The Digital Transformation of Investment Banking Client Services

Perpetual KYC: Transforming Client Risk Monitoring

This article is the second in a series of three pieces examining the digital transformation of legacy CIB client services operations; the first article explored how AI is reshaping Adverse Media Screening (AMS). Here we turn to an analysis of Know Your Client (KYC) protocols and the current trend of a shift toward perpetual, event-driven monitoring and how it represents one of the most significant changes in how CIBs manage client risk. 

In 2026, the KYC elements of a typical CIB’s overall state of client services operational functions find themselves increasingly under scrutiny – not because institutions lack controls, but rather because those controls have not been updated in many years to account for faster, less predictable risk environments. In a CIB, client structures are complex, ownership changes often follow transactions, mandates evolve and cross-border exposure is common. Yet many CIB KYC frameworks still rely on static, calendar-based assessments that assume customer risk remains broadly stable between review cycles. 

In emagine’s experience, the gap between how risk typically evolves in a CIB and how KYC reviews are performed creates major blind spots; for example, in the arena of financial crime prevention controls.

Perpetual KYC (pKYC) is emerging as the industry’s response to these realities. At its core, pKYC is a continuous, event-driven approach to customer due diligence replacing scheduled reviews with real-time risk monitoring triggered by material changes in a client’s profile or external environment. For many institutions, it represents a fundamental shift from periodic, one-off review as it stipulates the use of new technology to develop and execute a continuous, event-driven approach within a CIB to client monitoring and risk assessment. 

This article explores what a mature pKYC target state would look like within a CIB, why the transition from traditional approaches to KYC to pKYC is challenging and how AI can act as an accelerator when applied responsibly. 

Why Calendar-Driven KYC is No Longer Effective

Within many CIBs, there are three converging pressures that are making the periodic KYC increasingly difficult to sustain:

1. The pace of change in customer risk;
2. Rising compliance expectations; and
3. Operational inefficiency.

Static Calendar-driven Reviews in Dynamic Risk Environments

Periodic KYC assumes stability where little exists. Customers assessments take place during onboarding; they are assigned a risk rating and then often left untouched until the next scheduled review. In a CIB, material risk frequently emerges between cycles through deal activity, corporate actions, changes in ownership and controls, or through interactions with new counterparties.

Meanwhile, the risk environment around those customers continues to evolve – geopolitical volatility, rapidly changing sanctions regimes, new typologies of financial crime and shifting customer behaviour all introduce risk between review cycles. In a CIB, this is acutely visible in deal execution. Sanctions designations now shift faster than most periodic review cycles; a counterparty that was clean at onboarding may carry material risk exposure by the time a deal closes, forcing compliance teams into reactive re-verification under time pressure and with a client relationship on the line.

The conclusion is hard to avoid – periodic KYC is inherently backward-looking. It is calendar-driven rather than risk-driven.

pKYC focuses on the monitoring of meaningful risk signals as they arise. These signals may be triggered by transaction behaviour, deal usage, product usage, counterparty relationships, ownership changes or geographic exposures as they arise. This approach shifts the compliance paradigm. The focus thus shifts from arbitrary calendar dates to material events, allowing institutions to identify and mitigate risk in real-time. 

Regulatory & Business Drivers Raising the Bar

Across the EU and the UK, regulators are increasingly explicit about expectations for ongoing, risk-based monitoring. For a CIB, the supervisory focus often centres on complex structures, high exposure and governance over judgement-based decision making. There is a clear mandate for pro-active, risk-optimised, data-driven approaches as well as the use of analytics to support a robust risk environment.

Regulatory enforcement actions in 2024 and 2025 consistently cited outdated customer data and inaccurate risk classification as primary drivers – a pattern that reflects the structural limits of static review models.

From a business perspective, periodic reviews were highlighted by compliance leaders as a structural bottleneck that drove a high volume of low-value activity, creating rework and leading to remediation programmes when the backlog became unsustainable. This bottlenecking can also introduce friction with front-office teams and impact deal timelines.

pKYC takes the continuous, intelligence-led approach using automated data feeds and defined triggers to keep customer risk profiles updated. The objective is to maintain constant compliance rather than periodic reassurance. 

The Cost & Efficiency Model Under Strain

Periodic KYC is expensive. Work is concentrated in review cycles, creating peaks that require surge capacity, external resourcing and overtime. The same customer information is repeatedly collected and revalidated, even when little changes.

In a CIB where client complexity is high, manual processes, rework and remediation inflate per-review costs without proportionate risk benefit.

In contrast, event-driven monitoring distributes effort across ever-more events, and it directs resources toward cases that genuinely require attention. Meanwhile, CIBs that adopt dynamic monitoring models consistently report reductions in review times and improved cost efficiency.

