Our customers’ time is important to them and giving them meaningful content quickly builds trust and stronger relationships. Offering our customers access to the most pertinent products based on what matters to them delivers on our brand promise of being their trusted financial partner.
A cross-functional team at Discover Financial Services, led by the Digital Analytics team and including members from channel marketing, product owners, and technologists, embarked on a four-year journey to bring relevant, timelier offers to customers under the Personalization program. While the company uses advanced analytics and tools, at the core of everything is a human-centered approach to determine which offers the customers see at a given time.
This blog highlights a solution that this team built to improve customized offers for Discover customers.
A comprehensive solution to serve customers better
Since Discover is a digital bank, customers engage with the bank through its website and mobile applications. Therefore, the digital team at Discover needed to make faster and more accurate decisions about which offers to show customers on those virtual touchpoints. The team approached the problem with the premise that understanding and using the data appropriately would help them improve the selection of offers presented.
Data engineers and data scientists partnered to build a custom, secure platform behind Discover’s firewalls to collect and curate the necessary user data. A large ecosystem of mathematical models was built on this data to understand customer intent. Finally, a real-time machine learning engine uses these models and data to serve customized offers to customers as they visit Discover’s website and mobile applications.
The key to the success of this program was to merge the right technology with the correct analytical engines and algorithms in a cohesive way to serve each customer more effectively.
Making marketing decisions based on data produces great results
Aside from the data-driven, machine learning engine, the team also developed a solid test-and-learn platform that enables continuous improvements in the customer experience. Over the years, the test-and-learn process has reached a high level of maturity, to the extent that tests are always on with a backlog that is implemented in an agile manner.
This focus on testing has created millions of dollars in value, an expansion of the Personalization program to the end-to-end customer journey, and very nuanced optimizations of the customer experience. For example, there are certain products that are significant decisions for a customer, such as choosing to take a loan, and certain products that simply increase customer delight, such as choosing a card design of your liking. Balancing these experiences with the customer in mind while driving value for Discover is what the Personalization program strives to do.
Over the past four years, the Personalization program has quadrupled its profitability run-rate and there is a significant runway remaining to unlock incremental value. Since the entire program is built on a core of data-driven-decisioning for offers, the team can precisely measure the total value of the conversions that the program generates and the incremental profit that is a direct result of the personalized offers.
The team has steadily maintained about a 20% incremental profit rate throughout the life of the program.
Customer-centric focus is key to success
In today's world, the data and technology used to inform offer decisions are a commodity that is common to most firms with a mature data science practice.
What's the secret to success for Discover? The program’s commitment is to bring value to the customer. With a customer-centric focus, the teams felt empowered to make the program the best it could be for the customers.
The people who built this engine have been with this program from day one and never lost focus. They stayed, grew the program consistently, and followed the same guiding principles over the years that were set in place on day one.
This commitment to the product and to its customers resulted in customers seeing the best offers to help them build a brighter financial future.
Want to learn more about this project? Read our blog post about the 7 steps we took to build our personalization engine.
