Expedite regulatory compliance requirements and dispute resolution in the Banking Ecosystem with Robotic Process Automation (RPA)


The banking, financial services, and insurance (BFSI) sector is a multi-entity ecosystem. Hence they have to strictly follow regulatory compliances defined by authorities from time to time. They have to adhere to AML measures and resolve disputes in a highly dynamic environment. As a result, the need for speed and a real-time connection between the entities involved becomes mandatory. Robotic Process Automation (RPA) offers that much-needed speed, accuracy, and transparency and the nimbleness to quickly integrate the regulatory requirements in the business as usual workflows of the BFSI sector. It builds seamless pathways to optimize T+3 to near real-time processing. The technology also allows seamless Forensic Investigations in resolving cases related to ATM disputes.



Banking and financial services (BFS) and insurance (I) sectors have to tread a tight rope while managing customer service, risk, fraud, security, and dispute resolution – all at the same time. Federally-driven regulatory compliance measures offer a definite framework to banks, financial service (FS), and insurance institutions to perform this exemplary feat.


BFSI is fraught with innumerable challenges. Constantly changing global and domestic regulatory requirements, digitization of assets to ensure long term storage, classification of these digital assets for seamless storage, search, and access, continuous monitoring of customer transactions and their profiles for anti-money laundering (AML) compliance, know your customer (KYC) and ReKYC compliances, etc. Financial inclusion of the non-banking marginalized communities through prepaid cash cards and similar routes along with their KYC, batch refilling of the prepaid cards, extending the e-wallet paradigm to wearable payment technology and the related KYC. Resolution & management of disputes arising through interbank ATM and clearing house transactions, etc. The list goes on.


Managing the BFSI juggernaut involves extensive collaboration between third parties, such as government bodies, statutory bodies, federal agencies, and peer banks. The dynamic nature of the BFSI sector requires seamless and real-time connection with these entities. At the same time simplifying operations and not complicating them further becomes a must. Robotic Process Automation (RPA) allows to bring in the much needed granularity, abstraction, and hence simplification. Yet it is important to note that AUTOMATION is just the VEHICLE TO ACHIEVE THESE OBJECTIVES and not the ULTIMATE GOAL.


RPA is a non-invasive process. It is cost-efficient, scalable, and easier as compared to API integrations. It comprises surface level or user interface level integration. It thus allows high speed, two-way communication between the modern agile enterprise applications and the legacy systems within the distributed BFSI ecosystem in a highly secure framework. RPA automation offers multiple use cases for high velocity enterprises. Banking solution architects and domain experts from the System Integrators (SI) community, work closely with the stakeholders to understand the procedures and conduct the RPA deployment. The RPA integration is used for simple to medium complexity processes, which are repetitive and rule-based.


RPA automation is further leveraged for multi-layer, high complexity processes by using artificial intelligence / machine learning (AI / ML) algorithms. In popular terms, this leverage is called Intelligent Automation or Cognitive Automation. The pre-programmed AI / ML algorithms continuously learn through the exception handling procedures, such that they get better and deliver higher value with time. RPA and AI / ML enabled RPA (Intelligent Automation) enable BFSI enterprises to engage on the same innovative and competitive turf as born in the cloud (BIC) FinTechs. These technologies play the role of a change catalyst and enable digital transformation in the BFSI ecosystem comprising entities at varying levels of digital maturity.  RPA and AI / ML expedite processes and impart high speed and velocity to the BFSI enterprises allowing them to seamlessly operate in a wider distributed ecosystem and ensure all statutory compliances. These technologies set the highest standards of operations for the BFSI industry.


Check Truncation System case – Faster cheque clearance

The federal banking governance body in India has implemented a cheque truncation system or CTS for faster processing of inter-bank cheques. It proposed the discontinuation of physical cheques to expedite the cheque clearance. Here an image of the cheque is routed through the clearance house with some vital details extracted with AI / ML enabled RPA.


The adoption of the cutting edge technology offers immense benefits to the banking ecosystem, including improved efficiency, speed, security, cost effectiveness, cost savings along with improved customer service.

Such AI / ML enabled RPA use cases can be leveraged in all domains, which involve delayed paper-form related processing work and yet require the paper-forms to be in place due to statutory requirements. Invoice processing is one such similar use case, which mandates the physical invoices to be in place. Here also, the processing can be expedited with the use of truncated images and related data that is digitally extracted to expedite processing.


