As a leader of the bank, you now have two options:
- The first is to put big data into the liability or even the risk bucket, and fight it with primitive tools such as costly, rigid and complex core banking and RDBMS systems.
- The second is to embrace big data and come out in front of the regulatory-driven compliance tsunami.
Embracing big data will empower the front office staff with the controls they need to make decisions at the point of the transaction and, at the same time, eradicate the complexity fueled by silo-based point solutions.
What is your choice?
To date, 90 percent of your colleagues are choosing option one and only dipping their toes into option two out of plain fear. The latest issue of Intelligence Quarterly focuses on the 10 percent who are boldly embracing option two.
How are they doing it? Many are taking a factory approach to big data and analytics, constantly trying out new ideas in a big data lab, and then taking what works and repeating it with precision. In particular, the factory approach processes information and turns the raw material (data) into something useful.
More often than not, banking executives will point out that they have big data projects started, and they are working to cut through the complexity. With very few exceptions, however, they struggle to embrace the digital age and attempt to solve growing data-intensive problems with yesterday’s tools and approaches.
When it comes to big data, who is in charge?
Any given person, department or function alone cannot change the bank. To really change the bank, it will take the efforts of IT, risk, retail and other departments all working together.
IT does automation and infrastructure, not optimization, nor does it industrialize the model management. It’s not unusual to find dozens of custom-built solutions with millions of incompatible rules and hundreds of copies of the data floating around in the bank. Combine the short-term focus with these silo-based point solutions, and we have a picture of true complexity.
Find out how banks are moving away from this level of complexity and shifting to a strategic state. Part of this process involves moving away from vendor consolidation and asking vendors to take on more by providing software as a service or even as an appliance. Instead of asking the vendors that contributed to the problems in the first place to do more of the same, banking leaders are asking: What will it take for you to replicate results to larger parts of our increasingly complex value chain?
The way that banks look at IT is changing. Indeed, a major banker I respect recently asked, ”If I spend $300 million annually on AML alone, is it not reasonable to think that I should be receiving AML as an appliance?”
How can the risk department help lead the change? Throwing manpower at the problem will not break the growing wave of regulatory-driven compliance. Instead, we must let go of our investigative warehouses and the idea of offshore back offices. For an example, read how the chief model risk officer of Discover Financial Services embraced an analytical factory concept to cut through CCAR requirements with little effort. Others banks are using the analytical factory approach to empower front office staff to take ownership and responsibility of decisions at the point of the transaction.
Most heads of retail banking also are automating their customer interactions instead of optimizing the customer experience across touch points and channels. Why embrace hyperfragmentation through rigid, channel-based structures and systems when you could drive consistency across channels? Instead, you could use analytics to calculate risk-adjusted performance per client while creating personalized experiences and taking the costs out of the system.
Cut through the complexity
The best way to improve client performance while personalizing services to each client is to embrace the technology required to cut through the complexity. Unlike the rules-based technology of the past, ownership of the analytical factory calls for a new set of skills, be it social, demographic, economic or any other faculty conceivable. For example, SAS is working with one global, systemically important bank (G-SIB) that hired a team of astronauts to help find the extreme outliers in financial investigation data.
My favorite case study is about a UK bank that empowered the front office staff to make credit decisions. It asked, why not equip the customer-facing banker to make the credit decision that can be better for both the client and the bank? It makes sense, especially compared to the alternative of reducing the time it takes the risk department to process an internal request.
Why, then, are other banks not copying this approach? My guess: They applied the wrong skills to the job. The answer to too many databases isn’t to create another and call it an enterprise warehouse. The answer is a factory approach, as one insurance firm discovered when it separated risk models from the data, through the creation of a model factory. What do you know about your customer? Probably much less than this major insurer, since they now model customer behavior and analyze client perceptions to improve the customer experience and measure risk-weighted performance at a reduced cost.
Who should lead the digital transformation? It touches all aspects of the bank and needs a true champion at the top. The CEO should personally lead the way and change the bank — because the future of the bank is at stake.
And the best way for the CEO to lead the bank into the digital age is with a factory approach that automates, scales and manages data to support collaboration between departments and streamline analytics projects throughout the bank.