In April 2015, HDFC started using algorithms to disburse retail loans. “This will change the way in which retail lending is being done in the country,” said the bank’s executive. We might have come a long way in implementing technologies to help the banking industry, but we’ve hardly turned a corner when it comes to harnessing the true value of algorithms, the premise being discussed here.
Algorithmic banking will take the use of algorithms in retail banking to the next level. Simply put, algorithmic banking is the use of complex mathematical algorithms to drive better business decision making. But you’ve already heard this, right? Not quite. When you encapsulate big data and advanced predictive analytics with process-level automation, you drive competitive differentiation and create a niche for yourself. This helps deliver an efficient, flexible and unified customer experience orchestration.
If you’ve read our previous blog on Banking and IoT, we’ve discussed the need for retail banks and smart technologies to exist in unison. Smart machines, the emerging “super class” of technologies, bring in smart data discovery. The opportunities defined from these large troves of data is the starting point towards a comprehensive analytical decision making, which is where algorithmic banking comes into the equation. Its true impact though is yet to be felt.
Algorithmic banking finds a basic and primary use in offer management, by utilizing customer-specific knowledge to customize and deliver real-time offers. Integrating IoT with banking hugely increases data volume, and deep level algorithms are needed to not only find out what might the customers want next but also how effectively can this ‘want’ be fulfilled. This also opens up avenues for cross-sell promotions and product promotions.
One other important use of algorithmic banking would be in pricing optimization. This will help in targeting the best possible customer segments based on pricing strategies, and ensuring banking personnel optimize sales in a multichannel environment. Additional functionality like a price guidance system, to help the sales team, will also help in increasing margins and revenue.
Investment banks and hedge funds already use algorithmic trading using automated pre-programmed trading instructions, accounting for variables such as time, price, and volume, to execute large orders. This has not only reduced human intervention and brought down costs, but has also increased the risks on lightning fast decision making in banking.
The retail industry is one where the use and implementation of algorithms is expected to be much quicker as compared to other industries. Data-driven retailers capture new information about their shoppers every day and can devise new ways to use that data to complement the customer experience.
Algorithmic banking, following on from the synergy between IoT and banking, will transform retail banks into dynamic and digital entities and provide real-time micro analysis. For example, customer-facing bank branches and call-centers will embrace algorithms to enhance their customer support activities. Just imagine being able to know what the customer wants even before the customer asks for it.
Although the current usage and adoption of full-fledged algorithmic banking is practically non-existent, it is predicted to transform the banking foundation in the next five to ten years. This will require humans and smart machines to work alongside each other in the initial stages. But, expect these smart machines to replace humans, bring in a high level of predictability and effectively track customer behaviors.
Much of what has been discussed above about algorithmic banking significantly involves IoT. A sense of perspective and caution needs to be attached to where all can algorithms lead the banking industry to.
Effective banking involves providing the best experience to the customers. Customer’s response, on the other hand, is ever-changing and dependent on a host of factors and variables. Algorithms can use only a handful of such variables, with heavy dependence on them, and often the contextual information that really matters, isn’t considered.
Algorithmic banking can seem to border on privacy issues and may even backfire if the customer’s trust is risked in any way. It may also lead to complacency issues by quantifying large data, and hence leading bankers to believe that they know everything about the customer. A false sense of ‘know-all’ can be harmful in the long run and deter customers.
Algorithmic banking will change the way we see banks. The influx of smart machines, smart learning and smart data will eventually lead to smarter banking. This will involve the deepest-level of analytics and use of the most sophisticated algorithms to come with information that will not only solve customer’s banking problems but, in a way, delight the customers. But retail banks, already grappling with IoT and its intended effect, need to take a step back and visualize the full potential of algorithmic banking before trying to implement it. The customer experience transformation is some way off, but the introduction of IoT in banking is just the start.