Home
/ Insights
/ Articles
/ Smarter Compliance for Payment Service Providers: The Data & Analytics Advantage

Smarter Compliance for Payment Service Providers: The Data & Analytics Advantage

Leveraging Data and Analytics to Future-Proof Compliance in the Evolving Payments Landscape

This FTI Consulting article was written in partnership with Pushkar Dandekar, Director, FTI Consulting; Kenneth Pereire, LL.B. (NUS) (Hons), Managing Director, Advocate and Solicitor (Supreme Court of Singapore), KGP Legal; and Lin Yingxin (Mr.), LL.B. (SMU) (Hons), Associate Director, Head of Corporate Practice, Advocate and Solicitor (Supreme Court of Singapore), KGP Legal.

The broader use of digital payments has a dark side — e-commerce fraud is projected to rise from $44.3 billion in 2024 to $107 billion by 2029.¹ In this landscape, Payment Service Providers (“PSPs”) play a vital role in mitigating threats by embedding compliance measures powered by data analytics. With regulators increasing scrutiny, and the cost of non-compliance rising, leveraging data to detect risk, streamline processes and strengthen controls is not just best practice—it’s essential.

Regulated PSPs must comply with rigorous Anti-Money Laundering (“AML”) and Countering the Financing of Terrorism (“CFT”) obligations under local regulatory frameworks—such as those set out by the Monetary Authority of Singapore (“MAS”)² and equivalent regimes in other jurisdictions where they operate. The stakes are high, as regulatory breaches - particularly in KYC and transaction monitoring - can result in enforcement action, reputational damage and loss of licensing status. Regulators are demanding more transparency and tighter controls, as the volume of digital transactions has grown in recent years.³ As such, regulatory compliance is now a major operational focus for PSPs globally —and while many have already adopted data and analytics tools, there is a growing need to fully leverage emerging technologies like machine learning (“ML”) and artificial intelligence (“AI”) to embed compliance capabilities more deeply into core operations.

Using data analytics, PSPs can automate risk detection, streamline onboarding and future-proof regulatory reporting processes. This applied use of data analytics results in faster and smarter compliance that supports both resilience and growth.

Unlocking the Compliance Advantage Through Data Analytics

To meet rising regulatory expectations and mitigate risk, PSPs must go beyond traditional approaches and fully harness the power of data analytics. These tools can help streamline compliance functions, improve operational efficiency, and build future-ready frameworks. Specifically, data analytics can support PSPs in the following key areas:

Real-time transaction monitoring, AML and fraud detection
Smart customer onboarding and Know Your Customer (“KYC”)
Regulatory reporting automation

Real-Time Transaction Monitoring, AML and Fraud Detection
Real-time transaction monitoring is the process of analysing transactions and unearthing unusual patterns from the data in real time. It will allow PSPs to detect and respond to suspicious activities as they happen. Traditional systems coupled with AI and ML based systems will help PSPs safeguard their operations and help to create a robust defence mechanism against fraud and AML activities.

Traditional systems use predefined rules and thresholds to perform transaction monitoring. Modern systems, however, combine a traditional rule-based approach with advanced analytics-based solutions. Machine-learning (“ML”) models can analyse vast amounts of data and can find patterns and correlations that could indicate the potential fraud. Predictive models using artificial intelligence (“AI”) algorithms can be trained on historical data and can flag potential frauds. This will allow PSPs to be more proactive and mitigate risks. AI/ML models can analyse user behaviours by looking at patterns from the data, identifying deviations from the established patterns and flagging suspicious activities. The use of AI/ML enables quicker identification and escalation, reducing the risk of non-compliance with statutory deadlines—legally mandated timeframes for reporting obligations—as well as broader regulatory expectations set by authorities around timeliness, accuracy and proactive risk management. For example, PSPs in Singapore are required to file timely Suspicious Transaction Reports (“STRs”) under the Corruption, Drug Trafficking and Other Serious Crimes (Confiscation of Benefits) Act 1992 and the Terrorism (Suppression of Financing) Act 2002⁴ when potential red flags are identified. By learning from historical data and transaction patterns, ML/AI algorithms can detect these red flags with greater speed and accuracy than manual processes, enabling PSPs to meet reporting obligations more efficiently and reduce the risk of delayed compliance.