New BFSI compliances – Governance reports

As governments implement different norms and mandate their compliances, they need to continuously monitor the implementation and outcomes on almost a daily basis. As in the nation-wide demonetization exercise that was conducted a few years back, the Indian government had to continuously check on certain parameters to make the exercise a success. How much collection was achieved in which bank and branch, what was the denomination, social security number authentication fulfilment, number of regular customers and otherwise, so on and so forth. These numbers were required to be reconciled almost on a periodic basis along with other reports for statutory compliance. AI / ML enabled RPA allows the BFSI entities to be on the top of such tracking and reporting almost in a real-time environment even as the compliance implementation rolls outs and unfolds.


Insurance claim processing case – Fulfillment of SLAs and statutory norms

Insurance companies are mandated to quickly settle the claims. However, with the compliance requirement to evaluate each claim in detail on one hand and process them quickly on the other hand, they face a herculean task almost every time in fulfilling the SLAs. AI / ML enabled RPA allows to quickly extract the data as per pre-defined criteria and expedite the processing of the claims.


This solution can be applied to any use case that involves document analysis and decision, right from eligibility verification, KYC, claim processing, validation with internal / external databases, reducing error / rework, generating internal / external compliance reports, to improving productivity and OpEx reduction.


Due to the AI / ML enablement, the solutions also facilitate pattern mining thus unearthing outliers or fraud cases in a timely manner. The solution thus allows cost savings at multiple levels and hence results in revenue generation.


BFSI processing with AI / ML and RPA over the cloud – Remote Operations enablement

With the global tryst with pandemics and other natural as well as man made calamities, the businesses the world-over are looking at alternatives to conduct remote operations in a sustainable mode. Cloud is popularly used in SaaS and PaaS modes. It not only allows entities to retain their databases on cloud but also host entire digital offices in a SaaS mode to support remote operations. Similarly, the AI / ML enabled RPA can scrape and communicate data from the underlying systems hosted on cloud and automate repetitive processes related to compliance measures. This includes paper form processing by using Intelligent Document Processing, KYC, Video KYC, data processing, claim processing, etc., which can be performed remotely in a seamless manner.


Optimizing the Bill processing T+3 mandate – Near real-time processing

Most financial bills take a longer processing time, which is usually more than 3 days, while redeeming the proceeds. The process gets prolonged unnecessarily due to the requirement of updation of multiple internal and third party systems before the redemptions finally reach the customer. It becomes quite frustrating to all stakeholders and at times results in customer ire. In a post-pandemic economy, as the BFS operations stabilize and grow, the optimization of the T+3 processing mandate towards near real-time can be affected by using AI / ML enabled RPA operations. This effort goes miles in governments’ efforts to reboot, rebuild, and relaunch the post-pandemic economy.


CERSAI case – Mortgage compliance process automation – Third party government site updation


Some bad actors tend to take loans from multiple banks on the same immovable assets. Hence Central Registry of Securitisation Asset Reconstruction and Security Interest of India or CERSAI does a central tracking of all equitable mortgage loans. The Government of India uses this CERSAI mortgage compliance automation as a mandatory check for all retail and institutional mortgage loans. In this process, banks have to share the complete details of borrowers and the property that is used as a mortgage.


At any given point, banks lend out thousands of loans and they need to upload these details on the CERSAI site. Banks extract these details from the core mortgage system and use a bulk upload facility to upload the details on the CERSAI site. Due to formatting issues and data inconsistencies, the CERSAI site throws rejection cases in an Excel spreadsheet format.


The rejected cases tend to be on the higher side and banks need to review, rectify, and validate these details and upload them in multiple tabs and sections on the CERSAI website. This third party site or government site data updation and mortgage compliance is a tedious and error-prone process as it requires referring to data in multiple columns, which are not in a pre-defined sequence, and keying in the data in each tab and submitting it before going to the next tab.


After submission, the CERSAI site generates unique ids for transaction, asset, and security interest, which need to be captured and updated in the Excel file and relevant systems. RPA enables updation and 2-way communication between the systems involved. The process is completed with high accuracy and 200% faster speed as compared to manual operations.


