It is no secret that artificial intelligence (AI) is becoming a firm part of our everyday lives but not everyone is aware of how it is making its way to our bank accounts. Computers are becoming ever more powerful and with that, the opportunity for financial institutions to better use vast volumes of consumer data is accelerating.
Through this data, banks can analyze historical transactions and behaviors with the goal of predicting the future. Predictive banking is helping minimize costs whilst improving customer experience. In this post, we will loom at the key use cases for predictive analytics in the banking sector.
What is Predictive Banking?
According to Deltec Bank, Bahamas “In simple terms, predictive banking is using historical data can be used to forecast future events and trends.” Programs that rely on innovative artificial intelligence (AI) and its applications like machine learning can process enormous volumes of data quickly thanks to computing advances. For banking, financial institutions can predict what will happen next under current conditions. It is incredibly important to the industry where there can be huge risks and costs need to be minimized.
AI and machine learning specialist Dataiku, have created machine learning models that can analyze raw data like historical transactions to build a predictive framework. Using the software, banks can forecast and find behavioral patterns that lead to better efficiencies and cost optimization.
In 2017, Dataiku worked with BNP Paribas to create machine learning predictive models for fraud detection. All data collected by the bank was made accessible via a central source including transactions, locations, and transfers. Using the different sources, the models were able to accurately predict the likelihood of fraud by analyzing patterns and detecting anomalies.
Nordic Danske Bank uses the Teradata platform to predict fraud. Their existing rules-based engine was only detecting fraud at a 40% accuracy with as many as 1,200 false positives per day. In fact, 95% of cases they investigated turned out not to be fraudulent, costing an excessive amount of resource time.
The models deployed by Teradata were able to predict and identify fraud across multiple channels. It was able to spot anomalies quickly and with human support, after 5 months it performed significantly better than the previous framework in terms of finding false positives.
Perhaps the greatest risk to a financial institution is loan applicants undertaking long-term obligations. Banks will spend a huge amount of time trying to ascertain the right customer profile for them to lend to and avoid costly defaults or even legal action.
Predictive analytics in banking can be crucial in supporting the lending process. Lenddo is utilizing advanced machine learning models to predict the creditworthiness of individuals. Much of their work is focused on emerging markets that lack credit histories or bank accounts.
The Lenddo system looks at the entire digital footprint of applicants which spans over 12,000 attributes from social media to internet browsing, geolocation data, and smartphone usage. All of this data is turned into a credit score that they say is predictive of future risk. In using Lenddo, their partners are accepting over 50% more applications.
Using historical and third-party data, financial institutions can predict the likelihood of credit defaults. For example, in analyzing historic data about their borrowers, banks can predict future behaviors of those matching the same portfolio.
Crest Financial operates a “no credit needed” lease to own company offering microloans up to $5,000 with real-time approval. They have done this using the DataRobot automated machine learning platform. The innovative system predicts default rates through the use of data and is able to make a real-time decision.
Crest can successfully identify customers in high-risk and highly-competitive markets whilst detecting anomalies in customer transactions. All of this information combined, predicts the likelihood of a default for their applicants allowing them to make immediate decisions.
Wells Fargo has a mobile banking app with predictive features that can analyze user account information and provide them with personalized guidance. The insights predict the likely financial performance of customers following different transactions. For example, if a bill is suddenly higher than it has been in previous months, it will notify customers to review it.
If a customer purchases an airline ticket, they may be prompted to set up an account travel plan for their trip. This means the bank can automatically allow transactions from other countries without it being flagged as fraudulent perhaps. Wells Fargo customers are given access to the predictive insights via the mobile application.
Predictive banking can help financial institutions when it comes to the acquisition and retention of customers.
Citibank has adopted a data-led approach to its acquisition strategy. They do this by analyzing large volumes of data and targeting promotional spending via machine learning algorithms. They will figure out the process customers go through as part of the journey. For example, through deep customer segmentation and profiles, the marketing team will identify the right targets via the right channel.
The team at First Tennessee Bank leverages predictive analytics to optimize marketing strategies. From detailed customer segmentation, they send targeted campaigns that have resulted in a 3.1% increase and response rates and a 20% decrease in marketing costs. The algorithms ensure they market to the right people at the right time.
For retention, American Express has created complex predictive models that are able to forecast and then prevent customers from the churn. They achieve this by analyzing past transactions and mapping them against customers who have left previously. If the profiles have similar patterns, the predictive framework alerts the team to start preventative measures.
Predictive analytics offers banks a new and innovative way to manage their customers. Using data in this way is fundamental to success in a digital era, especially where new and exciting start-ups are attempting to revolutionize the sector. In order to remain competitive, traditional banks should make use of their years of data and take advantage of the opportunity that comes with it.
Disclaimer: The author of this text, Robin Trehan, has an Undergraduate degree in Economics, Masters in international business and finance and MBA in electronic business. Trehan is Senior VP at Deltec International www.deltecbank.com. The views, thoughts, and opinions expressed in this text are solely the views of the author, and not necessarily reflecting the views of Deltec International Group, its subsidiaries and/or employees.
About Deltec Bank
Headquartered in The Bahamas, Deltec is an independent financial services group that delivers bespoke solutions to meet clients’ unique needs. The Deltec group of companies includes Deltec Bank & Trust Limited, Deltec Fund Services Limited, and Deltec Investment Advisers Limited, Deltec Securities Ltd. and Long Cay Captive Management.