The Pandemic has forced a significant transformation in the banking industry - dependence on digital infrastructure and assets. The shifting consumer behavior, the need for personalized solutions, and the drive to provide operational excellence have led to this alteration.
Despite the push for digital excellence, a lot of banking operations still rely on manual methods. There is an underlying need to automate these processes, not just for speed and productivity but also to mitigate risks and prevent fraud.
Of the many processes, credit analysis and loan disbursements rely heavily on underwriters and manual assessments. This not only increases the overheads and resource costs but also heavily impends the risks associated with cybercriminals.
From the start, Perfios chose to run its data platform on Amazon Web Services (AWS) and in 2016, it became one of the first few users to move to the AWS Asia Pacific (Mumbai) region, ensuring that financial data stayed in the country as per local regulations. Subramani says, “AWS offered us a broad set of ML services in addition to the data security to meet financial regulations.”
The banks in India reported a total fraud of Rs. 4.92Tn as of March 31st, 2021. This is 4.5% of the entire bank credit as per RBI’s data. A total of 45,613 fraud cases were reported across 90 banks.
Identity theft is the single most significant contributor to bank fraud cases. For example, a case was reported in Hyderabad where a gang impersonated dead techies for loans. As a result, they managed to swindle a total amount of Rs. 53 Lakh from the bank.
In another fraud case in Coimbatore, a gang of four pledged forged documents against Rs. 2 Crore loan from five different banks. While the authorities eventually caught the fraudsters, it was too late, and the bank suffered heavy losses in the meantime.
The incompetent processes, legacy systems, and incumbent resources have led to a rise in frauds year-on-year. In addition, there has been a rise in the factors contributing to banking scams, from false addresses to false employment status.
The alarming surge in frauds has also endowed a lowered reputation on the banks, which they are fighting hard to overcome.
In the land of muggles, the CAM Pro is the wand that can slay any dementor and win the magical war. It ensures that you are safe before the fraudsters suck the last penny out of you.
CAM Pro is built on Machine Language and Artificial Intelligence to help credit analysts make more intelligent and quicker decisions. In addition, this tool ensures accurate fraud detection, which allows credit managers to analyze faster and stay productive.
Currently, the bank's underwriter uses manual evaluation methods to track the borrower's condition, capacity, and credit scoring. With CAM pro, credit managers can access the evaluation results within minutes.
An external database verification is when the system verifies all the documents scanned and uploaded by the borrower with legitimate bank documents, government proofs, and another external database. It helps identify if the person borrowing the sum has submitted the correct proofs or not. This will prevent the borrower from pledging fake documents.
It helps the lender get a detailed look into the repayment track record and the current capacity of the borrower. The external database also reveals the credit history and repayment behavior.
When the borrower provides multiple documents to prove they can repay the loan, the system scans and skims through them to identify the borrower's capacity. Manually, this process will take a lot of time as the credit manager has to conduct a thorough cash-flow analysis, check the debt-to-income ratio and even take a peek into their spending capability.
CAM Pro will analyze key data metrics along multiple sources to reveal more accurate results. For example, cross-validation occurs between the financial statements, the borrower's ITR, form 26AS, and other bank documents. This analysis helps detect fraud as early as possible.
The system will detect if all the documents submitted communicate the same numbers or not. Moreover, it helps identify if the borrower's capacity, collateral, and condition are as stated.
Industry insights are very relevant from a fraud detection perspective. These insights will guide the analyst into the typical income earned by the person employed in an individual capacity. For example, the person employed as a plumber will have a certain defined bracket as average income. If the analyst gets documents claiming a salary or income higher than the industry insights, the system automatically detects suspicious behavior and doesn't approve the document.
There are multiple data points that the lender needs to study, evaluate and interpret before disbursing the loans. Each data point offers a critical insight that enables quicker decisions.
Automating the journey reduces the time spent in analysis and reduces the length of the journey.