retail banking

Delivering Better Recommendations at a Fortune 50 U.S. Retail Bank

  • The result is a mobile app, as well as a number of administrator screens, so that personalized offerings are pushed towards the credit card holders.
  • Ability to track redemption rates across all merchants.
  • 50% – 100% lift in redemption rates.
Mobile App


This Fortune 50 US Retail bank has 10 million mobile users (running the bank’s mobile app on their smartphones). The bank has agreements with more than 10,000 merchants and wants to push coupons and discount offers to their credit card holders, depending on the user’s profile and location. The bank’s main goal was to improve coupon redemption rate through real-time, location-based personalized offers.

NGDATA partnered with the bank with its IEP to be used for its storage, search, user profiling and recommendation capabilities:

  • All data was uploaded into the solution utilizing the IEP’s ETL connectors and Data Loader.
  • The data was then stored in the consumer database, running on top of the IEP’s Big Data Repository.
  • Two of the IEP’s Recommendation Engines were deployed, thereby supporting multiple propensity scoring approaches, including: Collaborative Filtering and Knowledge-Based
  • The Recommendation Engine(s) calculates, in real-time, the customer propensity for all merchants created in the solution.
  • The IEP’s API’s were used to integrate with the bank’s mobile app running on the smartphone and the offer creation/management solution.
  • Finally, the IEP’s connectors to SAS and R were used to offer deep-dive analytical features.


 “Introducing big data and machine learning not only resulted in higher performance, but it allows us to introduce disruptive business concepts and opportunities.”
—Senior Vice President of the Bank 


The bank didn’t want to rely on their consumers to identify their preferences but wanted to develop Customer DNA profiles based on information available inside and outside the bank. The information identified was located in various source systems inside and outside the bank. The internal information included credit card transactions, debit card transactions, merchant information and logs of the mobile devices, and the external information included social data from a variety of sources.

All recommendations needed to be available in real-time, in order to present the right offers at the right time and location. For that number of users, and with the assumption that for each user 500KB of data per month will be logged, the amount of data was too high for conventional systems. Additionally, gathering data and recommendations was preferred to be as close to each other (and not in different systems or technologies).

The result is a mobile app, as well as a number of administrator screens, so that personalized offerings are pushed towards the credit card holders. Using NGDATA’s IEP, the bank is able to realize a LIFT between 50% – 100%, resulting in significantly increased redemption rates.

Next client story
Learn How Innogy Reduced Contact Center Calls by 13%