The Client Passed KYC. Why Can’t They Trade?

“Can the new fund trade on Monday?” should be an easy question.

The relationship manager knows the asset manager. The client record is active. KYC is complete. Legal has a signed agreement, and credit has approved a limit. Every team appears to have done its part.

But operations is still opening an account. A tax form is attached to the parent company rather than the fund. The agreement covers the right legal entity but not the requested product in that jurisdiction. A market permission has not reached the entitlement system. On Friday afternoon, every function can report that its task is complete—or at least within its service level—while the client still cannot place a trade.

No single system is necessarily wrong. Each system is answering a smaller question than the client asked.

This is the gap that process mining can address. The technology is often discussed in connection with claims automation, where it can identify slow routes, repeated work, and unnecessary handoffs. Those are useful applications. But the same method can be even more valuable in financial-markets onboarding, where one commercial outcome depends on several functions and systems agreeing at the same time.

The goal is not merely to complete KYC faster. It is to make the client usable—safely, lawfully, and with a clear record of how that decision was reached.

Claims Showed What Process Mining Could Do

Process mining reconstructs how work actually moves by reading timestamped events from operational systems. It can show when a case started, which activities occurred, how long it waited, where it was sent back, and which route produced the final outcome. Instead of relying on a workshop description of the process, leaders can see the process recorded in the data.

Claims made the idea easy to understand. A claim usually has a recognizable identifier, a sequence of actions, and a clear outcome. A team can compare straight-through claims with those that were reopened, escalated, or touched many times. It can separate actual handling time from days spent waiting in a queue.

That success can make process mining look like a claims-specific tool. It is not. Claims simply provided a clean example of a much broader problem: important outcomes are often created by a chain of work spread across teams and systems, while conventional reporting measures only one step at a time.

Financial-markets onboarding is a strong next use case because the commercial result is easy to state but difficult to assemble. The firm wants to know whether a particular entity can use a particular product through a particular account, in a particular jurisdiction, on a particular date. A “completed” KYC case answers only part of that question.

Why Onboarding is Harder Than a Claim

Institutional clients are not single records. One relationship may contain a parent company, investment funds, special-purpose vehicles, beneficial owners, controllers, signatories, and service providers. Each entity may require its own combination of evidence, agreements, tax records, credit decisions, accounts, settlement instructions, and product permissions.

The relationships among those records matter as much as their individual status. A document may support several entities. One agreement may govern several accounts. A KYC review may remain valid while a change of control triggers new work. A client can be fully known to the compliance platform and still be missing from the system that controls market access.

Traditional process mining often starts with a single case number. That remains useful for a straightforward workflow, but it can distort onboarding if the entire relationship is forced into one ticket. A better model connects the events to the business objects they concern: the client group, legal entity, beneficial owner, evidence item, agreement, credit approval, account, and permission.

This is what object-centric process mining is designed to do. In plain English, it lets the firm follow several connected records without pretending that they are one record. Celonis uses this approach in its process-intelligence model, relating objects and events so that a delay or decision can be traced across system boundaries.

“Ready to trade” requires connected approvals. All six must refer to the same entity, product, account, jurisdiction, and effective date.

The key point is simple: passing KYC is necessary, but it is not the whole activation outcome.

KYC is One Control Within Onboarding

KYC is sometimes described as a document-collection exercise that ends when the onboarding case is closed. The regulatory model is more durable. FATF Recommendation 10 calls on jurisdictions to require customer due diligence when a relationship is established and ongoing scrutiny throughout the relationship. In the United States, FINRA Rule 3310 requires FINRA member firms to use risk-based ongoing customer due diligence as part of their anti-money-laundering programmes.

That does not mean repeatedly asking every client for the same information. The better objective is to know what evidence is already valid, what has changed, and what the firm’s policy requires next. The Wolfsberg Group’s digital customer lifecycle guidance supports a connected customer profile and more purposeful use of trigger-based review where permitted and appropriately controlled.

Recent US policy illustrates the distinction. In February 2026, FinCEN issued optional exceptive relief allowing covered financial institutions, in defined circumstances, not to repeat beneficial-owner identification and verification every time the same legal entity opens another account. The relief does not remove the first-account requirement, checks prompted by facts that call earlier information into question, ongoing risk-based procedures, or the broader customer-due-diligence obligations.

The lesson is not that controls should be lighter. It is that duplicated activity is not automatically stronger control.

Process mining can help distinguish purposeful review from accidental repetition. It can show where a valid artifact is requested again, where a change fails to trigger the right review, or where an approval exists but never becomes operationally effective. Those are process issues with direct compliance, client, and revenue consequences.

What Process Mining Can Reveal

A conventional onboarding dashboard usually reports queue size, average age, and cases completed. Those measures are useful, but they rarely explain why an apparently complete client is still blocked.

Process mining can reconstruct the full journey and expose several patterns that local reporting hides.

It can identify duplicate client outreach when compliance, tax, legal, and operations ask for similar evidence through separate channels. It can separate working time from waiting time, showing whether a two-week delay came from a necessary review, an unattended handoff, or missing client information. It can reveal circular rework, such as a case returning to legal after an account detail changes and then returning to operations because the revised agreement was not propagated.

It can also detect false completion. A team may close its task when it approves a request, even though the approval has not reached the downstream account or entitlement system. From the team’s point of view, the work is done. From the client’s point of view, nothing has changed.

