How RPA and IPA Create Value For A Bank’s Digital Transformation

With the advent of mechanised software-based automation and artificial intelligence, the stream of Robotic Process Automation (RPA) found its increasing adoption in the banking industry.  RPA found its early adoption specifically in the areas of automating the mundane back-office operations in banking. Some of the earliest adopters of RPA in banking have experienced a significant boost in efficiency of their operations execution, have lowered cost of operations and have gained better reliability. Pandemic helped push the automation in back-office operations with greater speeds than ever before. RPA includes installing/ deploying a software robot i.e. a program equipped to take a certain set of actions when specific business scenarios occur, respectively. In this article we’ll explore why RPA is useful, it’s use cases in banking and more about its new improved avatar, Intelligent Process Automation (IPA).

RPA: For Automation by Design in Banking Back-Office Operations

While most digital transformation projects focus mainly on automating technology applications such as machine learning, cloud, social integration etc. , some of the core supporting workflows are often left alone to be done manually. This is the reason why business stakeholders often experience that digital transformation projects are leaving behind at least 2 to 3 internal supporting processes to be handled manually even after the project is live. Most often the back-office processes keep getting ignored from the larger scope of digital transformation projects at banks and financial institutions. In order for digital transformation to take its full effect, the automation solutions need to be thought through by design, including the back-office processes. To automate the back-office operations effectively, there’s a need for a solution to be both taking actions automatically and intelligently. Robotic Process Automation brings together the blend of automation with Artificial Intelligence (AI). While automation is about triggering actions spontaneously, intelligence in the context of RPA could be as simple as a set of rules that define which action to take when a certain scenario occurs.

Use Cases of RPA in Retail Banking

There are a plethora of use cases within retail banking where RPA has demonstrated boosting efficiency and productivity while lowering the cost and increasing reliability of operations. RPA in its true sense is a platform for innovation by banking and technology stakeholders. The list of use-cases where it is applied is only growing as the banking community continues to innovate with RPA. Here is a list of a few significant use cases that yielded instant returns on RPA investment.

Backend Check-list Verification

Check-list verification is a commonly used method to undertake quality assurance before initiating any back-end actions. For example, making a code live in the bank’s core processing system, creating a customer ID while onboarding a customer, archiving a customer account or initiating a new card print request, and so on. All these activities have an associated checklist of necessary and sufficient conditions to be checked before initiating them. Some of the checklists have hundreds of conditions to be verified from dozens of systems and this is cumbersome and manually an overwhelming task. An RPA program enables bank applications to swiftly go through the relevant checklists to find whether all the conditions are met or not, without any manual intervention. RPA is capable of choosing to take different actions, sending alerts to relevant stakeholders or systems, based on the checks that failed or are incomplete.

Central Bank Reporting and Alert Mechanism

Central bank reporting requirements keep changing frequently. Reporting is a combination of data readiness and formatting. For dozens of different reports including daily, monthly, quarterly, semi-annually, and annual reports, there’s enormous MIS man-power required to work round the clock to get the data ready and process it in the format required by the central banks. MIS teams have to follow-up with respective teams to get the data, if the data isn’t ready. Banks’ MIS teams successfully complete this mammoth task most often, only to find the reporting requirements changing the next day. Some of the central banks such as Monetary Authority of Singapore recently mandated the banks to only send the stream of data through an API call, and not send the reports in any specific format. All these changes mean that the automation of reporting activities is more effective for banks as compared to investing in manual efforts. RPA helps automated alerts, follow-ups for data completeness and building reports in the formats in which the data needs to be prepared.

Customer Onboarding

There are many steps in a banking customer onboarding process. These may include reading data from customer applications, customer data upload, multi-party data verification, notifying customers at each stage of the onboarding process, highlighting issues, interventions etc. All these steps done by different departments in the authority induce a delay in onboarding the customer in the banking system, resulting in a deteriorated customer experience. RPA helps relationship managers to automate most of the steps involved in the onboarding process by automating the workflow and taking steps as per scenarios. RPA has helped deliver a great customer experience, while reducing the overall time to onboard customers in the banking system. eKYC that involves automated scanning, verification and uploading of customer identification documents has been helping banks to facilitate digital only bank accounts.

Loan Processing

Underwriting is an important part of loan processing, many aspects of underwriting are repetitive and can be effectively done through an RPA. This includes reading through the stack of supporting documents presented by the applicant, verifying the credit history of the applicant etc. While there is a growing competition within banks and NBFCs to offer quicker loans, RPA is a useful tool to help automate parts of the loan processing system. The value of using RPA here is in delighting customers with better responsiveness during the loan processing.

Customer Service

80% of the load on customer service is due to the most common requests which are repetitive in nature, while 20% load is contributed by more challenging and complex requests. RPA helps in addressing the most common customer complaints and requests received from customers over emails, phone calls and chatbots. RPA is capable of handling most common customer service requests while leaving behind more complex ones to be handled by the back-end team. This helps the team to focus on the requests that need manual intervention and helps customers experience a prompt service.

AML Transaction Monitoring

Transaction monitoring is one of the important aspects of any AML application. There are hundreds of parameters pertaining to transaction monitoring such as countries involved, parties involved, amounts involved, pattern of transaction, time of transactions, day of transaction etc. The focal transactions need to be monitored for different permutations and combinations of these parameters as they appear. This is a time sensitive and extremely complex process to be monitored manually.  Most common problems of manual AML monitoring are into highlighting false positives, resulting in a bad customer experience. RPA helps conduct transaction monitoring more effectively and efficiently, reducing overall false positives and increasing the accuracy of AML monitoring. Oracle’s OFSAA uses this mechanism to monitor transactions for complying with AML regulations.

Intelligent Process Automation

While RPA is a mechanical automation of operations, Intelligent Process Automation (IPA), its more recent sibling, is steadily gaining traction within banking. IPA focuses on continuous learning and intelligence building to decide which actions to take. Intelligent Process Automation employs machine learning as opposed to using a set of rules to learn and identify best actions to be taken in a particular scenario. As banks move to the holistic automation from back-office to customer-centric front-end operations, more diverse scenarios are emerging. With diverse scenarios coming up, ascertaining a set of rules for all the scenarios is humanly impossible. In such workflows, intelligent process automation is far more valuable as compared to RPA. For example, some of the conversational chatbots that help bank’s customers on the website either in their buying journey or in servicing their service requests are using machine learning driven IPA. The IPA based chatbots are capable of triggering actions within the backend processing systems to authenticate users, pull out required information, and process requests such as cheque book request, statement request etc.


Whether banks employ RPA, IPA or a hybrid of the two, the industry is gearing to move into holistic automation for its back-end and front-end operations. Ability to provide amazing customer experience at scale, and reliable, high quality service delivery while saving avoidable operational expenses are the key value creation areas for process automation technologies. CIOs, CTOs and CDOs at the banks globally now appreciate the fact that long term digital transformation is here to happen by automating one process at a time. Frequent and constantly changing nature of process automation requires the banks to onboard a workflow automation platform that helps banking stakeholders to configure, re-configure workflows and processes themselves with little or no help from their technology partner. The platform approach to automation makes it a fertile ground for business process innovations.