Here, pKYC delivers ongoing confidence in robust, cost-efficient risk management rather than point-in-time comfort. 

From Periodic Reviews to Perpetual Monitoring: A Conceptual Framework

pKYC is a target operating state, not a process layered onto existing, periodic reviews. In this model, customer risk is continuously reassessed using internal and external signals, and KYC activity is aligned to material changes. It replaces calendar‑based reviews with real‑time insight, allowing institutions to focus their efforts on evolving risks rather than on where dates dictate action. 

In a mature pKYC model, not every signal results in intervention. The emphasis is on the relevance and proportionality, ensuring that action is taken only when it is justified by risk.

Effective pKYC operates as a closed-loop framework comprised of three core components:

1. Trigger monitoring;
2. Automated risk assessment; and
3. Dynamic due diligence. 

Trigger Monitoring – Detecting Material Events

In this framework, the starting point is an event and not a review date. Internal and external triggers continuously feed into the risk framework.

Internal triggers may include:

• significant transaction pattern changes;
• dormant account reactivation;
• product or service changes;
• ownership structure modifications; and
• credit limit increase requests.

External triggers may include:

• sanctions list updates;
• PEP status changes;
• adverse media alerts;
• corporate registry updates; and
• jurisdiction risk rating movements.

These signals are captured via automated data feeds, screening tools and alerting mechanisms. The objective is to detect any material changes early and reliably rather than waiting for their detection in the next review cycle. 

Trigger design is a critical factor. Poorly designed triggers generate noise and overwhelm operations. A well-designed trigger creates focus and confidence.

A mature pKYC model requires compliance leaders to clearly articulate why triggers exist and, equally, why certain detected events do not warrant further actions. 

Industry evidence suggests pKYC can reduce KYC maintenance costs by up to 40% while improving detection accuracy. For example, a leading EU-based bank demonstrated this in practice achieving 91.5% straight-through processing of eligible cases by deploying a perpetual behaviour monitoring platform across 3bn data objects with ML models continuously tracking behavioural shifts and automatically triggering CDD renewal when a material change is detected — no scheduled review cycle required.

Automated Risk Assessment – Recalculating the Risk Rating

Once the triggers are ingested, an analytics engine recalculates the customer risk in real-time. The assessment considers both the nature of the events and existing profile. For example: 

• a sanctions list hit or a negative legal development may drive an immediate move towards high risk;
• a limited product change, in isolation, may have no or only minor impact; or
• a series of smaller events, when viewed cumulatively, may tip a customer from low- to medium-risk.

The guiding principle is proportionality; risk updates must distinguish signal from noise and remain explainable, auditable and defensible.

Dynamic Due Diligence – Initiating the Right Response

Based on the refreshed risk assessment, the appropriate due diligence paths are triggered automatically. This triggering occurs via: 

• automated / STP (straight-through processing) where changes are low‑risk and fully supported by trusted data;
• simplified due diligence for low‑risk changes that require minimal validation;
• standard due diligence for moderate changes that require minimal validation;
• enhanced due diligence when risk thresholds are breached or cumulative risk increases materially.

In a CIB, enhanced due diligence is scoped to a transaction counterparty, jurisdiction or specific entry within a group, ensuring risks are addressed without unnecessary impact on deal execution.

This ensures that KYC effort is directed where it matters most; high-risk cases receive timely and deeper scrutiny, while low-risk cases are handled efficiently with minimal customer friction.

Transitioning from Periodic into Perpetual KYC 

Transitioning to pKYC is a marathon, not a sprint. emagine observes that most financial institutions are adopting a three-phased approach to avoid disrupting existing controls:

• Phase 01: Foundation (Validate Efficacy) – The initial focus is on data quality and consistency. This includes establishing a common data model and taxonomy across the organisation. Institutions typically pilot event-driven reviews with a defined client segment (e.g., high-net-worth individuals) to demonstrate value and control. 
• Phase 02: Scale (Expand Scope) – Once the foundation is in place, the model extends to more complex customer types and jurisdictions. Automation is increased and external data sources are integrated.
• Phase 03: Optimise (Advanced Analytics) – The final stage is the ‘steady state.’ It uses advanced analytics to refine materiality thresholds and improve decision accuracy. The goal is STP for low-risk changes and targeted human intervention where risk increases. 

For CIBs, approaching this transition in a structured, phased manner can typically demonstrate initial value within six to 12 months – covering a data foundation and a defined pilot segment. Full enterprise deployment across complex CIB populations and multiple jurisdictions realistically spans 18-24 months depending on data maturity and operating model readiness.

