Authors’ note: This is the third of a five-part series, in which we discuss the ways in which banks are adopting robotic process automation (RPA) today. Part 1 offered a list of seven deadly sins that can derail RPA adoption. Part 2 identified seven virtuous steps banks can take to avoid these deployment pitfalls. In this installment, we look at the RPA areas banks are exploring to gain maximum return on investment.
Given the challenges entailed by digital change, many banks are looking for quick and inexpensive alternatives that can deliver short-term wins while major initiatives are underway. RPA can help bridge this gap as a relatively inexpensive and even temporary alternative, both for customer-facing and middle- and back-office functions.
We see RPA helping banks engage customers in real time, increase efficiency and productivity, and tap into the benefits of straight-through processing without a significant overhaul of their underlying systems. In the past year, we’ve seen RPA applied most often in the following areas:
- Risk and compliance reporting: This represents an easy win for many banks given the number of different applications that need to be accessed to provide the required data for reporting. For example, we’re working with banks to create fiduciary risk management reporting capabilities, which involves going through multiple e-mail systems, external websites and broker statements to generate reports and highlight anomalies. Use of RPA has automated 90% of these processes, saving significant costs and time.
- Anti-money laundering (AML) and know your customer (KYC): Both AML and KYC are rules- and data-intensive processes, making them good candidates for RPA. We’re working with several banks to automate manual processes for inspecting suspicious transactions reported by AML systems. In these instances, RPA has proved a better option than pure-play business process management solutions in terms of cost and time to implement. In the KYC area, we’ve seen that while banks digitize the entire KYC process end-to-end, which requires changes in the systems involved, they’ve implemented an RPA solution to quickly bridge the integration gap between disparate systems that need be accessed. In addition, both KYC and AML have significant potential to leverage advanced artificial intelligence (AI) technologies in the form of machine learning to enhance the outcome of these functions, such as reduction of false positives, proactive fraud prevention and digitizing the end-to-end KYC process.
- Accounting: Several accounting and reporting processes involve repetitive daily activity that requires capturing data from multiple systems. These are well-suited for rules-based RPA. We’ve worked with banks to apply RPA to a number of accounting proceses, from relatively simple financial and management reporting, to more specialized functions such as automating a bank’s Fair Valuation of Assets Reporting process. In these cases, we’ve used a calculation of net asset values for investment funds and created efficencies in the product control function of investment banks.
- Mortgages: This is another area that’s ripe for significant distruption and transformation through automation and digital transformation. Given the number of third-party entities in the mortgage value chain, the significant use of paper (any proud owner of a mortgage can relate to this!) and the fragmented nature of the systems means RPA can play a key role in providing efficiencies while the industry undertakes a wider transformation. We’re working with multiple lenders in the area of loan origination and servicing, including document preparation, valuations, escrow, underwriting, foreclosure and tax management.
- Reconciliations: Even though reconciliation is already largely automated in most cases, the process to investigate and resolve reconcillation breaks is manual. For several clients, we’ve used RPA with predictive algorithms to reduce exceptions and automate the resolution process.
- Front office: Front-office and contact center staff often need to access multiple applications to work with customers. We’re working with several banks to use RPA to bring all relevent information from multiple systems to one screen for support staff to provide effective service.
- Other areas: We’re working on almost all areas in banking that have shown direct benefits through automation. These areas include cards and payments (card issuing, customer onboarding, card servicing) and asset and wealth management (AI-driven advisory, customer onboarding and helping financial advisors through automated research and market insights).
Many global banks started the automation journey early on and are now streamlining their RPA initiatives, while many regional banks are still testing tools and trying out RPA projects in specific areas. We see the following approaches being taken:
- Proof of concept (PoC): Almost all banks that began their automation journey in the last two years started with small PoC projects. In most cases, choosing the correct PoC had a bearing on how quickly the RPA initiative could be scaled.
- One line of business (LoB) at a time: Some banks start in a single LoB that exhibits high operational costs but simple processes. This approach can help organizations prove benefits quickly and manage change effectively.
- From simple to complex processes: Others have identified a small set of simple, rules-based processes that are common in multiple LoBs and that can be automated, and then measured their success across the organization. This approach has advantages over the previous two, but it needs enterprise support upfront and takes longer to plan.
- Biggest impact area first: Several banks have begun by identifying a single process that has a bigger impact when automated in terms of cost savings or efficiency. This approach has worked best when a bank wants to prove the real impact of automation and create a funding model for strategic initiatives through cost savings gained from automation.
With so many diverse and innovative uses of RPA in use today and planned for the future, it’s tempting to rush in, identify a PoC opportunity and start automating processes. Yet a more measured approach is advantageous, especially if it’s considered in the broader context of digital business.
In Part 4, we explore what banks should be thinking about when moving from proof of concept-based, siloed, task-level automation to overall operations transformation and how platform vendors are working to support these efforts.
Makarand Pande, a Digital Partner at Cognizant, contributed to this blog.
To learn more about how RPA is playing and intelligently evolving in the banking and financial services space, please read our report “Financial Services Automation: Taking Off the Training Wheels.”