On this page

Procurement teams writing an RFP for "automated UBO discovery across EU jurisdictions" routinely get responses from a dozen vendors, all of whom claim to support it. Most of those claims deserve significant scrutiny. The EU's post-2022 beneficial-ownership access environment is genuinely constrained, and the phrase "automated UBO discovery" can mean very different things depending on how a vendor has architected its data pipeline. This article gives you the evaluative precision you need to tell them apart, select a defensible shortlist, and structure an RFP that elicits honest answers.

What "automated UBO discovery" actually means (and why the definition matters for your RFP)

Before you can evaluate a vendor, you need to pin down what your organization means by "discovery." There are three architecturally distinct approaches vendors use, and they carry different accuracy, freshness, and auditability implications.

Direct beneficial-ownership register retrieval means the vendor queries the national UBO register for a given jurisdiction and returns structured data sourced directly from that register. This is the highest-evidential-quality model, but it's also the most constrained by access rules, as discussed in the next section.

Registry-derived ownership chain reconstruction means the vendor queries underlying company registers (not the dedicated UBO register), retrieves shareholder and director filings, then applies ownership logic recursively until a natural-person UBO is resolved. Evidence comes from company filings rather than UBO registers directly. Many vendors in the EU operate primarily this way.

Aggregated graph mapping means the vendor has pre-ingested large volumes of corporate data, built an ownership graph, and resolves UBO chains from that graph at query time. Coverage is broad and latency is low, but freshness depends on ingestion cadence, not live-register access.

Your RFP must specify which model (or combination) you require. It must also define the expected deliverables explicitly. A UBO name and percentage alone is insufficient for most regulated-enterprise use cases. Auditors and regulators typically expect the full ownership path, percentage per hop, source documents per hop, timestamps for each data point, and the logic used to apply the 25% ownership or control threshold. If you also need a REST API or webhook integration for automated workflow triggering, that's a separate specification from what data is produced.

Understanding these distinctions is also the foundation for automated compliance decisioning workflows, where the quality of inputs into an ownership resolution step determines how much analyst escalation follows.

EU register access constraints every vendor claim must be tested against

No vendor can honestly claim fully open, automated, API-level access to every EU member state's beneficial-ownership register. The legal environment makes that impossible.

The BORIS (Beneficial Ownership Register Interconnection System), as described by European e-Justice (June 29, 2026), is an EU interconnection tool that links national central registers containing information on beneficial owners. BORIS improves discoverability across borders, but it doesn't grant unlimited programmatic access. Each member state's register governs its own access rules.

The 2022 CJEU ruling fundamentally changed the landscape. Before it, many EU member state registers allowed broad public access to UBO data. The court's decision restricted that access on privacy grounds, and member states have progressively tightened their rules since. The practical result is that most beneficial-ownership register access now requires demonstrated "legitimate interest" (LIA) or a classification as an "obliged entity" (a bank, fintech, law firm, or other AML-regulated business). Some registers have tiered access with materially different data fields per tier.

The Netherlands is a concrete example. KVK (the Netherlands Chamber of Commerce) describes three access tiers: UBOB1, UBOB2, and UBOB3. Each tier unlocks different fields, and the afschermen (shielding) rules mean that certain UBO records can be partially hidden for privacy protection reasons. A vendor claiming "full Netherlands UBO register access" needs to specify exactly which access tier they operate under and what shielding scenarios affect their output.

The implications for your RFP are direct. You need to ask each vendor:

  • What access mechanism do you use per EU member state (direct register, company registry, or graph)?

  • What obliged-entity or legitimate-interest credentials are required from the customer, and who establishes eligibility?

  • For countries where direct access is restricted, what is your fallback strategy, and how is the fallback clearly flagged in the output?

Vendors that cannot answer these questions with specificity for your target jurisdiction list are signaling a gap you should not accept at contract signature.

How UBO chain resolution should be evaluated

Once you've established access model and jurisdiction coverage, the evaluation moves to the technical quality of ownership chain resolution. Four dimensions matter most.

Entity resolution quality is the baseline. Before any ownership traversal begins, the vendor must correctly identify which corporate entity in a registry matches the entity you submitted. Weak entity matching, especially across multilingual name formats (German, Polish, and Romanian company names in the same portfolio, for instance), generates missed connections and false UBO attributions. Ask vendors for their match rate on your specific entity population, not a generic benchmark.