This is where platforms such as Celonis can add value. They can discover the routes work actually takes, compare them with an intended path, connect events around multiple business objects, and surface the conditions most strongly associated with delay or rework. Those findings give the firm evidence for changing the process rather than simply adding another escalation meeting.

But a conformance score is not a regulatory opinion. A clean route can still rely on weak evidence, while an unusual route may be the correct response to unusual risk. Process mining explains how work moved. Compliance, legal, and credit authorities still decide what is permitted.

Visibility Has to Change the Operating Model

A process map alone will not fix onboarding. The firm needs someone accountable for the result that crosses the functions.

That role might be called client activation owner, onboarding product owner, or lifecycle lead. The name matters less than the remit: coordinate the path from commercial intent to first permissible activity, maintain shared definitions, resolve cross-functional blockers, and improve the journey as a whole.

This does not weaken control independence. Compliance retains authority over financial-crime decisions. Legal retains authority over contractual acceptability. Credit retains its approval rights. The activation owner does not overrule those decisions; the owner prevents necessary independence from turning into unmanaged collective delay.

Celonis or a comparable platform should sit above the systems of record as a measurement and orchestration layer. It should not become a shadow KYC database or an automatic compliance officer. Its job is to connect events, explain the current blocker, identify the accountable owner, and route an approved next action with the relevant context.

Process evidence becomes useful when it reaches the right owner under the right authority. Process mining informs the decision; it does not grant approval.

The first actions do not need to be ambitious. Request the right evidence once. Reuse it only where policy permits. Alert the owner when an activation date is no longer credible. Reopen a case when a downstream permission fails. Send genuine judgment to the appropriate specialist. Update the client state only after the relevant account or entitlement is effective.

AI can support this model by summarising a case, classifying a document, predicting a likely delay, or recommending the next approved action. It should work from connected, current client data and keep predictions clearly separate from formal decisions. An automated agent acting on fragmented process data will only move the fragmentation faster.

Measure Whether the Client Was Able to Work

Average onboarding time is not enough. It can improve because the firm removed waste, rushed a necessary review, or simply moved the delay from an internal queue to the client.

A stronger scorecard follows the outcome from commercial request to actual use. Useful measures include:

  • intent to ready-to-trade;
  • ready-to-trade to first activity;
  • avoidable firm waiting time, separated from client waiting and mandatory control time;
  • repeated client requests and first-time-right evidence;
  • reopened reviews and false-ready incidents;
  • cost per activated entity or product; and
  • revenue delayed or abandoned while activation remains blocked.

These measures should be segmented by client complexity, risk tier, product, and jurisdiction. The purpose is not to force every client through one standard path. It is to preserve necessary variation while removing delay created by poor data, unclear ownership, and disconnected systems.

The Same Foundation Reaches Beyond Onboarding

Customer onboarding and KYC are compelling starting points, but they should not become another narrow technology programme. The same connected model of clients, entities, agreements, accounts, permissions, and events can support other financial-markets processes.

Trade lifecycle and settlement prevention. Connect trades, allocations, affirmations, settlement instructions, securities, counterparties, and obligations. Process mining can reveal the combinations that lead to late matching or settlement failure before the break appears in an operations queue. The objective moves from repairing exceptions to preventing them.

Counterparty credit and limit management. Connect the counterparty hierarchy, requested product, group exposure, netting opinion, collateral terms, limit, and approval. The firm can reduce incomplete requests, repeated review, expired decisions, and limits that are approved but unavailable where the commercial team needs them.

Client lifecycle changes. A new fund, signatory change, settlement-instruction update, product expansion, or event-driven KYC review can create several unrelated tickets. Process mining can show the full change journey, including circular handoffs and repeated client requests, while reusing what the firm already knows about the relationship.

Payments, collateral, reconciliations, and controls. These areas also produce event trails across systems and teams. Celonis’s Standard Bank customer story describes using a process digital twin in cross-border payments and reports straight-through processing above 90 percent. That is a vendor-published example, not an industry benchmark, but it shows how the method can extend well beyond claims and onboarding.

The best candidates share a common feature: an economically important outcome crosses enough systems and owners that no local dashboard can explain the whole journey.

When the Market Opens

The relationship manager should not need six screenshots and four phone calls to answer whether the new fund can trade on Monday. The answer should exist as a current, governed state supported by connected evidence. If the answer is no, the blocker should have an owner, a reason, an economic consequence, and a permitted next action.

That is the broader promise of process mining in financial markets. It can turn the event trails already produced by the firm into a shared account of how work actually runs. It can connect approvals that ticket-based reporting separates. It can show where a valid decision is stuck between systems and where automation would genuinely help.

It does not remove the need for KYC, legal judgment, credit authority, or operational control. It makes their combined result visible and manageable.

When the market opens on Monday, success will not be measured by how many internal workflows say “complete.” It will be measured by whether the client can trade—and whether the firm can explain exactly why.

Unknown's avatar

Author: Jason Miles

A solution-focused developer, engineer, and data specialist focusing on diverse industries. He has led data products and citizen data initiatives for almost twenty years and is an expert in enabling organizations to turn data into insight, and then into action. He holds MS in Analytics from Texas A&M, DAMA CDMP Master, and INFORMS CAP-Expert credentials.