USPTO trademark search
A faster, smarter way to search USPTO trademark records. Qevatrix Trademark OS indexes live and historical USPTO filings and applies phonetic, fuzzy and dominant-word matching so you do not miss the refusal that a basic TESS string search would leave behind.
Why the official TESS search is only a starting point
The USPTO Trademark Electronic Search System (TESS) is authoritative, but it is designed for examiners, not brand owners. An exact string search misses marks that sound the same, that share only the distinctive word, or that cover related goods in a neighbouring class. Applications are refused over those marks every day.
Qevatrix Trademark OS is built to catch the marks that block a filing. It normalizes, phoneticizes and compares the dominant portion of each mark, then scores every result so you can decide what to do next.
How the search finds conflicts TESS can miss
Exact and normalized matches
Marks are normalized the way USPTO examiners read them — punctuation, spacing, casing and common design wording stripped — so ACME-CO, Acme Co. and ACMECO land on the same result.
Phonetic matches
Marks that sound alike are surfaced even when spelling differs: KWIKSET against QUICKSET, VERIDIAN against VIRIDIAN. Sound-alike marks are a routine basis for USPTO refusal, so a basic TESS string search is not enough.
Fuzzy and misspelling matches
Edit-distance matching catches inserted, dropped and transposed characters — the space where deliberately tweaked marks and typo variants hide.
Dominant-word analysis
Generic and descriptive wording is discounted so the distinctive portion of a mark drives the comparison. A search for a two-word mark still finds a registration that shares only the distinctive half.
Conflict Score™: every hit scored factor by factor
Each result carries a Conflict Score built from four factors, each shown separately so you can see why a hit ranked where it did rather than trusting a single number:
Mark similarity
How close the marks are in appearance, sound and overall commercial impression, weighted by the distinctive portion.
Goods and services overlap
The wording actually claimed in each filing is compared, not just the Nice class number — two marks in the same class can serve entirely different buyers.
Class relatedness
Classes that are commonly treated as related are scored as such, so a conflict in an adjacent class is not silently dropped.
Registration strength
A live, incontestable registration on a coined mark weighs more heavily than a descriptive mark or a lapsed filing.
The score is a triage tool for a result list — it ranks and explains risk. It is not a legal opinion on likelihood of confusion.
Three search depths, from quick to attorney-ready
Results can be filtered by registration status and by owner — useful when you already know which company's marks you need to exclude — and any screening exports as a branded clearance report PDF with the conflicts, risk levels and score breakdowns included.
Where the USPTO data comes from
The searchable index is built from USPTO's public trademark data services and refreshed on a schedule, covering live and dead records — a dead registration still tells you a name was used in your field, and a recently abandoned one can be refiled by its former owner. Case status and documents on marks you docket come from USPTO's own status and document system.
Results are informational. Always confirm what you find against the official USPTO record: Qevatrix Trademark OS is software, not a law firm, and does not provide legal advice or representation.
Run a USPTO trademark search
Searching is free to start, with no credit card. If the mark is clear you can move straight into a guided filing package for a word mark or a logo; if it is not, you have just avoided a rebrand. Costs are broken down on the pricing and fees page, and common questions are answered in the FAQ.