ATM dispute management case – Real-time connect with peer banks and third party systems

Customers withdraw money from ATMs of other banks that are near to their residence or workplace and not particularly from the ATMs of their own bank. At times disputes arise in cases, where the transaction does not deliver the complete amount or does not deliver at all but still registers a delivered notice.


The parent bank has to resolve such issues within a predefined time limit. However, they have to verify whether it is a genuine request and not a fraudulent attempt at money coercion. RPA enables banks to read the data from the multiple systems involved in the ATM transaction and delivery of the monetary amount.


This ATM dispute management process is a type of RPA & AI enabled Forensic Investigation. Here the parent bank has to access data from different systems and set the RPA trigger to fetch the systemic details for that fateful day. This includes the core banking system for the verification of ATM id, debit card id, and the related transaction details. Electronic journals (EJs), system errors, transaction details, exception details, etc., from the other bank’s ATM machine. Reconciliation reports from third parties and vendors, etc.


If the Forensic Investigation reveals fraud, then the RPA & AI sends auto-triggers along with corresponding report and supporting evidence to the stakeholders and video surveillance vendor to fetch zoomed-in video clippings with date and time stamps and then set off the further process. In case it is a genuine issue, then the report and the supporting documents are routed to the credit department to credit a payment to the customer’s bank account.


RPA & AI enabled Forensic investigations related to ATM dispute management, is a fast paced process. It allows banks to resolve ATM dispute cases within a matter of two hours, thus registering a quick turnaround time cycle as compared to semi-automated processes.


Paycheck Protection Program (PPP) USA case – The race against time

As the novel coronavirus triggered a global economic carnage, Governments over the world stepped in with SOPs in a deadline driven format. The Small Business Administration Paycheck Protection Program or SBA PPP or simply PPP program was one such mammoth scale effort initiated in the US.


The PPP program intended to revive the small businesses by doling out loans and sanctions. The small businesses needed to fulfill some documentation in order to enroll into the program and enable banks to take the processing forward.

However, tight deadlines to execute the program made it difficult for the banks to quickly fulfill the statutory documentation and validate the small business criteria due to the avalanche of high volumes of PPP loan applications.


RPA enabled the banks to reduce the repetitive steps in data entry in different systems and reduce process latency. It also enabled the banks to extract the data from different form types, including Excel files, pdfs, e-forms, scanned documents, etc., and transfer the data to the core banking system.


RPA has a capability to execute a process in a 24 x 7 mode to deliver highly accurate output. The high flexibility and scalability of RPA enabled the banks to quickly tailor it to suit their requirement and scale it up to suit the increasing volumes of the applications as the deadline approached.


RPA allowed the banks to quickly process the loan applications as per statutory requirements and make the program a success. The program was delivered on a mammoth scale in US and was executed in conjunction with the banks to re-start the small businesses on a war footing.


Federal Compliance related RPA use cases

RPA is a highly flexible technology and can be customized to suit a wide range of BFSI requirements. It cuts across all divisions and sections in banking, financial services, and insurance. With the continuous evolution of the 5G and LTE spectrums, high velocity banking and insurance automation practices are on the rise.


RPA, AI / ML, and improved network security is giving the BFSI sector an unprecedented confidence and conviction to make digital platforms and transactions a success story. High volume fast data transfers and integrations across digitally connected platforms is offering the global economy a profound thrust.


Some common RPA use cases that ensure fast implementation of regulatory compliance measures include the following:


Trade Finance RPA use cases

  •         DOWAR compliance – RPA allows auto-recovering commissions on bank guarantees and routes them to concerned government departments according to the Domestic Outward Guarantee Auto Renewal process.
  •         Bill of Entry – RPA allows auto-upload of bill of entry for all EXIM cases as per the regulatory mandates. It also allows one time updation of backlogs.
  •         Data Entry & Validation – RPA auto-verifies documents received from the involved parties. It ensures consistency of information, regulatory checks, and AML screening checks.
  •         SWIFT Validation – RPA allows to create auto-alerts for pending transactions related to inward remittances. It also allows integration between core banking systems and the SWIFT alliance.
  •         XLC Advising – RPA allows monitoring of the Trade Finance applications throughout the lifecycle of the transaction to eliminate the risk of entertaining any bad actors through integration with global AML regulatory third parties.
  •         FIBG pending reports – The foreign inward bank guarantees that are pending for more than 15 to 30 days are escalated through regulatory mechanisms with RPA integrations.   

