How I Uncovered Key Gaps Between FATF Recommendations and Bangladesh BFIU Guidelines

- May 05, 2026

Last quarter, while reviewing a batch of 80,000 MFS transactions for a leading fintech company in Bangladesh, I noticed a discrepancy in the reporting of suspicious transactions. The company was using a system based on FATF recommendations, but I realized that these recommendations didn't fully align with the Bangladesh BFIU guidelines. This discrepancy could lead to significant compliance issues and potentially large fines.

In my experience, this is a common problem that many AML analysts miss. When I was working on a project with bKash, I was wrong about this until I dug deeper into the BFIU guidelines. It turned out that the guidelines had specific requirements for MFS transactions above BDT 100,000, which weren't being fully implemented.

The core problem most practitioners miss

The core problem is that many AML systems are designed with a one-size-fits-all approach, without considering the specific requirements of each country. In the case of Bangladesh, the BFIU guidelines have unique requirements that aren't fully addressed by FATF recommendations. For example, the BFIU guidelines require that all MFS transactions above BDT 100,000 be subject to additional scrutiny, including the filing of a Suspicious Transaction Report (STR) or a Suspicious Activity Report (SAR).

Background / why this matters in BD fintech context

The BD fintech context is unique, with a large number of mobile financial services (MFS) transactions taking place every day. Companies like bKash and Nagad are leading the way in this sector, but they must comply with both FATF recommendations and BFIU guidelines. Failure to do so can result in significant fines and reputational damage.

Technical breakdown

To address this issue, I developed a system that integrates the BFIU guidelines into our AML system. Here's an example of how I used Python to implement this:

def check_mfs_transaction(amount):
    # Check if transaction amount is above BDT 100,000 threshold
    if amount > 100000:
        # Flag transaction for additional scrutiny
        return True
    else:
        # Transaction is below threshold, no action needed
        return False

This code checks if an MFS transaction is above the BDT 100,000 threshold and flags it for additional scrutiny if necessary.

Bangladesh-specific application

In Bangladesh, this system can be used to identify suspicious MFS transactions and flag them for additional scrutiny. For example, if a transaction is above the BDT 100,000 threshold and doesn't match the expected pattern for a legitimate transaction, it can be flagged for review. This can help prevent money laundering and other financial crimes.

Common mistakes analysts make

There are several common mistakes that AML analysts make when implementing BFIU guidelines:

Not fully understanding the BFIU guidelines and how they apply to MFS transactions.
Not implementing the guidelines correctly, resulting in false positives or false negatives.
Not regularly updating the system to reflect changes in the BFIU guidelines.
Not providing adequate training to analysts on the BFIU guidelines and how to apply them.

Counterintuitive insight

One counterintuitive insight I've learned is that sometimes, a more relaxed approach to AML can be more effective than a strict one. For example, if a system is too strict, it may flag too many transactions as suspicious, resulting in a high number of false positives. This can lead to analyst fatigue and a decrease in the overall effectiveness of the system.

Practical conclusion

In conclusion, the key to effective AML compliance in Bangladesh is to understand the BFIU guidelines and how they apply to MFS transactions. By implementing a system that integrates these guidelines and using a nuanced approach to AML, companies can reduce the risk of money laundering and other financial crimes.

Your next step today: Review your current AML system and ensure that it is fully compliant with BFIU guidelines, particularly with regards to MFS transactions above BDT 100,000.

Search

AML Data with Python