•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
The surge of digital payments in recent years has not only changed consumer spending behavior but also opened a new data source for the banking sector. From QR and POS transaction histories to cash inflows and outflows in accounts, banks are gradually using payment data to assess credit, particularly for individuals and micro-businesses that have long struggled to access capital due to lack of collateral.
Data-driven lending is widely viewed as a step toward financial inclusion. However, the approach carries risks that are not easy to identify and have not been fully tested through economic cycles.
In traditional credit models, banks rely on financial statements, collateral, and credit history to make decisions. This leaves many micro-entrepreneurs—who often lack formal accounting—almost outside the formal credit system.
Cashless payments are changing that. Recent data indicate digital payments in Vietnam continue to grow at double-digit rates, with QR payments rising by more than 50% in 2025. As more micro-business cash flows move into digital channels, banks can estimate real revenue on daily, weekly, and monthly horizons, analyze cash-flow stability, and build alternative credit scoring models.
Theoretically, this shifts lending from “data-scarce” to “data-rich.” In practice, the key issue is not only the amount of data, but the quality of the data and how it is used.
For retail banks, lending based on cash flow offers three main benefits.
Some banks in Vietnam have begun offering cash-flow-based lending products using funds linked to POS or QR payments. These products are designed to enable flexible credit limits based on customer profiles, automated disbursement, and automatic debt collection when cash flows return. The model is often compared to developed markets, where payment data functions as a form of “soft collateral.”
The model’s biggest weakness is input data quality. Not all micro-entrepreneur revenue passes through digital channels, and cash remains significant in many industries. This creates the risk of misjudging financial capacity if banks rely only on electronic data.
There is also the risk of manipulation. Payment data can be fabricated to inflate apparent revenue, funds can be shuffled among related accounts, and cashback programs can be exploited to “pump” numbers. If scoring models are not sufficiently sophisticated, banks may produce biased credit decisions from the outset—an outcome described as “garbage in, garbage out.”
Another challenge is the cyclical nature of the information. Payment data typically reflects behavior over the past few months, which may be too short to capture business cycles. Yet credit commitments are often mid-term. This mismatch can lead to over-extending credit, and when market conditions turn, cash flows can deteriorate quickly, resulting in higher bad debts.
Unlike collateral, data is not a recoverable asset.
Using payment data raises questions about privacy and the regulatory framework. Key issues include who owns the payment data—banks, payment intermediaries, or customers—and whether data sharing among parties requires explicit consent. The boundary between “personalized service” and “privacy invasion” is not always clear.
Against the backdrop of the State Bank of Vietnam progressively refining the legal framework for digital finance, data regulations are expected to be pivotal in balancing innovation with user protection.
Lending based on payment data is a trend that is difficult to reverse. For the model to develop sustainably, several conditions are highlighted:
Lending based on payment data can expand credit and support financial inclusion. But without adequate controls, it can also become a new source of risk—“insidious but pervasive.”
The experience of earlier periods of rapid credit growth shows that risk does not disappear; it becomes more sophisticated. The central question is not whether to use payment data, but to what extent and with what controls—because competitive advantage depends not only on harvesting data, but on understanding and managing what data cannot fully express.
Premium gym chains are entering a “golden era” that is ending or already in decline, as rising operating costs collide with shifting consumer preferences toward more flexible, community-based ways to exercise. Long-term memberships are shrinking, margins are pressured by higher rents and facility expenses, and competition from smaller, more personalized…