AI-based Anomaly Detection Models can detect complex patterns from underlying data which traditional rule-based systems might miss. AI models assess customer data and can dynamically classify AML risk levels. ML based clustering methodologies can discover unknown fraud patterns and club together these transactions to detect outliers and potential frauds.

Smart Onboarding and KYC
One of the challenges in onboarding a new customer is collecting the right data about them. KYC regulations are tedious and complex, and PSPs are required to adhere to them. The traditional ways to perform KYC are often manual, error-prone and quite time-consuming. Poor data collection and verification often lead to compliance gaps and regulatory scrutiny. To address these issues, data analytics-led solutions are necessary. Data analytics can play a major role in automating the whole KYC process with better accuracy and efficiency and reduction in risks. Natural Language Processing (“NLP”) based techniques, coupled with AI models can quickly analyse vast amounts of data and flag potential risks. AI-based systems can quickly scan and verify identity documents and images and can then match them with the customer details. AI-powered name screening tools can automatically compare customer names against sanction lists and politically exposed person (“PEP”) lists and can then help to reduce the number of false positive hits which normally occur during manual reviews. Finally, AI- and ML-based systems can help to automate data collection and the verification process and can fast-track the overall compliance checks cycle, resulting in faster customer on-boarding.

Importantly, automating onboarding and KYC is not simply about efficiency, it is vital for compliance with legal obligations such as verifying beneficial ownership screening against global watchlists and ensuring customer risk assessments are auditable and up to date.

Regulatory Reporting Automation
PSPs are required to submit accurate compliance data to authorities such as the MAS, in accordance with obligations under the Payment Services Act 2019 and AML regulations.⁵ This regulatory reporting often involves a lengthy and complex manual process that requires multiple levels of review by stakeholders across finance and compliance departments. Additionally, one of the key challenges of manual reporting is the need to collect, compile and analyse large volumes of data from various systems and sources. This can result in non-compliance with statutory timelines for submitting STRs and compliance attestations and licensing disclosures under the Payment Services Regulations 2019 — each of which carry potential regulatory consequences. Compliance teams must also constantly adapt to frequent changes in laws, creating operational strain and a higher risk of non-compliance. Furthermore, data security remains a critical concern under the Personal Data Protection Act 2012.⁶

To mitigate these risks, leveraging the full power of data and analytics enables organisations to ensure data integrity and significantly reduce human errors. Advanced analytics tools, such as data integration platforms, regulatory reporting software and AI-powered analytics engines, can automate the validation of data against the latest regulatory requirements, minimising the burden on compliance teams and the risk of oversight. For example, ML algorithms can identify trends and anomalies in transactions data to flag potential compliance breaches before they escalate. Additionally, NLP can automate the review and interpretation of regulatory texts — extracting key requirements and mapping them to appropriate policies and datasets. This not only streamlines the generation of regulatory reports but also helps to ensure that reports are always aligned with current legal and policy standards. These technologies also support audit-readiness and traceability, which are essential for PSPs during MAS inspections or enforcement reviews under the MAS Act 1970.⁷ Together, these technologies improve both the accuracy and responsiveness of regulatory compliance processes, enabling organisations to adapt more efficiently in an increasingly complex regulatory environment.

Establishing Compliance by Design

As regulatory expectations grow across the digital payments ecosystem, PSPs must prioritise a compliance-by-design approach. Leveraging data and analytics is no longer just a differentiator, it is a legal and operational imperative. By embedding AI-driven tools into transaction monitoring, KYC and reporting processes, firms can future-proof their compliance posture while reducing enforcement risks. With regulators such as MAS and FATF tightening oversight,⁸ PSPs who invest in these capabilities today will be better positioned to operate securely and sustainably tomorrow.

Footnotes:

1: “Tracking the Evolution of Payment Fraud in 2025,” Sift (May 22, 2025).

2: For example, MAS Notice 626 on Prevention of Money Laundering and Countering the Financing of Terrorism (June 2025).

3: MAS Technology Risk Management Guidelines (TRM Guidelines) updated in 2021 for digital payment systems.

4: Corruption, Drug Trafficking and Other Serious Crimes (Confiscation of Benefits) Act 1992, and Terrorism (Suppression of Financing) Act 2002.

5: Payment Services Act 2019.

6: Personal Data Protection Act 2012 (No. 26 of 2012) for data security obligations.

7: MAS Act 1970.

8: Virtual Assets: Targeted Update on Implementation of the FATF Standards on Vas and VASPs FATF July 9, 2024.