One legal entity. Multiple IDs. Three business risks.

Charlie Browne, head of sell-side solutions at GoldenSource, explains how fragmented entity IDs undermine risk, compliance and revenue decisions

The entity ID problem starts when the same legal entity is represented by different names, identifiers and records across a bank’s systems. Each record may be accurate on its own but, without a consistent way to establish that they refer to the same legal entity, the bank is left with fragmented views of who it is dealing with.

One system may use the legal name. Another may record a trading name or local subsidiary. A third might retain an identifier created before an acquisition or restructuring. Client onboarding, credit, compliance and reference data teams could each be working from a different version.

Unlike securities, where banks typically anchor records to widely adopted industry identifiers such as International Securities Identification Numbers and Committee on Uniform Securities Identification Procedures numbers, legal-entity identification is far less standardised in practice. External identifiers such as Legal Entity Identifiers, Dun & Bradstreet D‑U‑N‑S Numbers and Business Identifier Codes provide important common standards, but they are not used consistently across every source, jurisdiction or internal system.

Banks also maintain proprietary, system-specific entity IDs, often for the same organisation. One legal entity could legitimately appear under several identifiers, names and records across the bank.

Imagine the same patient registered under several different medical record numbers. Each record may be accurate, but no doctor has access to the complete history of the patient. Until those identities are reconciled, every decision is being made with only part of the picture in place.

Entity ID fragmentation creates the same problem for banks. If one legal entity is represented by multiple IDs, a bank might fail to see its full counterparty risk and it may apply compliance obligations consistently or recognise the complete customer relationship and the revenue opportunities within it.

When identity is fragmented, so is the business view

Entity ID resolution depends on both data quality and data context. Accurate records are essential, but so are the hierarchies, relationships and business meanings that establish how each legal entity connects to the wider organisation.

If the same counterparty is represented by multiple entity IDs, downstream risk systems could struggle to aggregate activity consistently against the correct legal entity, sector, country and rating bucket. The quality of the downstream risk view therefore starts with getting the identity right and the contextual data behind it upstream.

That same foundation is critical to regulatory compliance. Frameworks – including single-counterparty credit limits, BCBS 239 and related risk data aggregation and risk reporting expectations, as well as Basel III credit risk requirements – all depend on banks being able to identify counterparties consistently and aggregate and report risk accurately.

Entity ID resolution sits at the intersection of three important outcomes: better visibility of risk, stronger regulatory compliance and better-informed revenue opportunity

Fragmented entity data can also weaken other compliance processes. Ownership, jurisdiction, onboarding status or regulatory information may sit against different records, making it harder to establish a consistent view of the entity and demonstrate that the correct controls have been applied.

There is also a commercial impact.

Banks’ entity databases do not merely contain organisations they already serve. They can also contain legal entities across the markets that a bank wants to address. If identities and corporate relationships are fragmented, a bank may have an incomplete picture of its existing customers and an inaccurate view of its target market.

That can affect questions such as: Who are our target customers? How many are there? Which products do they already use? Where are the gaps in the relationship? Where are the strongest cross-selling opportunities? And which prospects represent new revenue potential?

Entity ID resolution therefore sits at the intersection of three important outcomes: better visibility of risk, stronger regulatory compliance and better-informed revenue opportunity.

Matching records is only the beginning

Entity ID resolution requires more than comparing names.

Organisations change names, acquire subsidiaries, merge, split and restructure. These corporate actions can create new legal entities, change ownership relationships and alter the identifiers and hierarchies on which downstream systems depend.

They can also have consequences beyond the entity master. A merger or restructuring may create or retire securities, affect pricing information or change other product and reference data relationships.

Two records with similar names could therefore represent the same entity; two records with identical names may not.

A reliable entity ID resolution process needs to consider legal names, aliases, internal and external identifiers, addresses, jurisdictions, registration details, and parent, subsidiary and ownership relationships.

But establishing identity is only the starting point.

A unique master entity ID also becomes the anchor for the rich entity, reference and market data the business depends on. That can include:

• Issuers and links to issued securities
• Credit ratings
• Legal agreement definitions, including International Swaps and Derivatives Association agreements, credit support annexes, Global Master Repurchase Agreements and loan agreements
• Netting sets
• Collateral and haircut data
• Bond prices
• Issuer spreads
• Credit default swap spreads
• Proxy pricing methodologies

This connected context is what allows downstream systems to use the entity ID with confidence.

