Similar company names, shared addresses and overlapping ownership details are making it harder for financial institutions to identify businesses accurately, increasing the risk of false positives and missed financial crime threats in anti-money laundering (AML) and Know Your Business (KYB) checks, according to ZIGRAM.
These businesses, often described as corporate doppelgängers, are legally distinct organisations that share several identifying characteristics. They may operate under nearly identical names, use the same registered address or have directors with matching surnames and similar industry classifications.
Such similarities can confuse automated screening tools and compliance teams, making it difficult to establish whether two records relate to the same business, connected organisations or entirely separate entities.
This is where entity resolution becomes important. The process establishes whether different records refer to the same real-world organisation and helps financial institutions maintain more accurate information across sanctions screening, transaction monitoring and customer risk assessments.
Without reliable entity resolution, errors can spread throughout compliance operations. Financial institutions may face unnecessary sanctions alerts, fragmented customer records and risk assessments based on incomplete or misleading information.
Entity resolution also serves a different purpose from identity verification and name screening. While identity verification checks whether documentation is authentic and name screening identifies potential matches against watchlists, entity resolution determines whether those matches genuinely relate to the same business.
To establish this, institutions need to assess multiple identifiers, including current and former company names, registration numbers, incorporation jurisdictions, registered addresses, tax identification numbers and Legal Entity Identifiers (LEIs). Director details and beneficial ownership records can provide further evidence when investigating potential connections.
The cost of getting these decisions wrong can be substantial. Poor data quality costs organisations an average of $12.9m annually, while one analysis found that 99.1% of resolved screening hits were false positives.
However, inaccurate matching creates risks beyond unnecessary investigations. Weak entity resolution can allow shell companies to pass onboarding checks or enable sanctioned businesses to avoid detection when their details appear differently across databases because of spelling variations or transliteration.
Company names alone offer limited evidence of identity. Generic terms such as Global Trading Solutions may be used by unrelated businesses in multiple jurisdictions, while differences in legal suffixes and non-Latin alphabets can make matching even more complicated.
Registered addresses can create similar problems. Corporate service providers and virtual offices may accommodate hundreds of unrelated businesses at one location. Nominee directors can also hold positions across multiple company boards without establishing a direct relationship between those organisations.
ZIGRAM highlights the issue through a hypothetical example involving Baltic Trade Solutions Ltd, registered in the UK, and Baltic Trade Solutions OÜ, based in Estonia. A sanctions alert linked to the UK company could be incorrectly attributed to its Estonian namesake because of their similar names and nearly identical director surnames.
Examining the underlying corporate records would reveal differences in registration authorities, jurisdictions, incorporation dates and directors. This additional information would help compliance teams identify the UK business as the relevant sanctions match while recognising the Estonian organisation as a separate entity.
Artificial intelligence could help institutions improve the accuracy of these decisions. A 2025 Federal Reserve study found that large language model-based matching reduced false positives by approximately 92% and increased detection rates by around 11% compared with traditional fuzzy matching techniques.
Nevertheless, a higher matching score should not be treated as definitive proof that two records represent the same organisation. Human oversight remains necessary when reviewing ambiguous cases and deciding whether further investigation is warranted.
ZIGRAM recommends that financial institutions use multiple attributes when matching entities, establish risk-based confidence thresholds and maintain clear records of data provenance. Documenting decisions involving uncertain matches and continuously monitoring changes to corporate information can also help strengthen the process.
Its Entity Hero platform uses resolved entity information to deliver configurable and explainable customer risk scoring. The approach highlights how accurate business identification underpins effective financial crime controls, as unreliable source data can undermine even sophisticated compliance systems.
As financial institutions increasingly automate AML and KYB processes, the ability to distinguish between similar-looking businesses is becoming central to reducing false positives, identifying genuine threats and maintaining reliable risk assessments.
Copyright © 2026 FinTech Global
Copyright © 2026 RegTech Analyst