Recursive traversal depth determines whether the vendor can follow ownership chains across multiple hops, including chains that cross jurisdictions. A Dutch holding company owned by a Luxembourg SOPARFI owned by a Cayman trust requires the vendor to traverse three distinct legal regimes. Many vendors handle two-hop chains competently; three-to-five-hop chains across jurisdictions are where coverage thins materially.

UBO determination logic covers how the vendor applies the 25% ownership-or-control threshold and how it handles ambiguity. When ownership percentages aren't disclosed, or when a nominee structure complicates the chain, the vendor must have documented logic for what it outputs and what confidence score it assigns. This logic should be available for audit review.

Evidence and auditability is non-negotiable for regulated enterprises. The vendor must link each element of the resolved chain to a source document, record the method used (direct register, derived from filing, or computed from graph), and timestamp each data point. Evidence must survive re-examination months or years later when regulators ask why a particular UBO was accepted. Platforms that build policy-driven compliance workflows with detailed audit trails treat this as the output standard, not an optional enhancement.

Shortlisted vendors: what they claim and what to probe

Kyckr positions its approach around automated identification of UBOs and discusses the challenges of navigating restricted EU beneficial-ownership registers in its KYC API documentation (updated November 2025). The relevant probe: for which EU member states does Kyckr access the dedicated beneficial-ownership register directly, versus reconstructing ownership chains from company registry filings? Ask for a country-by-country matrix with the access method labeled per jurisdiction.

Encompass claims automation of UBO verification and integration with data from "over 175 trusted global public registries" (Encompass UBO Verification page). Encompass's strength is orchestration: it acts as an automation layer that can traverse multi-registry chains rather than relying solely on a single data source. The probe here is depth versus breadth. How many EU jurisdictions have dedicated beneficial-ownership register connectivity (not just company registry), and what does the integration footprint look like for a customer's tech stack?

Sayari markets beneficial ownership chain resolution across "250+ jurisdictions" (Sayari KYC and Beneficial Ownership enterprise page). Sayari's architecture is graph-based, which gives it strong cross-border chain resolution capabilities, particularly useful where direct register access is constrained. The critical question is data freshness: Sayari's outputs likely reflect ingested data rather than live register pulls, so you need to establish the update cadence per jurisdiction and how source provenance is recorded in the output.

Moody's (via its UBO Due Diligence Platform / Orbis ecosystem) positions a dedicated automated UBO discovery product for identifying corporate and shareholder structures of legal entities. Moody's strength is scale, particularly for larger-entity portfolios where Orbis ownership data has historically had strong EU corporate coverage. Probe for: timeliness of EU UBO register data versus Orbis ownership links, the API/integration model, and whether EU-specific UBO register fields (as distinct from general shareholder data) are sourced from registers or inferred.

LSEG World-Check One "UBO Check" explicitly combines beneficial ownership data sourced from Dun & Bradstreet or Sayari with World-Check risk intelligence (LSEG UBO Check product page). The integrated risk-plus-UBO workflow is genuinely useful if your RFP covers both UBO discovery and sanctions/PEP screening in a single output. The important caveat: LSEG's UBO Check is architecturally an enrichment and check layer sitting on top of underlying data providers. If your RFP requires direct EU register evidence, you need to trace that requirement to the underlying provider (Sayari or D&B) and confirm their EU coverage, not just LSEG's wrapper.

A note on ComplyAdvantage, which appears frequently in compliance RFPs: its core capability is sanctions, PEP, and adverse media screening. It is a downstream screening platform, not a primary EU UBO-register discovery engine. Evaluating it against an "automated UBO discovery" requirement will produce a capability mismatch. The correct framing is whether your chosen UBO discovery vendor integrates cleanly into ComplyAdvantage (or your equivalent screening platform) for post-discovery risk checks. Distinguishing UBO discovery from watchlist screening is one of the most common clarity gaps in RFPs of this type.

Dow Jones Risk & Compliance is worth noting separately: it sources its beneficial ownership data through a Dun & Bradstreet partnership and delivers it via risk data feeds and APIs. Like LSEG's UBO Check, this is an aggregated data model rather than a live-register-access model. Useful for broad portfolio enrichment; less suited to requirements demanding primary-document evidence per EU jurisdiction.

