Impactsure implements an AI-enabled trade finance solution for an Indian private sector bank

Impactsure implements an AI-enabled trade finance solution for an Indian private sector bank

March 20, 2023

Impactsure deploys a solution that can easily process complex trade finance documents using artificial intelligence.

Technology adoption in the banking sector has seen an impetus in recent times, with an increased focus on advanced technologies and their impact on all aspects of banking. AI-based solutions solve many challenges in banks and help in increasing productivity, minimizing turnaround times, improving compliance, as well as reducing costs.

Trade Finance covers many financial instruments that banks and companies utilise to make trade transactions feasible. Multiple documents are involved in these transactions that the bank, buyer, and seller must scrutinize for accuracy.

A leading private sector bank headquartered in India was already using a trade finance workflow application from a major software provider. However, they still had to manually analyse documents required for Bank Guarantees, costing them time and effort. For this reason, the bank chose Impactsure’s SureMatch to identify, classify, match and trigger exceptions based on multiple documents required in Trade Finance.

Project and Solution Overview

Since the bank had already invested a significant amount in the existing software to process the workflow and in training their staff, they could not replace the existing system entirely. They had to find ways to improve its efficiency without disrupting the existing workflow while retaining the security and compliance requirements of the bank.

Impactsure studied the requirements of the bank and trained its AI solution – SureMatch – to solve a critical bottleneck in the Bank Guarantee vetting process.

SureMatch is developed with Python and Java using a microservices architecture and uses an advanced NLP matching engine to match clauses by understanding the language and context to scrutinize the documents and highlight onerous clauses, unmatched content, and potential discrepancies.

Implementation Process

Implementation of the solution followed principles of Agile and DevOps. along with the bank’s standard IT guidelines, security, and infrastructure policies to ensure alignment with prescribed governance requirements.

The time taken to implement SureMatch was approximately three months and it was deployed at the bank as an on-premise solution, with the project going live in March 2021.

There were two main challenges that Impactsure needed to contend with while implementing their AI solution:

Integration with the existing trade finance workflow platform

Training the AI software to read documents with poor scan quality

However, the solution from Impactsure was flexible enough to overcome these challenges.

All the key aspects of the implementation, from configuration to deployment, training and support were done remotely due to the pandemic.

Benefits Achieved

The bank achieved several key benefits through the implementation Impactsure’s AI solution:

The bank wanted to decrease time and effort taken to analyse documentation. SureMatch reduced the average time taken to scrutinise exceptions in Bank Guarantee clauses. The turn-around-time improved dramatically – from one hour to less than 10 minutes. This reduction of manual effort enabled the bank’s officers to spend more time with customers.

SureMatch could examine documents, compare them with previously vetted ‘best-matched documents’ and identify onerous clauses. These functions gave bank officers the confidence to vet documents quickly and accurately.

Compliance is a key requirement in trade finance for every bank. SureMatch has a built-in trainable AI engine that the bank staff can update and train with new regulations, permissions, and newly allowable pre-vetted clauses. This ensures compliance with international, national, bank-specific, sector-specific and client-specific regulatory requirements. This enabled the bank to maintain SureMatch independently and reduce the total cost of ownership.

SureMatch also allowed the creation of a repository for documents and past bank guarantees that served as a data library.

Conclusion

The project aimed at implementing an AI-based solution that could easily scrutinise and vet complex documents used for Bank Guarantees. Turn-around-time, attending to customers, and adhering to regulations were crucial factors for the bank’s solution choice. With the help of SureMatch’s AI technology, the bank was able to streamline and enhance trade finance activities.

Key Facts