Quality Issues: The journey began with challenges in the data provided by the brand. Collaborative efforts with the brand and the Data Engineering team were crucial in resolving these quality issues.
2. Low Response Rate (RR): The initial RR of 0.9% indicated a significant imbalance between positive and negative classes. Where, positive (1) indicates account "opened" and negative (0) indicates "not opened" in the next 12 months.
VALUE DELIVERED: HOW WE HELPED?
Increased Account Open Rate by 15%
We build a machine learning powered PTO model to accurately identify the audience with high probability of opening a new card, along with excluding customers less likely to open a card. This strategy led to a remarkable 15% increase in new accounts.
Improved Customer Acquisition
Leveraging the predictive strategy to select the right target audience increased the likelihood of customers opening a card account by 1.53x times compared to an untargeted approach.
Future Focused Campaigns
The success of the model has paved the way for future applications. It will be instrumental in targeting audiences in upcoming 2024 campaigns.
CONCLUSION
Our client's journey showcases the transformative power of machine learning in the BFSI sector. By overcoming data challenges and leveraging predictive analytics, they not only increased the efficiency of their marketing campaigns but also laid the foundation for future endeavors.
This success story stands as a testament to the impact of innovative technologies in reshaping customer acquisition strategies in the financial services domain.
If you are facing similar problems and looking to apply advanced strategies like this in your business and marketing systems, please don't hesitate to reach out to us at: info@maticsanalytics.com
Comentarios