RFP evaluation matrix

Use these eight dimensions as your scoring rubric. Each should be answered per-jurisdiction, not in aggregate.

Dimension

What to score

Jurisdiction coverage

Country-by-country matrix for all 27 EU member states; register type per country (company registry vs. dedicated UBO register)

Access model

Direct register, derived from filings, or graph/cached; LIA/obliged entity requirements per country

Automation and integration

REST API specification, response schema for ownership chain output, rate limits, bulk processing capacity

Evidence and audit trail

Source document linkage per hop, method tag, timestamps, re-verification behavior

Ownership logic transparency

25% threshold application, control-clause handling, confidence scoring, ambiguity documentation

Data quality

Entity match rate, name standardization across languages, false-positive rate

Operational KPIs

Completion rate, fallback rate, average resolution time, analyst time saved per case

Security and compliance

Data handling, logging, retention periods, and whether the vendor's outputs are auditability-ready for your regulator

The jurisdictions where direct beneficial-ownership register access exists versus where it's constrained should be treated as separate scoring pools. A vendor with strong coverage across 20 EU states via company registries but no UBO-specific register access in Germany, France, or the Netherlands carries a material gap for many regulated enterprise portfolios.

What to request in vendor demos and questionnaires

A demo that only shows the vendor's prepared example set is not sufficient. For automated UBO discovery, the demo must run against your actual entity types.

Prepare a demo entity set that includes: a multi-layer holding structure (at least three hops), at least one EU-to-non-EU ownership route (e.g., a Dutch operating company held through a UK or Cayman parent), and at least one known edge case from your portfolio (a nominee arrangement, a foundation, or a structure with no disclosed UBO percentage).

For each entity in the set, ask the vendor to produce: (a) the resolved UBO list with percentages, (b) the full ownership path from operating company to natural person, (c) the evidence document or source record for each hop, (d) the timestamp for each data element, and (e) a written explanation for any hops where data is missing or estimated.

Separately, ask for a quantified coverage report for your target EU countries. "We cover the EU" is not an answer. The acceptable answer looks like: "For the 27 EU member states, here is our access method, our data update frequency, and our fallback approach per country, with a completion rate from a comparable client portfolio in the last 90 days."

Teams building EU corporate onboarding workflows with AI-assisted ownership resolution are already using this level of specificity as their baseline vendor requirement, not an aspirational one.

Choosing the right vendor for your specific risk posture

There's no universal answer, because the right vendor depends on what your compliance function and your regulators actually require.

If your primary obligation is maximum evidential quality with strict auditability, the selection priority should be vendors who can demonstrate primary document access per jurisdiction, with an explicit access model that ties each data output to a source. Encompass and Kyckr are most aligned with this model for the EU, but both require rigorous jurisdiction-by-jurisdiction confirmation during procurement.

If your primary obligation is broad cross-border coverage with fast resolution, particularly for portfolios that include both EU and non-EU entities in the same ownership chain, the graph-based approach from Sayari is the stronger fit. The trade-off is that you must establish acceptable evidence provenance standards in the contract and validate them during the demo.

If you need UBO discovery and risk screening in a single integrated workflow, the LSEG UBO Check or a Moody's Orbis solution may reduce integration complexity. The critical verification step remains the same: confirm whether the beneficial-ownership data underpinning those platforms satisfies your EU register evidence requirements, or whether you're getting aggregated data that needs supplementary primary-document collection.

Whatever architecture you select, the orchestration layer connecting UBO discovery to your downstream compliance workflows matters as much as the discovery capability itself. Platforms like Duna are built precisely for this: connecting data providers across 210+ local registries, routing evidence through policy-driven workflows, and producing audit-ready outputs with traceable sources, without requiring you to manually stitch together each vendor's API response into a compliant case file. The KYB/UBO onboarding flow sits on top of the discovery layer, not beside it, which means evidence gaps surface as workflow exceptions rather than silent compliance failures.

The practical takeaway is this: write your RFP to require a jurisdiction-by-jurisdiction evidence matrix, an explicit access model declaration, and a live demo against your entity set. Any vendor that can satisfy all three with specificity has cleared the first filter. Then use the evaluation matrix above to separate the shortlist on evidence quality, integration fit, and operational performance.

