Methodology

How the data on this site is sourced, parsed, normalised, and refreshed.

Primary sources

  • USPTO Trademark Bulk Data - published daily as XML zip files at bulkdata.uspto.gov. Each daily file is a delta against the previous day's record set. We download, decompress, and SAX-parse the XML stream to extract approximately twenty fields per mark (name, serial number, status, filing date, registration date, Nice classification, owner name and country).
  • EUIPO Open Data Portal - weekly bulk CSV exports plus a REST API for incremental polls. EU trademark records are simpler to parse than USPTO (flat CSV rather than nested XML), refreshed once a week.

Status normalisation

Both offices use different vocabularies for trademark status. We map them to a single normalised set:

  • registered - currently in force
  • pending - application filed, examination in progress
  • abandoned - application withdrawn or rejected before registration
  • cancelled - previously registered, removed for non-renewal or other cause
  • expired - past the renewal deadline without renewal

The raw office-issued status is retained on every record so audit trails remain clean.

Refresh process

  1. A scheduled job runs daily at 03:15 UTC to fetch the latest USPTO daily file.
  2. A weekly job at 04:30 UTC on Mondays pulls the latest EUIPO bulk export.
  3. Both ingestors stream-parse the source and write to a new staging table.
  4. Indexes are built on the staging table after the bulk write (faster than incremental updates during load).
  5. A single atomic ALTER TABLE … RENAME swaps the staging table into place, replacing the previous live table.
  6. The previous table is dropped in the same script - no orphaned _old tables left behind.
  7. Per-owner aggregates and per-Nice-class counts are recomputed after the swap.

Cross-referencing UK Companies House

For marks owned by UK-registered companies, we attempt to match the owner's name and country code against the UK Companies House register and, on a confident match, link directly to the corresponding company profile on ukcompanyfinder.com. Confidence comes from a combination of exact name match (after standard suffix stripping) and a country-code restriction.

Known limitations

  • The site does not currently surface trademark images or logos. The text mark, owner, classification, and status are shown; the actual logo is not. This is a deliberate decision pending a review of image licensing across both offices.
  • WIPO Madrid Protocol designations that target jurisdictions other than the US or EU are not included. Madrid designations into the US or EU are reflected in USPTO/EUIPO data and so do appear here.
  • Owner address details are limited to country code. Full address blocks are intentionally not displayed to limit personal-data exposure.
  • Office-internal events (Office Actions, prosecution timelines) are retained in JSONB metadata but are not currently surfaced on the live site.

Corrections

If you spot an error in our presentation of the data, email contact@worldtrademarksearch.com. We'll fix or remove the issue within 48 hours. For changes to the underlying record (status, owner, classification), file with USPTO or EUIPO directly; our next refresh picks up corrections automatically.

More