The UBO Investigation Handbook:
Finding the Real Owner

Checking a register isn't investigating an ownership structure.

Every AML guide defines a beneficial owner: a natural person who ultimately owns or controls an entity, usually above a 25% threshold. Almost none teach you how to actually find one when the paperwork is designed to obscure them. This guide covers the practical investigative technique, what the registers genuinely show you, where they mislead by default, and how nominee arrangements, layered structures, and trusts are actually unwound in practice.

One thing worth knowing before anything else: the landscape changed in a way a lot of existing AML content hasn't caught up with. On 22 November 2022, the Court of Justice of the European Union struck down the EU's requirement for unrestricted public access to beneficial ownership registers, ruling it violated fundamental privacy rights. Reaction was fast: Luxembourg, Austria, Belgium, the Netherlands, and Malta among others restricted or suspended public access within days, and most other member states followed within the following months. If your mental model is "just look up the foreign UBO register," that model is largely out of date, most EU registers now require demonstrating a legitimate interest, and the picture keeps shifting, Italy's register, for example, spent years suspended after the ruling and only began a phased reopening in 2026 under a new tiered access model, following a further, Italy-specific CJEU judgment and implementing decree. Confirm the current position country by country before relying on any assumption here.

Before going further, it's worth being precise about who actually qualifies as a UBO, because "more than 25%" isn't a universal rule the way a lot of training treats it. FATF itself doesn't mandate a specific percentage, its guidance mentions 25% only as an illustrative example of how a threshold could work, not a fixed international standard. The UK applies 25% as the default. In the EU, the substantive beneficial-ownership definition now sits in Regulation (EU) 2024/1624, the directly applicable AML Regulation, not AMLD6 (Directive (EU) 2024/1640), which is a separate instrument governing national supervisory architecture, FIUs, and registers, and requires transposition into national law by 10 July 2027. The Regulation sets ownership and control as two separate limbs: Article 52 sets a 25% ownership or voting-rights threshold, Article 53 addresses control independently. A lower ownership threshold, down to a floor of 15% for higher-risk sectors, can only be set by a future European Commission delegated act under Article 52(2) and Article 85, following an EU-wide risk assessment due by 10 July 2029, individual member states cannot set their own lower threshold with ad hoc Commission approval. Thresholds vary meaningfully outside the UK and EU too, this is genuinely one to confirm in current local law rather than work from a fixed list, since these figures are actively revised by individual jurisdictions. The US Corporate Transparency Act still applies a 25%-plus-control test in principle, but a March 2025 FinCEN interim final rule, made final in August 2026, exempts domestic US entities and US persons from CTA reporting entirely, only entities formed under foreign law and registered to do business in a US state or tribal jurisdiction remain in scope. That control dimension matters beyond the US: a shareholder sitting just under whatever the applicable threshold happens to be, deliberately structured at 24.9% rather than 25%, but holding board appointment rights or an effective veto, may still be the beneficial owner in substance under a framework that includes an independent control limb, though that isn't a universal rule that applies regardless of the specific jurisdiction's own legal test, and a UBO analysis that stops at the percentage figure on the cap table can miss exactly the structuring pattern the threshold was designed to catch.

Four patterns, four visual memory cards

Most beneficial ownership misjudgments trace back to one of these four shapes. Learn to recognise them before the scenarios test them.

At a glance

Four patterns, the trap that makes each one look routine, the tell that actually gives it away, and the response that fits.

The register is where an investigation starts, not where it ends.

Every pattern in this guide comes back to the same discipline: a register lookup is a data point, not a conclusion. FinCrimeRadar's screening tool lets you check each individual beneficial owner you actually identify against live sanctions and PEP data, free, no account required.

Related topics

• Money Mule Networks, where this guide's nominee and shell-structure investigation technique applies directly to the Business Front pattern.
• The Onboarding Dilemma, where UBO identification sits inside the wider due diligence decision, KYB always contains KYC, and every UBO identified must be individually verified.