Procurement teams writing an RFP for "automated UBO discovery across EU jurisdictions" routinely get responses from a dozen vendors, all of whom claim to support it. Most of those claims deserve significant scrutiny. The EU's post-2022 beneficial-ownership access environment is genuinely constrained, and the phrase "automated UBO discovery" can mean very different things depending on how a vendor has architected its data pipeline. This article gives you the evaluative precision you need to tell them apart, select a defensible shortlist, and structure an RFP that elicits honest answers.

What "automated UBO discovery" actually means (and why the definition matters for your RFP)

Before you can evaluate a vendor, you need to pin down what your organization means by "discovery." There are three architecturally distinct approaches vendors use, and they carry different accuracy, freshness, and auditability implications.

Direct beneficial-ownership register retrieval means the vendor queries the national UBO register for a given jurisdiction and returns structured data sourced directly from that register. This is the highest-evidential-quality model, but it's also the most constrained by access rules, as discussed in the next section.

Registry-derived ownership chain reconstruction means the vendor queries underlying company registers (not the dedicated UBO register), retrieves shareholder and director filings, then applies ownership logic recursively until a natural-person UBO is resolved. Evidence comes from company filings rather than UBO registers directly. Many vendors in the EU operate primarily this way.

Aggregated graph mapping means the vendor has pre-ingested large volumes of corporate data, built an ownership graph, and resolves UBO chains from that graph at query time. Coverage is broad and latency is low, but freshness depends on ingestion cadence, not live-register access.

Your RFP must specify which model (or combination) you require. It must also define the expected deliverables explicitly. A UBO name and percentage alone is insufficient for most regulated-enterprise use cases. Auditors and regulators typically expect the full ownership path, percentage per hop, source documents per hop, timestamps for each data point, and the logic used to apply the 25% ownership or control threshold. If you also need a REST API or webhook integration for automated workflow triggering, that's a separate specification from what data is produced.

Understanding these distinctions is also the foundation for automated compliance decisioning workflows, where the quality of inputs into an ownership resolution step determines how much analyst escalation follows.

EU register access constraints every vendor claim must be tested against

No vendor can honestly claim fully open, automated, API-level access to every EU member state's beneficial-ownership register. The legal environment makes that impossible.

The BORIS (Beneficial Ownership Register Interconnection System), as described by European e-Justice (June 29, 2026), is an EU interconnection tool that links national central registers containing information on beneficial owners. BORIS improves discoverability across borders, but it doesn't grant unlimited programmatic access. Each member state's register governs its own access rules.

The 2022 CJEU ruling fundamentally changed the landscape. Before it, many EU member state registers allowed broad public access to UBO data. The court's decision restricted that access on privacy grounds, and member states have progressively tightened their rules since. The practical result is that most beneficial-ownership register access now requires demonstrated "legitimate interest" (LIA) or a classification as an "obliged entity" (a bank, fintech, law firm, or other AML-regulated business). Some registers have tiered access with materially different data fields per tier.

The Netherlands is a concrete example. KVK (the Netherlands Chamber of Commerce) describes three access tiers: UBOB1, UBOB2, and UBOB3. Each tier unlocks different fields, and the afschermen (shielding) rules mean that certain UBO records can be partially hidden for privacy protection reasons. A vendor claiming "full Netherlands UBO register access" needs to specify exactly which access tier they operate under and what shielding scenarios affect their output.

The implications for your RFP are direct. You need to ask each vendor:

  • What access mechanism do you use per EU member state (direct register, company registry, or graph)?

  • What obliged-entity or legitimate-interest credentials are required from the customer, and who establishes eligibility?

  • For countries where direct access is restricted, what is your fallback strategy, and how is the fallback clearly flagged in the output?

Vendors that cannot answer these questions with specificity for your target jurisdiction list are signaling a gap you should not accept at contract signature.

How UBO chain resolution should be evaluated

Once you've established access model and jurisdiction coverage, the evaluation moves to the technical quality of ownership chain resolution. Four dimensions matter most.

Entity resolution quality is the baseline. Before any ownership traversal begins, the vendor must correctly identify which corporate entity in a registry matches the entity you submitted. Weak entity matching, especially across multilingual name formats (German, Polish, and Romanian company names in the same portfolio, for instance), generates missed connections and false UBO attributions. Ask vendors for their match rate on your specific entity population, not a generic benchmark.