More RPA use cases for Trade Finance>>


Know Your Customer (KYC) RPA use cases

  •         KYC & Re-KYC norms – RPA allows to auto-capture data from different information channels, including helpdesks, internet portals, financial crime regulatory bodies, excel files, etc., and ensure KYC FIT status of its onboarded customers at regular time intervals.
  •         CKYC norms – RPA integrates the banking records with the central KYC portals and reports the acceptance or rejections to the customer.
  •         Customer Due Diligence – RPA allows compliance with increasingly complex KYC mandates and statutory norms by performing detailed customer due diligence across internal and external applications and systems.
  •         KYC cheque return – RPA bots allows auto-closure of non-KYC compliant customer accounts as per the federal and bank level KYC policies.
  •         Remote KYC and Account Opening – RPA facilitates major processes related to remote banking, including remote submission of KYC supporting documents, and auto approve/reject the account opening without the customer having to visit the bank.
  •         Loan Processing KYC – RPA plays a major role during loan origination and KYC by auto-validating details as per the continuously revising regulatory norms to eliminate the risk of non-performing assets.

More RPA use cases for KYC>>


Anti-Money Laundering (AML) RPA use cases

  •         AML screening – RPA allows screening of new customers and account detail modifications by bulk uploading data to AML applications and weeding the data for True positives and False negatives.
  •         AML investigation – RPA allows to retrieve data from multiple internal and third party systems, integrate, and perform data massaging to get the final output.
  •         AML & Fraud Detection – RPA and AI/ML algorithms enable detailed investigation to retrieve true positives and false negatives and then generate Suspicious Transaction Report (STR) for the true matches and file it with federal authorities.

More RPA use cases for AML>>


Retail assets RPA use cases

  •         Salary upload – RPA expedites salary payments by using bulk upload facility to remit monies to corporate prepaid cards or corporate salary accounts.
  •         Charge reversal – RPA facilitates discretionary reversal of charges on different services offered by the bank from time to time.
  •         Remittance procedures – RPA allows to automate the various remittance processes, such as Nostro, Vostro, RTGS, and NEFT.
  •         Electronic wallet – RPA facilitates two-way money transfer between the e-wallet tied to the bank portal and the bank account from any device.
  •         Loan Foreclosure – RPA enables validation of loan foreclosure requests received from the banking system.
  •         Trading account opening – RPA expedites the process of opening Trading, DMAT, Trading & DMAT accounts, etc., to a matter of some minutes as compared to days earlier by using RPA bots.

More RPA use cases for Retail assets>>                


Retail liabilities RPA use cases

  •         ATM non-bank charges – RPA allows to auto-reverse the charges levied on using other bank’s ATM as per the request from customer.
  •         Limit extension – RPA allows to expedite the process of extending the limits for cash credit and overdraft for customers.
  •         RTGS processing – RPA facilitates money transfer between accounts after validation of beneficiary details. It allows auto-rejection and reversal of amount to source in case of invalid transactions.
  •         Tax refund – RPA auto-processes refund of tax deducted at source despite the customer having submitted the required forms.
  •         Credit card processing – RPA allows auto-execution of requests for credit card and updation of limits for existing customers.
  •         Debit card re-issuance – RPA allows to auto-cancel existing debit card and update the information of the newly issued one in the banking systems.

More RPA use cases for Retail liabilities>>


Corporate Banking RPA use cases

  •         Term deposit creation – RPA allows creation of individual term deposit certificates for a corporate customer by using the certificate creation requests and the total amount received.
  •         Audit confirmation – RPA allows banks to retrieve company-wise details from multiple sources, merge the details about a company into a single file for audit.
  •         Dividend data – RPA facilitates auto-downloading dividend data of thousands of customers in account wise manner and then share with the customer post reconciliation.

More RPA use cases for Corporate Banking>>


Wholesale Banking RPA use cases

  •         Service Tax receipt – RPA bots allow to book charges, including service tax, processing charges, etc., on bank products.
  •         Priority sector lending – RPA facilitates shortlisting of priority cases as per pre-defined criteria and update the details in the core banking systems.
  •         Standing instructions – RPA reads standing instructions for NEFT / RTGS and auto-updates the details in the web application for payment.

