Consulting firm McKinsey has issued a report on AI application in banks, Frank Media reports.

The report notes that the ongoing AI application enables the banks to serve more customers, expand market share, and increase revenue at lower cost. Deployed at scale, these technologies can give the bank a decisive competitive edge, also in global markets.

Banks that use AI have potential to increase value in four ways: strengthen customer acquisition, increase customer lifetime value, lower operating costs, and lower credit risk.

This is especially important in the competition for customers and in when entering large markets, where banks are forced to compete with tech companies and ecosystems.

Customer acquisition

The use of advanced analytics is crucial to the design of journeys for new customers, who may follow a variety of paths: some may head directly to the bank’s website, mobile app, branch kiosk, or ATM, others may arrive indirectly through a partner’s website or by clicking on an ad. Many banks already use analytical tools to understand each new customer’s path to the bank, so they get an accurate view of the customer’s context and direction of movement, which enables them to deliver highly personalized offers directly on the landing page.

AI technologies allow banks to abandon mass messaging. The bank can select customers according to their responsiveness to prior messaging - also known as their “propensity to buy” - and can identify the best channel for each type of message, according to the time of day. Improvement of the quality of service affects the conversion.

Crediting

Traditional banks’ customers may wait anywhere from a day to a week for credit approval, McKinsey analysts write. AAI banks have developed lending schemes based on real-time analysis of customer data, which allows them to make loan decisions much faster.

The use of artificial intelligence provides several more advantages when issuing loans:

- Limit assessment. Leading banks are also using AA/ML models to automate the process for determining the maximum amount a customer may borrow. These loan-approval systems, by leveraging optical character recognition (OCR) to extract data from conventional data sources such as bank statements, tax returns, and utilities invoices, can quickly assess a customer’s disposable income and capacity to make regular loan payments. Therefore, banks can quickly assess the client’s income and their ability to make regular loan payments.

- Pricing. Banks generally have offered highly standardized rates on loans, with sales representatives and relationship managers having some discretion to adjust rates within certain thresholds. However, fierce competition on loan pricing, particularly for borrowers with a strong risk score, places banks using traditional approaches at a considerable disadvantage against AI-and-analytics leaders. Fortified with highly accurate machine-learning models for risk scoring and loan pricing, AI-first banks have been able to offer competitive rates while keeping their risk costs low.

- Fraud management. Concentration of credit relationships in digital channels opens new opportunities for fraud. The costliest instances of fraud typically fall into one of five categories: identity theft, employee fraud, third-party or partner fraud, customer fraud, and payment fraud (including money laundering and sanctions violations). Banks should continuously update their fraud detection and prevention models. Ping An, for example, uses an image analytics model to recognize 54 involuntary microexpressions that occur before the brain has a chance to control facial movements. In general, better identification of suspicious customers will allow banks to increase loan approval rates without increasing credit risk.

Increasing customer engagement

Leading banks use advanced analytics to identify the least engaged customers, who might leave the bank, and find a way to keep them. Every personalized offer must be delivered through the right channel at the right time.

For example, by analyzing the structure of expenses and search queries of the client, the bank can recognize a need for a loan for the purchase of household appliances. Analysis of the data on the use of bank products can also identify areas, in which the bank can make a better offer to the client in accordance with the client’s current need.

The context

Artificial intelligence technologies are also in demand in Russian banks. In 2021, this technology might be one of the main areas, in which banks will invest actively. “Banks are becoming conglomerates of huge amounts of information, and now we see tough sharp competition in the banking segment. Banks need an effective product that will allow them to further reduce costs and increase profits,” said Sergei Voronenko, director of S&P Financial Institutions.

Large Russian banks have been using AI for a long time. Back in 2019, First Deputy Chairman of the Board of Sberbank, Alexander Vedyakhin, said that there was not a single unit left in the bank, which did not feature artificial intelligence technologies. Back then, the bank actively used AI to make loan decisions. “Now, on the basis of AI-made decisions, we issue 100% of credit cards, more than 90% of consumer loans and over 50% of mortgage loans, since it is a more complex product,” he said.

The partner of Fintech section is