Sanctions list data: dates of birth, names and IDs

What information do the lists contain?

Sanctions lists can give you more than a name. We looked at which fields contain details and how those details are written. The study covers 19,452 OFAC SDN records and 6,339 UK designation records captured on 30 September 2026.

We count source records, not unique people or organizations. The files cover different populations and regimes. Their percentages do not rank the two sources.

Among individual records, 13.21% in the OFAC snapshot and 30.95% in the UK snapshot had no full calendar value in their structured date-of-birth fields. Most OFAC records in that group still had a partial date or another non-full value.

The UK source also treats aliases and primary-name variations as separate categories. Keeping both categories changes which names an integration can show a reviewer.

This study measures structured-field availability at one point in time, not screening accuracy. Download the research package for the report, methodology, aggregate data, source hashes, analysis code and figures. It contains no original source records.

A populated birth-date field can still be partial

We put each individual record into one category after checking all its structured date-of-birth values. A full calendar value must be a valid day, month and year, with nothing else in that value. A year, range or approximate date does not qualify. This describes the field format. It does not prove that the date is certain or correct.

A record with a year-only value and a full date goes in the full-calendar category. A year alone can still help a reviewer. Treating it as an exact birthday invents detail; treating it as empty loses information. That is an implication for software design, not a measured improvement in matching.

Name categories change the answer

The UK source separates aliases from variations of a primary name. Of its 4,032 individual records, 761 had an alias and 1,248 had a primary-name variation. The groups overlap: 98 records had both. Counting each record once gives 1,911 records, or 47.40%. Adding the two group totals would count that overlap twice.

In that population, 1,150 individual records had a name variation but no alias. An integration that reads only the alias category would miss those variations. This counts available names that the integration would omit. It does not estimate missed sanctions matches.

OFAC's legacy XML has a different alias structure. At least one populated alias appeared on 4,298 individual records (56.81%), including entries marked a.k.a., f.k.a. or n.k.a. That is not the same category definition as the UK alias field.

Document fields need equally careful labels

A passport-number field had a value on 2,103 OFAC individual records (27.80%) and 587 UK individual records (14.56%). This does not count every type of identity document. It does not validate the number, show that a passport is current or mean that other identifiers are absent.

In particular, OFAC's general idList also carries attributes such as gender, websites and sanctions information. Its presence is not a sound substitute for counting passport entries. The analysis selects passport-typed entries with a populated number, as specified in the methodology.

When you assess sanctions list data quality for an integration, keep source labels and partial values. Empty, partial, multiple and fully formatted values mean different things. Your software and reviewers need to handle each state.

For a practical implementation, use the OFAC API integration guide. It connects source selection, partial identity information and potential-match review to explicit application outcomes, with a Python example for testing failure handling.

Methodology and evidence

We examined all records in two official XML files. This is a census, not a sample or a screening benchmark. The OFAC file reports a publication date of 30 September 2026. The UK file reports a generation date of 29 September 2026. Both were retrieved on 30 September at about 19:19:59 UTC. A file date does not mean that every record changed on that date.

Each metric counts records with at least one qualifying value. We divide that count by all records of the relevant source and entity type. We handle whitespace and explicit placeholder values consistently. We do not extract extra facts from free-text remarks, match or deduplicate records across sources, or calculate a combined quality score.

Read the complete methodology, aggregate CSV, JSON results and source manifest. The analysis code verifies pinned SHA-256 hashes before calculating. Synthetic parser tests cover partial dates, invalid dates, multiple DOB values, placeholders and field typing.

The original official files are kept privately for audit and are not redistributed here. A download URL can return new content later. Exact reproduction needs the matching snapshot bytes. The public hashes can verify an input version, but they do not supply historical data. The method is reproducible with the pinned private evidence archive; historical raw files are not publicly hosted.

Use and cite this research

Suggested citation: SanctionsKit (2026), What structured sanctions records tell us about dates and names, 30 September 2026. Link to this report and identify the snapshot date when quoting a finding. Keep source names, record denominators and field definitions with any percentages.

The complete research package includes editable SVG figures, PNG figures, the report and reproduction files. Individual PNG figures are available as DOB precision and UK name categories. Preserve the figure attribution and its limitations.

Contains public sector information licensed under the Open Government Licence v3.0. Analysis by SanctionsKit. Independent work; no government endorsement. Original records, narratives, logos and seals are not republished.

These findings do not measure source correctness, list completeness, screening accuracy, false positives, false negatives, legal compliance, ownership restrictions or SanctionsKit's production coverage. One snapshot per source cannot establish a trend. Read the coverage limitations before applying list-screening results to a real-world decision.