More RPA use cases for Wholesale Banking>>


Treasury Management RPA use cases


  •         Exchange house balance advising – RPA allows to generate balance reports for foreign exchange houses with respect to different currencies.
  •         Foreign exchange counterparty generation – RPA allows to auto-verify if the counterparty exists in the banking systems for Treasury Management, Core Banking, and statutory listings. Where it does not exist, it auto-creates the counter-party.
  •         Risk exposure report – RPA bots enable auto-creation of risk exposure report for the customers in context of forex, bullion, bonds, derivatives, securities, etc., and auto-mail the report to the authorities.

More RPA use cases for Treasury Management>>


Moving beyond – AI / ML enabled RPA or Intelligent Automation

As the need for speed becomes a “must” and not just a “nice to have”, manual exception handling also becomes tedious due to mounting workloads in the BFSI sector. It is here that Artificial Intelligence / Machine Learning (AI / ML) algorithms take the simple rule-based models powered by RPA to the next level. AI / ML enabled RPA or Intelligent Automation learns continuously on an on-going basis as and when exception handling cases surface. It builds on the pre-configured algorithms and grows intelligent with time.


Mega merger of US banks case study – Auto-analysis, auto-indexing, and auto-classification of mortgage documents in the DMS of the acquiring bank


A leading US bank went through a merger and acquisition process, wherein it acquired multiple smaller entities. As a result, they sat on a pile of 35+ million pages in an unstructured data format that needed to be duly analyzed and classified into 275 categories so that they were searchable and accessible. The bank needed to be quick with the classification as they had to implement certain statutory requirements for their customers.


The bigger problem was that each of the smaller entities had a different mode of document storage. Some stored the data in paper format and some as scanned documents with each of them having low to medium resolution. The scenario was such that manual classification would have required onboard huge teams to manually analyze, segregate, and classify these documents, requiring many months to take the task to closure.


The bank used AI / ML enabled RPA as a solution to this business conundrum. They converted all the paper documents into digital assets by using Intelligent Document Capture or IDP. They then ran all the documents through a pre-designed and pre-programmed AI / ML enabled RPA engine, which auto-classified the unstructured mortgage documents into the required categories. However, the exercise threw up many exceptions that did not fall into any of the predefined categories. Here, domain experts manually classified the exceptions. The AI / ML algorithms were quick to learn from the exception handling. As a result with each new batch, the AI / ML enabled RPA engine became progressively more intelligent such that the subsequent batches ran without throwing any exceptions.


Trade Finance case – Data-intensive, multi-player ecosystem replete with unstructured documents from different sources


As the trade opens up in the COVID and post-COVID world, banks cannot limit themselves to purely paper based transactions especially in the fast paced Trade Finance sector. Besides, the Trade Finance sector involves a high number of players and inter-company transactions by using financial documents in unstructured format from different sources, multiple banks, and entities. In addition, banking regulations and statutory norms in this domain are becoming stringent by the day and require continuous monitoring of the data transfer between the entities involved.


The Trade Finance scene demands quick resolution of issues as they crop up. The solution is executed through an integrated approach, which uses multiple technologies including RPA, AI / ML, Intelligent Document Processing (IDP), and Automated Decision Management Workflow. The banking personnel have to keep a hawk’s eye on the transactions and validate the data with the AML guidelines and regulatory authorities such as OFAC, Hunter, bank black lists, etc., where the use of the aforesaid technologies becomes indispensable.


Using AI / ML enabled RPA and the allied technologies, improves processing time and turnaround time, reduces the instances of missing documents, and reduces the back and forth that ensues between the entities involved in the Trade Finance scenario. Besides, the automated scenario offers 100% regulatory compliance due to the audit trails and logs that are generated in the process.


Simply put

The BFSI sector has to deal with many challenges amidst constantly changing regulatory requirements and strict AML regulations. The need for speed in a highly dynamic BFSI ecosystem makes AI / ML enabled RPA the silver bullet that offers a seamless and real-time connection among the entities involved. It offers tremendous granularity, abstraction, simplification and hence visibility and transparency in the transactions. It also allows seamless Forensic Investigations in cases of disputes between the entities. It brings all the entities in the distributed BFSI ecosystem, which exist at varying levels of digital maturity, on the same level turf.