Common Challenges in the Transition

Despite its benefits, CIBs face recurring challenges when moving toward pKYC:

• Data Quality & Consistency – Without strong data integrity, continuous monitoring amplifies existing issues. CIBs often underestimate the effort required to align data models and eliminate duplication.
• Trigger Governance – Triggers must be explainable, defensible and aligned to risk appetite. Too many triggers overwhelm operations; too few create blind spots.
• Operating Model Readiness – Transitioning to event-driven work requires a cultural shift. Teams must adapt to variable volumes and prioritisation based on risk, not deadlines.
• Regulatory Comfort – Supervisors expect innovation as well as transparency. Institutions must explain how pKYC decisions are made, validated and controlled.

Recognising these challenges early allows for a more measured and defensible transition.

How AI Accelerates the Transition: Moving from Automation to Intelligence

AI can be a powerful accelerator in the transition to pKYC where continuous monitoring generates volumes of data beyond human capacity to process effectively. 

AI supports pKYC in four key areas:

1. Data fusion;
2. Continuous risk scoring;
3. Analyst augmentation; and
4. Narrative generation.

Each one of these focuses on reducing the time and increasing the precision of checks or reviews. For example: 

• Data Fusion – AI and ML models can be a solution to the resolution of fragmented customer identities across systems. Graph analytics link customers, account, directors and counterparties, allowing a single event to update risk across related relationships. In practice, client data is rarely consistent across onboarding, credit and product systems which blocks creating a 360-degree view of client risk.
• Continuous Risk Scoring – AI-driven models that update risk scores dynamically as new signals arrive, incorporating both structured data and unstructured sources such as adverse media. 
• Analyst Augmentation – AI models supporting the analyst by triaging alerts, surfacing related entities and summarising evidence. Natural language processing extracts relevant information from documents and open-source material, reducing manual efforts while preserving human judgement.
• Narrative Generation – GenAI can assist in drafting structured, auditable narrative for case files, SARs and regulatory responses based on validated inputs and evidence. 

Agentic AI, where systems can autonomously orchestrate multi-step KYC tasks such as entity verification and risk narrative generation, represents the next frontier for pKYC. 

However, emagine believes that AI should be applied responsibly and in a controlled manner, with clear problem definitions and a phased approach that includes starting with simple, interpretable models, maintaining training and validation datasets, and testing for bias and skewness. 

AI acts as a technical catalyst providing the insights, ensuring humans are kept in the centre of a loop that allows them to take the high impact decisions. Explainability remains critical. When done properly, AI helps to strengthen decision-making by ensuring risk detection occurs faster, consistently and transparently.

UK regulators, including the FCA, are consistent in their message that the sophistication of a pKYC model matters far less than how well it is governed and controlled.

Key expectations are: 

• clear definitions of material risk change;
• explainable and auditable decisions; 
• ongoing monitoring of model performance and bias;
• clear accountability as automation increases; and
• the protection of customer outcomes and proportionality.

Already, EU regulations like the EU AI Act are set to classify AI-driven risk scoring as high risk from August 2026, requiring explainability, bias testing and documented human oversight – considerations that should be designed from the outset, not retrofitted later.

CIBs that address these expectations clearly and consistently tend to have constructive and confident regulatory engagement around pKYC.

The Future of KYC: Accountability, Governance & Human Oversight

While pKYC accelerates the detection and assessment of customer risk, it does not dilute accountability. In practice, continuous monitoring and automation often increases responsibility, and ownership of risk outcomes remains within the compliance team.

In a mature pKYC model, compliance leadership remains accountable for:

• defining a material change in customer risk and the controls associated;
• oversight on trigger design, calibration and ongoing tuning;
• risk thresholds and escalation pathways;
• ensuring the risk decisions are reviewed and vetted by humans; and
• maintaining effective human oversight for high-impact outcomes.

pKYC changes how risks are detected and assessed; it does not change who owns the consequences.

As such, pKYC reframes customer due diligence from a periodic obligation into a continuous risk control embedded within CIB operations, perfectly aligning it closely with how financial crime risk evolves. AI plays a critical role in enabling this shift, but it does not remove the need for judgement, governance, or ownership.

CIBs that approach pKYC with discipline strengthen both risk management and regulatory confidence. Those that do not may find existing KYC models increasingly difficult to defend. While this article focuses on CIBs, the underlying framework applies equally across retail, wealth and other business lines. The methodology is common, but the complexity of application is not.

Would you like to know more?

Talk to emagine to learn more about how perpetual KYC can transform client risk monitoring across your CIB operations.

In our next article in this series, we explore how Agentic AI represents the next frontier of risk and compliance in CIB, focusing on systems capable of autonomously orchestrating multi-step tasks.