Recursive traversal depth determines whether the vendor can follow ownership chains across multiple hops, including chains that cross jurisdictions. A Dutch holding company owned by a Luxembourg SOPARFI owned by a Cayman trust requires the vendor to traverse three distinct legal regimes. Many vendors handle two-hop chains competently; three-to-five-hop chains across jurisdictions are where coverage thins materially.

UBO determination logic covers how the vendor applies the 25% ownership-or-control threshold and how it handles ambiguity. When ownership percentages aren't disclosed, or when a nominee structure complicates the chain, the vendor must have documented logic for what it outputs and what confidence score it assigns. This logic should be available for audit review.

Evidence and auditability is non-negotiable for regulated enterprises. The vendor must link each element of the resolved chain to a source document, record the method used (direct register, derived from filing, or computed from graph), and timestamp each data point. Evidence must survive re-examination months or years later when regulators ask why a particular UBO was accepted. Platforms that build policy-driven compliance workflows with detailed audit trails treat this as the output standard, not an optional enhancement.

Shortlisted vendors: what they claim and what to probe

Kyckr positions its approach around automated identification of UBOs and discusses the challenges of navigating restricted EU beneficial-ownership registers in its KYC API documentation (updated November 2025). The relevant probe: for which EU member states does Kyckr access the dedicated beneficial-ownership register directly, versus reconstructing ownership chains from company registry filings? Ask for a country-by-country matrix with the access method labeled per jurisdiction.

Encompass claims automation of UBO verification and integration with data from "over 175 trusted global public registries" (Encompass UBO Verification page). Encompass's strength is orchestration: it acts as an automation layer that can traverse multi-registry chains rather than relying solely on a single data source. The probe here is depth versus breadth. How many EU jurisdictions have dedicated beneficial-ownership register connectivity (not just company registry), and what does the integration footprint look like for a customer's tech stack?

Sayari markets beneficial ownership chain resolution across "250+ jurisdictions" (Sayari KYC and Beneficial Ownership enterprise page). Sayari's architecture is graph-based, which gives it strong cross-border chain resolution capabilities, particularly useful where direct register access is constrained. The critical question is data freshness: Sayari's outputs likely reflect ingested data rather than live register pulls, so you need to establish the update cadence per jurisdiction and how source provenance is recorded in the output.

Moody's (via its UBO Due Diligence Platform / Orbis ecosystem) positions a dedicated automated UBO discovery product for identifying corporate and shareholder structures of legal entities. Moody's strength is scale, particularly for larger-entity portfolios where Orbis ownership data has historically had strong EU corporate coverage. Probe for: timeliness of EU UBO register data versus Orbis ownership links, the API/integration model, and whether EU-specific UBO register fields (as distinct from general shareholder data) are sourced from registers or inferred.

LSEG World-Check One "UBO Check" explicitly combines beneficial ownership data sourced from Dun & Bradstreet or Sayari with World-Check risk intelligence (LSEG UBO Check product page). The integrated risk-plus-UBO workflow is genuinely useful if your RFP covers both UBO discovery and sanctions/PEP screening in a single output. The important caveat: LSEG's UBO Check is architecturally an enrichment and check layer sitting on top of underlying data providers. If your RFP requires direct EU register evidence, you need to trace that requirement to the underlying provider (Sayari or D&B) and confirm their EU coverage, not just LSEG's wrapper.

A note on ComplyAdvantage, which appears frequently in compliance RFPs: its core capability is sanctions, PEP, and adverse media screening. It is a downstream screening platform, not a primary EU UBO-register discovery engine. Evaluating it against an "automated UBO discovery" requirement will produce a capability mismatch. The correct framing is whether your chosen UBO discovery vendor integrates cleanly into ComplyAdvantage (or your equivalent screening platform) for post-discovery risk checks. Distinguishing UBO discovery from watchlist screening is one of the most common clarity gaps in RFPs of this type.

Dow Jones Risk & Compliance is worth noting separately: it sources its beneficial ownership data through a Dun & Bradstreet partnership and delivers it via risk data feeds and APIs. Like LSEG's UBO Check, this is an aggregated data model rather than a live-register-access model. Useful for broad portfolio enrichment; less suited to requirements demanding primary-document evidence per EU jurisdiction.

