Independent service. Not affiliated with Companies House or GOV.UK.

Official register: Find and update company information

Transparency

Data sources

Every figure on this site comes from a named source with a recorded check time. Here is exactly what we use and how we handle it.

Company records

Company details are retrieved from the Companies House Public Data API at https://api.company-information.service.gov.uk. For this release we use three endpoints:

  • GET /search/companies — company name and number search.
  • GET /company/{companyNumber} — the company profile.
  • GET /company/{company_number}/filing-history — recent filings.

Requests are authenticated server-side with HTTP Basic authentication, using our API key as the username and an empty password. The key is stored as a server secret. It never appears in browser code, in public environment variables or in logs.

Where the API offers advanced search, we would only use the filters its specification actually supports. We do not filter one page of fetched results and describe it as a filter across the whole register.

SIC codes

SIC descriptions come from the condensed SIC 2007 list published by Companies House at resources.companieshouse.gov.uk/sic. Each SIC page shows the dataset label and the date our editor checked it. Our interpretation of a code is always presented separately from the official description.

Official records and further datasets

Caching, freshness and request budget

  • Search results are cached for about 10 minutes; profiles and filing history for about 24 hours. Both are configurable.
  • Identical simultaneous requests are shared rather than repeated, and searches are debounced.
  • We keep a shared request budget well below the documented limit of 600 requests per five minutes, counting all server activity and any scheduled jobs. We back off when the service returns a rate-limit response, and we never rotate keys to work around a limit.
  • Every page shows the actual time we checked the record. If a refresh fails, we serve the cached copy and label it as potentially out of date. We never present a failed refresh as a fresh check.

Monthly snapshots versus a live check

A monthly CSV snapshot is a different thing from an API check, and we label them differently. A snapshot describes the register at the point it was published, it does not necessarily contain every historical or dissolved company, and it is not a substitute for the current record. Bulk import is designed as a resumable background job run by an administrator, never a browser download or a long web request, and we do not attempt to crawl the whole register through the API.

What is not in our data

Turnover, profit, cash position, credit ratings, risk scores, regulatory approvals, staff numbers and reviews are not part of the basic public record, so they do not appear here. Officer and person-with-significant-control enrichment is outside the current scope, and we do not publish person-search pages.

Corrections

Register facts can only be changed by the company at the official source. If our presentation of a record is wrong, please tell us and we will fix it and record the change.

Reuse and personal data

Public availability does not remove obligations around personal data and reuse. We limit what we publish to registered company facts, and we ask users to do the same when reusing anything from this site.

Page path: /data-sources. Independent service. Not affiliated with Companies House or GOV.UK.