Merchant monitoring across the lifecycle

The challenge in merchant monitoring

Modern PSPs and marketplaces onboard thousands of merchants with very different risk profiles. Underwriting teams run deep checks before go‑live, but once merchants are active, monitoring often relies on periodic reviews or manual checks triggered by complaints and incidents. In practice, risk and compliance teams struggle to keep up with how quickly merchants can change.

Signals are scattered across many systems. External data providers update insolvency status, ownership, sanctions, negative media or financial scores. Internal sources generate their own signals: chargebacks, delivery issues, customer disputes, risk tickets, and support notes. Each dataset tells part of the story, but no one has the time to connect everything into a coherent merchant‑level view.

Without a central, dynamic view, monitoring becomes reactive. Teams discover that a merchant is in trouble only after volumes spike in risky categories, a regulator raises questions, or a major incident occurs. Reviews are often ad‑hoc; some merchants are re‑assessed too often, others not at all. Documenting why a merchant was allowed to stay active, despite accumulating weak signals, is difficult when evidence is spread across tools.

Risk and compliance teams need a way to turn all these signals into a clear, explainable risk score per merchant, and to focus attention on the few merchants whose situation is actually changing.

Solution: an AI agent for continuous merchant monitoring

Alphaguard introduces an AI agent that continuously monitors merchants using a configurable scorecard, built from both external and internal signals. Instead of replacing existing providers or rules, the agent orchestrates them into one risk narrative per merchant and keeps that narrative up to date over time.

Signals are grouped into clusters such as financial indicators, business verifications, merchant reputation, and directors’ profile. Each signal contributes points to an overall risk score: new sanctions matches, insolvency events, delivery delays, negative media, internal watchlist hits, or changes in ownership structure, for example. The AI agent ingests these signals as they are refreshed, recalculates the score, and evaluates whether the change is material enough to require human review.

When a merchant crosses a threshold or accumulates concerning patterns, the agent opens a monitoring case. It compiles the relevant signals, summarises what has changed since the last review, and proposes a recommended action: keep monitoring, adjust risk limits, request updated documentation, or initiate offboarding. Each case includes a structured explanation that ties specific signals to the recommended decision.

For low‑risk merchants, the agent quietly maintains the scorecard without creating noise. Improving signals such as stable delivery performance, or improved third‑party scores can even reduce the risk score, reflecting a healthier profile over time. This allows teams to reserve manual reviews for merchants where something has truly shifted.

Every monitoring decision is logged with its underlying signals and rationale. Risk leaders can audit how many merchants are in each risk band, see why a merchant was escalated or de‑escalated, and refine scoring rules without losing the history of past decisions. The result is a portfolio‑level view where the riskiest and fastest‑changing merchants naturally float to the top of the queue.

Designed outcomes for merchant portfolios

This merchant monitoring workflow is designed to reduce the time spent on routine reviews while increasing coverage across the entire portfolio. By automating score updates and case creation, the AI agent lets risk and compliance teams focus on complex merchants and policy decisions rather than chasing data.

Consistent scoring, explanations and decision trails make it easier to demonstrate to regulators and partners that merchant risk is monitored continuously, not just at onboarding. Most importantly, merchants are assessed on how they behave over time, not only on who they were on day one.

‍