PwC India’s analysis predicts a major increase in digital financial suspicious transactions, with volumes rising from 159 billion in FY 2023-24 to 481 billion in FY 2028-29.
This sustained growth pattern indicates that the number of transactions in 2025 will be higher than the 159 billion recorded in FY 2023-24. This trend represents a 3x growth of digital financial activities over a five-year period.
However, the latest trends also suggest that bank fraud is on the rise in the same period. For instance, the total number of bank fraud cases in the first half of FY 2024-25 rose 27 % year-on-year to 18,461.
So, lenders must be equipped to identify and report suspicious transactions. This will help them avoid fraud early on.
In this blog post, we outline the tools and strategies that enable lenders to do so with speed, accuracy, and efficiency.
What are Suspicious Transactions?
Suspicious transactions mean financial activities that follow unusual or irregular patterns. In an effort to meet massive sales targets, Indian digital lenders, banks, and NBFCs might not be able to fully assess the risks during their underwriting procedures.
This often leads to:
- Transactions connected to serious illegal activities such as money laundering, tax evasion, and terrorist financing.
- Unhealthy financial behaviours of customers, such as bounced checks, EMI defaults, circular transactions, and too many bank overdrafts.
However, timely access to suspicious transaction data can help lenders safeguard their businesses from these outcomes. However, there are some roadblocks to achieving this.
Tracking Suspicious Transactions: Key Challenges Lenders Face
India’s MSME loan portfolio saw 17.8 % growth, reaching a value of INR 64.1 trillion by March 2024, in comparison to the same period in the previous year.
As the demand continues to grow, lenders are capitalising on this opportunity by expediting loan approvals.
However, without the right technology in place, they are also vulnerable to loan fraud and money laundering.
Here’s a snapshot of factors that make it challenging for lenders to identify and report suspicious transactions:
1. Manual Underwriting Processes
Manual loan application reviews are often prone to oversights. The sheer volume and intricate nature of these applications mean human reviewers can miss fraudulent transactions.
2. Lack of Data
Lenders may not have access to all financial data due to delays and omissions on the part of the loan applicants.
Hence, the underwriting process may be completed without access to comprehensive data and financial analysis. This could result in:
- Inaccurate Risk Assessment: Incomplete data may misrepresent a borrower’s creditworthiness, leading to over- or underestimation of default risk.
- Higher Default Rates: Approving loans without full insight could increase the likelihood of borrowers failing to repay.
3. Poor Data Governance Processes
Lenders lack access to specialised tools that extract, categorise, and organise data in an efficient, accurate manner.
Hence, there is significant scope for data to be duplicated, omitted, or misrepresented.
Flagging Suspicious Transactions: 6 Strategies Lenders Should Adopt
To tackle the issues above, lenders must adopt a holistic list of best practices to identify and report suspicious transactions:
1. Extract All Transactions
In FY 2021-22, the Financial Intelligence Unit-India (FIU-IND) received approximately 433,000 STRs (Suspicious Transaction Report). This significant volume reflects the growing vigilance within India’s financial sector, particularly among lenders like banks and non-banking financial institutions, which are key contributors to these filings.
Clearly, lenders can detect threats early on, provided they have the technology to do so.
Notably, the process of identifying and reporting suspicious transactions starts much before these specific tasks in the lending lifecycle. Lenders must gather all data efficiently and accurately.
For instance, they must extract transactions from a borrower’s bank statement documents for the purpose of in-depth analysis.
However, manual extraction is an inefficient process that consumes time and resources. It also results in errors, duplications, and omissions of extracted data, which can trigger an inaccurate analysis.
By adopting AI-powered bank statement analysis software tools, lenders can automate data extraction, bringing efficiency and accuracy to the process.
For example, with Precisa, data can be extracted from over 700 document formats.
2. Categorise Every Transaction
Lenders must ensure that all transactions are accounted for by segregating them into specific inflow and outflow categories.
- Every transaction can be further classified into over 25 to 30 inflow and outflow categories.
- Outflow categories may include employee salaries, fees, fines and charges, EMIs, customer refunds, supplier payments, credit card payments, and taxes.
- Inflow categories may include revenues, tax refunds, interests, and capital gains.
- The financial analysis software can also record payment modes such as NEFT, IMPS, and credit cards.
3. Comprehensive Transaction Analysis
In the current fiscal year, bank fraud rose to 18,461 cases. By a thorough financial analysis, banks detect signs of loan fraud early on.
Having access to transaction data that is accurately extracted and categorised lets lenders thoroughly analyse all bank transactions.
Also, the analysis process factors in various aspects of consumer behaviour—spending patterns, income sources, and prior credit-related behaviour.
4. Flagging Irregular Transaction Patterns
A superior bank statement analysis software tool leverages AI to detect and flag irregular transaction patterns, which lenders can further investigate. These patterns are often invisible to the human eye.
HSBC’s AI systems monitor over a billion transactions monthly, enabling the bank to process and analyse data more efficiently.
It has also reduced the occurrence of false positives—instances where legitimate transactions are flagged as suspicious—by 60%.
5. Customised Dashboards for Reporting
Lenders can customise reports as per their needs. They can isolate transactions based on various parameters.
Examples of such parameters include:
- Inflow transactions where the sources are unknown
- Outgoing transactions where the destination accounts are unknown.
- Repetitive transactions to and from specific accounts.
- Transactions that relate to specific activities, such as bounced cheques, fines, and penalties.
6. Use Alternate Credit Scoring
Adopting bank analysis software enables lenders to leverage the benefit of alternate credit scores in decision-making.
The scores are automatically generated, taking into account all transactions. These scores provide a benchmark for lenders to make underwriting decisions, including borrowers whose financial data reflects suspicious transactions.
The Takeaway
By adopting AI-powered financial transaction software, lenders can make data-driven decisions. In turn, they reduce the potential for fraud and NPAs.
Here’s how Precisa’s financial data analysis solution simplifies and speeds up the process:
- Automated Data Extraction: Extracts bank statement data instantly, cutting manual entry time.
- Multiple Format Support: Handles over 700 formats and 350+ bank integrations, skipping reformatting.
- Intelligent Analysis: AI categorises transactions and flags risks, replacing manual checks.
- Real-Time Insights: Delivers instant dashboard results, avoiding delays.
- Error Reduction: Validates data automatically, minimising mistakes.
- System Integration: Links with LOS via APIs, streamlining workflows.
- Scalability: Automates tasks to handle more volume with less effort.
The software provides actionable insights on a customisable dashboard, thus helping companies make informed business decisions.
Request a free demo today!
Mumbai