RFP evaluation matrix

Use these eight dimensions as your scoring rubric. Each should be answered per-jurisdiction, not in aggregate.

Dimension

What to score

Jurisdiction coverage

Country-by-country matrix for all 27 EU member states; register type per country (company registry vs. dedicated UBO register)

Access model

Direct register, derived from filings, or graph/cached; LIA/obliged entity requirements per country

Automation and integration

REST API specification, response schema for ownership chain output, rate limits, bulk processing capacity

Evidence and audit trail

Source document linkage per hop, method tag, timestamps, re-verification behavior

Ownership logic transparency

25% threshold application, control-clause handling, confidence scoring, ambiguity documentation

Data quality

Entity match rate, name standardization across languages, false-positive rate

Operational KPIs

Completion rate, fallback rate, average resolution time, analyst time saved per case

Security and compliance

Data handling, logging, retention periods, and whether the vendor's outputs are auditability-ready for your regulator

The jurisdictions where direct beneficial-ownership register access exists versus where it's constrained should be treated as separate scoring pools. A vendor with strong coverage across 20 EU states via company registries but no UBO-specific register access in Germany, France, or the Netherlands carries a material gap for many regulated enterprise portfolios.

What to request in vendor demos and questionnaires

A demo that only shows the vendor's prepared example set is not sufficient. For automated UBO discovery, the demo must run against your actual entity types.

Prepare a demo entity set that includes: a multi-layer holding structure (at least three hops), at least one EU-to-non-EU ownership route (e.g., a Dutch operating company held through a UK or Cayman parent), and at least one known edge case from your portfolio (a nominee arrangement, a foundation, or a structure with no disclosed UBO percentage).

For each entity in the set, ask the vendor to produce: (a) the resolved UBO list with percentages, (b) the full ownership path from operating company to natural person, (c) the evidence document or source record for each hop, (d) the timestamp for each data element, and (e) a written explanation for any hops where data is missing or estimated.

Separately, ask for a quantified coverage report for your target EU countries. "We cover the EU" is not an answer. The acceptable answer looks like: "For the 27 EU member states, here is our access method, our data update frequency, and our fallback approach per country, with a completion rate from a comparable client portfolio in the last 90 days."

Teams building EU corporate onboarding workflows with AI-assisted ownership resolution are already using this level of specificity as their baseline vendor requirement, not an aspirational one.

Choosing the right vendor for your specific risk posture

There's no universal answer, because the right vendor depends on what your compliance function and your regulators actually require.

If your primary obligation is maximum evidential quality with strict auditability, the selection priority should be vendors who can demonstrate primary document access per jurisdiction, with an explicit access model that ties each data output to a source. Encompass and Kyckr are most aligned with this model for the EU, but both require rigorous jurisdiction-by-jurisdiction confirmation during procurement.

If your primary obligation is broad cross-border coverage with fast resolution, particularly for portfolios that include both EU and non-EU entities in the same ownership chain, the graph-based approach from Sayari is the stronger fit. The trade-off is that you must establish acceptable evidence provenance standards in the contract and validate them during the demo.

If you need UBO discovery and risk screening in a single integrated workflow, the LSEG UBO Check or a Moody's Orbis solution may reduce integration complexity. The critical verification step remains the same: confirm whether the beneficial-ownership data underpinning those platforms satisfies your EU register evidence requirements, or whether you're getting aggregated data that needs supplementary primary-document collection.

Whatever architecture you select, the orchestration layer connecting UBO discovery to your downstream compliance workflows matters as much as the discovery capability itself. Platforms like Duna are built precisely for this: connecting data providers across 210+ local registries, routing evidence through policy-driven workflows, and producing audit-ready outputs with traceable sources, without requiring you to manually stitch together each vendor's API response into a compliant case file. The KYB/UBO onboarding flow sits on top of the discovery layer, not beside it, which means evidence gaps surface as workflow exceptions rather than silent compliance failures.

The practical takeaway is this: write your RFP to require a jurisdiction-by-jurisdiction evidence matrix, an explicit access model declaration, and a live demo against your entity set. Any vendor that can satisfy all three with specificity has cleared the first filter. Then use the evaluation matrix above to separate the shortlist on evidence quality, integration fit, and operational performance.