The question is therefore not merely: Are these two records the same? It is also: What information, controls and downstream decisions depend on getting that identity right?

Resolving identity through trusted context

Getting that identity right requires more than a better matching technique. Artificial intelligence can help uncover potential matches, investigate discrepancies and explain relationships, but reliable outcomes depend on the underlying data being mastered, governed and connected. It needs the context to understand both who or what an entity is, and which exposures, obligations and client relationships depend on that identity.

GoldenSource brings enterprise data management (EDM) and AI together in a single capital markets data platform, wrapped in a Trusted Contextual Data Layer. The platform combines three capabilities: mastering and governing data, making it available for analytics and downstream workflows, and applying AI to investigate and interpret it.

The Trusted Contextual Data Layer connects entity records with hierarchies, ownership relationships, instruments, accounts and relevant market data, while preserving the firm’s overrides, validation rules, lineage and approvals. This gives risk, compliance and commercial teams the connected business context they need to use entity information with confidence.

GoldenSource EDM establishes the trusted entity record. GoldenSource AI helps teams investigate potential matches and exceptions, understand the evidence behind an identity decision and explore its implications across the business. Together, they support AI-assisted outcomes grounded in governed data, with human oversight for consequential changes.

Putting entity intelligence to work

Within GoldenSource EDM, the two-step match-and-merge process creates a trusted master entity ID. GoldenSource AI supports investigation and explanation around that governed process.

Matching identifies candidate records that refer to the same legal entity and links them to the master identity. Merging determines which attributes should populate the master record, using the firm’s source priorities, rules and approvals.

GoldenSource Scout uses AI grounded in trusted entity, reference and market data to match, merge and resolve identity while preserving the context downstream decisions depend on

GoldenSource’s entity resolution combines established matching techniques with machine learning and AI, including deterministic matching, machine learning-enhanced scoring and more advanced deep-learning approaches, together with defined rules and human review. GoldenSource AI helps teams investigate candidates, compare the supporting evidence and understand the relationships behind a potential match.

Where confidence is insufficient, cases can be routed for human review. GoldenSource EDM preserves the lineage, matching evidence and approvals supporting the final master record. GoldenSource AI makes that evidence easier to investigate and explain without bypassing established controls.

This enables teams to resolve identities more consistently, reduce manual investigation and provide downstream systems with a stronger data foundation for risk, compliance and revenue-related decisions.

GoldenSource EDM establishes the trusted entity record. GoldenSource AI puts its governed context to work through AI-assisted investigation and explanation, preserving the evidence downstream decisions depend on.

From entity identity to business impact

A definitive entity view gives a bank more than just a cleaner entity master. It creates a governed foundation for the systems and teams that need to understand the correct legal entity and the data connected to it.

For risk teams, accurate entity IDs and corporate hierarchies provide the upstream foundation downstream systems need to aggregate counterparty and credit risk consistently.

Trusted decisions downstream begin with trusted entity IDs upstream

For compliance teams, the same trusted identity helps connect ownership, jurisdiction, legal agreements and regulatory information to the correct entity, strengthening controls and supporting more consistent regulatory processes.

For customer-facing businesses, trusted entity ID resolution can provide a much clearer view of existing customers and the wider addressable market. By connecting entities across corporate structures and relationships, banks can better understand who their customers are, how many target prospects exist, which products they already use and where cross-sell or new business opportunities may lie.

That commercial value is relevant across the industry. For a global bank, it can strengthen customer intelligence across retail and other client-facing businesses. For a regional bank, it can provide the foundation for understanding its customer base, defining its target market and identifying future revenue opportunities.

The use cases might differ, but the underlying requirement is the same. Risk, compliance and commercial teams all depend on getting the entity ID right upstream.

Entity ID resolution is therefore not the end-goal. It is the foundation for better risk visibility, stronger compliance and better-informed revenue decisions downstream.

Trusted decisions downstream begin with trusted entity IDs upstream.

スポンサーコンテンツ