There is one prevalent trend sweeping the entire business spectrum: everyone is talking about big data, data analytics and machine intelligence. Many organisations are now launching a big data initiative and forming their own data science units. Data scientist and data engineering roles have become among the hottest jobs on the market, attracting the brightest minds from academia and industry. In the US, a data scientist is the most sought-after position, according to a popular company review site Glassdoor. Yet many of you must still be wondering what’s driving this data phenomenon.
Data analytics is not new. All businesses, banking included, have been using data to provide insight and improve productivity for decades. We are familiar with marketing officers looking at a graph for sales projections or customers receiving phone calls from service representatives for a post-sale satisfaction survey. However, what is new for data analytics is its increased capability and potential impact to business, economy, and society. Two factors are seen as transforming data analytics. The first is the advent of big data technology that allows computers to capture, transfer, and store data on a massive scale. The second factor is an unprecedented advancement of machine learning technology, now enabling computers to digest and find patterns in data. Machines can now be used to help us find far more complex patterns and insights from data at a much higher speed, scale and accuracy than what we have ever seen before. In many cases, a computer can learn the insights from data and choose to perform appropriate actions automatically, without any human intervention at all.
Supercharged by big data and machine learning technologies, data analytics can be leveraged in many business areas including banking. One example is customer feedback acquisition and analysis. Instead of human operators gathering and interpreting customer survey results manually, computers can now automatically scan a collection of customer survey data and use natural language understanding technology to summarise overall levels of customers satisfaction, as well as all pending issues that need corrections.
Another example is personalised marketing. Computers can now understand each user’s taste, preference, and lifestyle based on the customer’s purchase history, and hence can offer the right product to the right user – at the right time – through the right channel. The same approach applies to classical operation management. Optimisation problems such as service branch allocation, task scheduling, and logistic planning can now be solved much more quickly and accurately with big data analytics and machine learning.
On the human resource management and development side, computers can now analyse each employee’s data to assess performance, suggest skills improvement, and assign that person the right role.
Aside from generic uses that are applicable to any business, big data and machine learning may be applied to a few more areas specific to banking and finance. Automated customer identification and authentication is one area that can improve banking customer experience. When a customer walks into a bank, a camera can learn to identify the user using intelligent facial recognition technology so that the teller can automatically retrieve the user’s account details and predict the user’s intent in advance. Financial fraud activities can also be tagged more precisely and handled more quickly, thanks to advanced anomaly detection on usage behaviour.
For banking customers with a positive account balance, a bank can employ what is called a “robo-advisor” – a machine intelligence technique to automatically provide personalised investment suggestions that ts the customer’s assets, financial goals and approach to risk. For banking customers who need credit, computers can also help a bank assess a customer’s credit score more accurately by detecting complex, underlying patterns in customer data that inherently affects the customer’s ability and willingness to repay the debt, which leads to a more efficient lending system that minimises non-performing loan rate and increases financial productivity at a macro scale.
Here at KASIKORNBANK, we use big data and machine learning techniques to automatically identify customers in need of credit and then offer appropriate loan products with customised pricing based on each customer’s credit score. This personalised loan product pricing and offering has proved to be more successful when compared to the traditional method.
Data disruption is here to stay
The cases discussed within this article are just the tip of the iceberg. While such applications have already been implemented or are currently being implemented by several financial institutions, there is potentially much more banks can achieve with big data and machine learning. This is a global phenomenon across countries and industries. IDC, a global leader in industry and marketing research, reported that widespread adoption of cognitive systems and AI across a broad range of industries will drive worldwide revenues from nearly $8bn in 2016, to more than $47bn in 2020 with banking named as one of the top two industries to lead this adoption. Banks have no choice after all.
Data disruption, like all other technological disruptions, is relentless. It will sweep aside those standing in its path, and reward those who embrace it for the bene ts of their customers. Finance is no exception. It is not a question of “if” but “when” an industry will be disrupted by data. It is not a question of “how” but “who” will emerge as winners from this disruption. One thing is clear. The data revolution is here.