Documentation · Bulk data

Using the bulk data

Every collection we publish can be downloaded in full as a set of files, rebuilt from the register and published at fixed addresses. This guide helps you decide whether bulk files suit your work, which collection to take and where to read next.

The bulk files are free to download, with no account. Non-commercial use is covered by our data licence; commercial use needs a licence. To check names one at a time instead, use the screening API.

Should I use bulk data?

There are three ways to use our data. Pick the one that matches how many names you check and where your matching should run.

Bulk files give you every record at once, but the matching is then yours to build: name variants, transliteration, dates and identifiers all need handling on your side. If you only need to know whether a few thousand names appear on a list, the API is usually less work.

Understanding the data model

Each row in a source register becomes one record (a Person, Company, Organization, Vessel, Airplane or LegalEntity), with an id of the form crown-<number>. Its listing on that register is a separate Sanction entity, crown-<number>-sanction, carrying the register, programme, source link and listing date. When we have matched records from different registers as the same party, they share a canonical_id.

Every record carries one risk topic, taken from the list type of its register:

Read more in the model reference, entity structure, risk topics and statement data model. The same description is published as JSON at /downloads/meta/model.json, so a loader can read it.

Choosing a collection

A collection is a set of registers published together. Five are available as bulk files:

We recommend default. The smaller collections are subsets of it, so you lose nothing by taking the whole and filtering it yourself, and you keep one pipeline when your needs change. Filter on the topic or on the source register (the datasets field). For example, with jq:

# only politically exposed persons
jq -c 'select(.properties.topics // [] | index("role.pep"))' entities.ftm.json

# only records from one register, by its short identifier
jq -c 'select(.datasets | index("ofac_sdn"))' targets.nested.json

Only want PEPs? See Using PEP data.

The files hold only registers whose terms allow redistribution (see licence positions), so their totals are lower than the searchable figures on the site. For example, search holds 1,303,162 PEP records, the peps download —.

Registry reference data (company registers, beneficial ownership, securities identifiers) is not published as bulk files: see Know your business and Securities.

Data formats and updates

Each collection is published in several formats from the same build, so the record ids agree across files. Data formats compares them; Keeping data fresh explains the addresses, the build index and checksums; Delta files explains incremental updates.

The catalogue of all collections, with versions, figures and file sizes, is at /downloads/collections/index.json.

Licensing

Downloading needs no account and no key. Under our data licence, non-commercial use is free, provided you credit the source. Commercial use, such as screening your customers or building the data into a product, needs a commercial licence; see also free and non-commercial use. Each register's own terms prevail where they are stricter.

Wherever you reproduce the data, credit it in this form (from section 5.1 of the data licence):

Source: [Source Publisher name], compiled by Crown Intelligence (crownintelligence.net).
Data as of [date shown on record]. Verify against the original source before relying on this information.

Our legal texts are drafts awaiting review by Canadian counsel; the published versions will replace them.

Related: Crown Default · Sanctions · PEPs · Maritime · Know your business · Securities · Data enrichment

Last updated 2026-10-03