Bank Automates XML Cleansing Process

Combating money laundering is a global priority for financial institutions, and Brazil is no exception. Banks are required to notify the Financial Activities Control Council (Coaf) of any suspicious transactions that may indicate illicit practices. This notification must be timely and in the correct format as mandated by regulations. A crucial part of this process is generating detailed reports in XML (Extensible Markup Language) format, a widely used markup language for exchanging structured data between systems.

XML is a format that allows for the organization, storage, and transportation of data in a way that is readable by both humans and machines. In the context of Coaf reports, XML files contain detailed information on financial transactions, which must be accurate and complete for authorities to effectively analyze them. However, creating and maintaining these files can be challenging, particularly in ensuring that all data is correct and formatted properly.

To overcome these challenges, it is necessary to perform enrichment and cleansing of XML files. Enrichment involves adding relevant and complementary information to the data, ensuring that all necessary details are included. Cleansing involves removing incorrect, unnecessary, or duplicate data and correcting formatting errors. These steps ensure that reports meet regulatory requirements and are useful for detecting suspicious activities.

In this context, one of Brazil’s largest financial institutions implemented the Garnet microservice developed by act digital. This system automates the enrichment and cleansing process for XML files using advanced technologies like RPA (Robotic Process Automation). The automation improves process efficiency, reducing the time spent on these tasks and errors associated with data handling, ensuring that reports are sent to Coaf quickly and accurately.

Client: Leader in Digital Financial Solutions

The financial institution that engaged act digital for this project is recognized as one of the largest digital banks in the world. It currently serves over 100 million customers across Latin America, including Brazil, Colombia, and Mexico.

Founded in 2013 in Brazil, the bank has experienced rapid and consistent growth, now ranking as the fourth-largest financial institution in Latin America. Its extensive product and service offerings aim to impact customers’ lives, from financial inclusion to access to new investment opportunities.

Challenge: Manual Processing of Reports for Coaf

The bank faced various challenges due to manual processes related to analyzing and reporting suspicious money laundering cases to Coaf. Reports were submitted to the agency in XML format, which often contained errors.

One of the main obstacles was the complexity of the XML format itself, which demands precision and detail in data structuring. Each XML file must contain all necessary information for authorities to effectively perform their analyses.

Another difficulty was the high volume of suspicious cases processed by the bank monthly, around 4,000. This manual process was not only time-consuming but also prone to human errors, such as incorrect or incomplete data entry and inconsistency in applying analysis criteria.

Ensuring regulatory compliance was also a challenge. Reports to Coaf must be submitted in a specific format and within designated deadlines (e.g., within 24 hours of detecting suspicious activities). Any delay or error in submission could result in penalties for the bank and damage its reputation.

Solution: Garnet Microservice for XML Automation and Cleansing

The solution developed by act digital involves fully automating the enrichment and cleansing process of XML files, essential for reporting potential money laundering cases to Coaf. The Garnet microservice is the core component of this automation, ensuring that all data is processed correctly and in a timely manner.

Methodology

Garnet’s architecture is based on a State Machine, a computational model used to design logical systems that can be in a finite number of states, responding to specific events or conditions.

In this process, each state represents the stages of extraction, transformation, transportation, and reporting of data generated by Sherlock to Coaf following its API structure with compatible data on suspicious cases and their transactions, including analyst feedback.

- Extraction: The system retrieves initial analysis information on suspicious cases and related clients, extracting and storing it in a database (DynamoDB) for future reference.

- Transformation: The system retrieves all investigation information from Sherlock, checking if the investigation was manually completed by an analyst. If so, the state is marked as “manually processed.”

- Transportation: The transformed data is then prepared and transported to the next stage.

- Report: The data is reported to Coaf according to the required API structure, with analyst feedback integrated into the process.

Garnet is managed by a scheduled task (cron job) that schedules executions and calls the corresponding APIs, ensuring that all stages are completed sequentially and without failures.

Technologies Used

- XML: The primary data structure used to describe, store, and transport financial information.

- RPA (Robotic Process Automation): Automates repetitive and manual tasks, enhancing efficiency and reducing human error.

- DynamoDB: A NoSQL (non-relational) database used to store state information and references, ensuring data integrity.

Results: Enhanced Efficiency and Recognition from Coaf

The implementation of the Garnet microservice provided the bank with several benefits. The automation resulted in increased efficiency in performing activities and improved the institution’s reputation with Coaf, demonstrating that the solution also helped mitigate regulatory risks.

Key Achievements

• Savings of 666 hours per month.

• Recognition as the financial institution with the best reporting of suspicious cases to Coaf.

• Increased accuracy and standardization of reports.

• Reduction of manual errors and rework.

• Compliance with regulatory deadlines.

• Processing of 400 to 500 reports daily.

How We Can Help Your Company

act digital leverages its expertise in innovation and technology to assist financial institutions in transitioning to the digital environment, ensuring that this migration occurs with maximum security and integrity. The company optimizes data management, significantly improving the operational efficiency of its clients’ businesses.

With a focus on comprehensive solutions, we offer new operational models that enable financial institutions to serve multiple digital channels. From conception to implementation, our complete solutions ensure that each step of the process is executed with excellence and innovation.

act digital is dedicated to driving the success of its clients in the digital world, helping them excel in a competitive market. We aim to deliver solutions that meet our clients’ current needs while preparing them for future challenges.

Transform the efficiency and compliance of your reporting processes with act digital’s solutions. Contact us today.