Retail banks and AI: An easy walkthrough or rough weather ahead?

Published: March 23, 2018

Terms like speech recognition, image recognition, gesture recognition, machine learning, and robotic process automation are gradually entering the traditional banking business lexicon. As a result, the data-intensive banking industry is experiencing pressure to dive into the data pile it's sitting on to uncover meaningful inferences about existing and prospective customers. The above terms are more or less subsets of what we call artificial intelligence (AI).

For banks, especially retail banks, AI can help analyze customer data to find out how they are interrelated; how to consider different combinations of data points; how to interpret data to make meaningful inferences that can serve the customer better; and how to reduce anomalies, improve operational efficiency, and increase business by adding new customers.

AI adoption in retail banking is reaching a more mature state and gradually being adopted by banks worldwide. Royal Bank of Scotland, Swedbank, and Bank of America are some of the early adopters of AI-based technology being extended to serve retail banking customers. However, from the industry standpoint, it marks only the starting point because the adoption of such technology is essentially confined to the tier-1 banks. The greater transformation will occur over the next few years as more and more banks are willing to take up AI-based transformations. While banks experience the transformation journey, it is prudent to highlight some of the key considerations that retail banks may have to evaluate judiciously:

Extent of adoption: AI adoption, of course, requires a lot of investment. Retail banks need to carefully evaluate the extent and areas of AI adoption versus the possible benefits.

Data quality: Banks need to determine if the data they have is worth being applied to AI analysis methods or not. If not, they need to first work toward cleansing and making the data more structured, so they can apply AI methods to analyze them and reap the full benefits of AI.

Feel of the solution offerings: Getting existing customers to adopt AI-based service offerings can be a challenge. Providing only the technology available to customers may not be sufficient. The interaction should be enjoyable and satisfactory to the customers. They should not feel they are only talking to the machine on the other end who does not feel their expressions. Therefore, integration of human emotions to machines intelligence is a challenge.

Personalized offerings: The goal of implementing AI in retail banking is to make the product or solution offerings more personalized for customers. Given the same scenario, different customers may react to the same situation in different ways. The challenge is how granular the personalization can be.

Internal threat of automation: Due to the automation of some of the processes, there may be a situation where some of the human effort may become redundant. Careful planning of such employees to upgrade their skills to some other function will be helpful; otherwise, the whole process of AI adoption will be considered employee unfriendly by the workforce. It will have a negative impact on the entire process.

AI will provide retail banks with a very powerful tool to optimize efficiency, reduce frauds, serve customers more efficiently, solve anomalies in a shorter period, reduce human errors, and improve customer relationships and loyalty. However, retail banks need to identify and implement the best AI technologies that suit the bank's strategy to leverage the maximum of AI's transformative capabilities.

If adopted and implemented in proper ways, AI can change the retail banking scenario dramatically. Absorbing the drastic change for customers and employees is what lies in the hands of banks- able leadership and their strategic decisions.

On the one hand, we debate the success of AI-based transformation, and on the other hand, the banking industry deals with financial inclusion, i.e., bringing the unbanked population into the banking system, in certain geographies. Such aspects of banking reality also do factor in the adoption of AI-based transformation now or in the future for banks.

Even if it is quite evident that AI-based automation will dominate the retail banking industry for years to come, the adoption and implementation of the same with fewer or measured hassles will be the learnings in the future. This will help banks decide to what extent the man versus machine battle should and will go.

Data Quality Checks (DQC) Framework

Unlock cost-friendly and unrestricted data quality checking

Related content