USPTO Trademark Search: bulk phrase lookup (live marks, pay per phrase)

Check up to 500 trademark phrases per run against a snapshot of live (pending and registered) US federal trademarks built from official USPTO bulk data. $0.002 per phrase checked, no start fee. A record lookup, not legal advice or trademark clearance. Ideal for print-on-demand and merch phrase screening, brand-name brainstorming, and bulk checks for sellers. Runs on Apify schedules and integrations. The index is refreshed daily from new USPTO filings.

Use cases

• Print-on-demand and merch phrase screening before you commit to a design
• Brand-name brainstorming: check many candidate names in one run
• Bulk checks for sellers against live US federal marks
• Filtering candidates by Nice class (e.g. 25 = clothing) before deeper review

What this is (and is not)

A record lookup: it searches live federal marks in the USPTO's official bulk data and tells you what matches exist and their status. The snapshot date is in every result (indexSnapshot), so you always know how fresh the data is.

It is not legal advice and not a trademark clearance. State-law marks, common-law rights, dead (abandoned/expired/cancelled) marks and foreign marks are not included, and similar-sounding or similar-looking marks are not detected. For a real clearance, have a trademark attorney review the results.

What you get (one dataset item per phrase)

phrase, status (ok or invalid_input), matchCount, totalMatches (reported as 10000+ when capped), totalMatchesCapped, and matches[] — each match has serial, registrationNumber, mark, statusCode, statusCategory, statusDescription, filedDate, registeredDate, classes[], and a tsdrUrl deep link to the official USPTO TSDR status page. Every item also carries indexSnapshot and the disclaimer text.

Matching and filters

• matchMode: words (default) — all words of the phrase must appear in the mark, in any order; exact — only marks identical to the phrase.
• classes — optional Nice-class filter (1-45, e.g. 25 = clothing); a match must be listed in at least one of the given classes.
• maxMatchesPerPhrase — how many matches to include per phrase (default 25, max 100); totalMatches still reports the full count.

Pricing: pay per phrase checked

One event per phrase checked successfully (status: ok, $0.002), no matter how many matches it has. Invalid phrases are not charged, and there is no start fee. The run checks your spending limit before each phrase and stops cleanly if it is reached.

Limits

• Max 500 phrases per run (duplicates removed); up to 100 matches returned per phrase.
• Data is a snapshot of the USPTO bulk files, not a live API; the snapshot date is in every result.

FAQ

Is this legal advice or a trademark clearance? No. It is a record lookup of live federal marks. State-law, common-law, dead and foreign marks are not included, and similar-sounding or similar-looking marks are not detected. For a real clearance, have a trademark attorney review the results.

Which marks are included? Live (pending and registered) US federal trademarks from official USPTO bulk data.

How fresh is the data? It is a snapshot of the USPTO bulk files, not a live API. The snapshot date (indexSnapshot) is in every result, so you always know how fresh it is.

Will it find similar-sounding or similar-looking marks? No — matching is words (all words of the phrase appear in the mark, in any order) or exact (identical marks) only.

Can I filter by product class? Yes — the optional classes filter takes Nice classes 1-45 (e.g. 25 = clothing); a match must be listed in at least one of the given classes.

What does it cost? $0.002 per phrase checked successfully, no start fee; invalid phrases are not charged.

Input example

{
  "phrases": ["just do it", "gold core"],
  "classes": ["25"],
  "matchMode": "words",
  "maxMatchesPerPhrase": 25
}