A trademark clearance search is the investigation you run before you commit to a name, to find out whether someone else already owns rights that your use would infringe. The question it answers is not “can I register this?” but the harder one: can I safely use this at all? Those are different questions with different answers, and confusing them is the most expensive mistake in branding.
Clearance is a discipline, not a database lookup. A search retrieves records. Clearance interprets them: it weighs how close the marks are, how related the goods are, how strong the senior mark is, and how likely an ordinary buyer is to be confused. The retrieval is the easy half.
The two are not interchangeable, and treating them as if they were is where most trouble starts.
A knockout search is a screen. You run your candidate through the USPTO’s Trademark Search system looking for identical or nearly identical marks in related classes, and you discard the candidates that come back bloodied. It takes minutes, costs nothing, and exists to stop you from falling in love with a name that was never available. What it cannot do is clear anything. Passing a knockout means only that the obvious conflict did not surface.
A full clearance search is an investigation. It expands well past exact matches into phonetic equivalents (KWIK and QUICK), alternate spellings, translations, marks that differ only in a weak or descriptive tail, and design elements coded by their visual content. It reaches outside the federal register into the sources described below. It then produces an availability opinion: a reasoned assessment of the risk of adopting the mark for specific goods in a specific market, usually sorting conflicts into clear, cautionary, and blocking tiers.
The opinion is the deliverable. A list of 400 hits with no analysis is not clearance, and vendors selling “comprehensive searches” as raw report dumps are selling the retrieval while leaving the judgment to you.
The USPTO’s Trademark Search tool, which replaced the long-running TESS system in November 2023, is authoritative for exactly one thing: federal registrations and pending applications. Three large categories of enforceable rights sit outside it entirely.
Common-law rights. This is the big one, and it surprises nearly everyone. In the United States, trademark rights arise from use in commerce, not from registration. A bakery that has sold under a name in Tucson since 2009 and has never filed a single form owns enforceable rights in that name in its trading area. Federal registration adds powerful things, including nationwide constructive use as of the filing date under 15 U.S.C. § 1057(c) and constructive notice to the world under § 1072, but it is not the source of the right. It is a layer on top of a right that use created.
State registrations. Each state runs its own register. These are cheap, common among small businesses, and absent from federal search results.
Foreign marks with U.S. reach. A mark used abroad generally does not create U.S. rights on its own, but foreign owners can claim priority into the United States under the Paris Convention and can file under sections 44 and 66 of the Lanham Act. Applications pending on those bases can be sitting ahead of you in line.
A professional clearance search therefore also pulls state registers, business name filings, domain registrations, app store listings, industry directories, trade press, and general web usage. The goal is to find people who are using the name, whether or not they ever told the government about it.
The governing principle is priority: the first user in a geographic area wins there, even against a later federal registrant. The doctrine comes from the Tea Rose-Rectanus line, Hanover Star Milling Co. v. Metcalf (1916) and United Drug Co. v. Theodore Rectanus Co. (1918), and it survives in codified form at 15 U.S.C. § 1115(b)(5), which preserves a limited-area defense for a good-faith prior user who began before the registrant’s filing.
The practical shape of this: you can obtain a federal registration, build a national brand on it, and still be unable to enter one metropolitan market because a senior local user got there first. Your registration is not invalid. It is just carved out. That carve-out can be exactly the market you needed.
Two cost curves, both of which get worse with time.
Rebranding. Everything downstream of the name gets rebuilt: packaging, signage, uniforms, the domain, the app listing, the email addresses, the printed inventory, the SEO equity, the customer recognition you paid to build. A conflict found in week one costs a conversation. The same conflict found in year three costs the brand.
Legal exposure. Under 15 U.S.C. § 1117(a) a successful plaintiff can recover the infringer’s profits, its own damages, and the costs of the action, and the court may enter judgment for up to three times actual damages. Attorney fees are available in exceptional cases. In Romag Fasteners, Inc. v. Fossil, Inc. (2020) the Supreme Court held unanimously that willfulness is not a precondition to an award of profits under § 1125(a), while noting that a defendant’s mental state remains a highly important consideration in the equitable calculation. Counterfeiting carries its own mandatory treble regime under § 1117(b).
Read those two points together and the incentive is stark. Clearance is not only about avoiding the conflict. It is about the record you build. A documented search followed by a reasoned decision to proceed looks like diligence. Receiving a cease-and-desist letter, ignoring it, and scaling the brand anyway looks like the willfulness that moves a judge toward the ceiling rather than the floor. Some businesses avoid searching precisely so they can claim ignorance, which is a strategy that fails on contact: the register itself is constructive notice under § 1072, so “I never looked” is not the shield people imagine.
Clearance is staged, not binary. Screen a long list of candidates with knockouts, cheaply and early, while switching names still costs nothing. Run full clearance on the two or three finalists before any money touches the name: before the logo, before the domain purchase, before the packaging run, and certainly before the application. Re-clear when you expand into materially new goods or a new country, because clearance is scoped to what you sell and where, and both of those change.
The proportion is a business judgment. A local service business with a coined name and a modest budget sits in a different risk posture than a consumer product launching nationally under a suggestive name in a crowded class. The searching effort should track what a mistake would actually cost you.
What is a trademark clearance search? A trademark clearance search is an investigation into whether a proposed name, logo, or slogan can be adopted without infringing someone else’s existing rights. It is broader than a registrability check. It looks for federal registrations and pending applications, state registrations, and unregistered common-law uses, then assesses the likelihood of confusion between those marks and yours in the goods and services you plan to sell.
Is a knockout search the same as a full clearance search? No. A knockout search is a fast, cheap screen that looks for identical or near-identical marks in the USPTO database, and its only job is to kill obviously doomed candidates before you spend money on them. A full clearance search adds phonetic and spelling variants, common-law and state sources, trade press, and domain and social usage, and it ends in a reasoned assessment rather than a hit list. Passing a knockout means nothing more than that you have not yet found the obvious problem.
Does searching the USPTO database mean my name is clear? No. The USPTO Trademark Search system, which replaced TESS in November 2023, covers only federal registrations and applications. In the United States trademark rights arise from use, not registration, so a business that has never filed anything can still hold enforceable priority. State registrations, foreign marks with U.S. reach, and unregistered common-law users are all invisible in that database, and any of them can force you to rebrand.
What happens if I skip clearance? Two categories of cost. The first is rebranding: signage, packaging, inventory, domains, app listings, and accrued goodwill all get written off, and the expense scales with how long you waited. The second is legal exposure. Under 15 U.S.C. § 1117(a) a court can award the infringer’s profits, the owner’s damages, up to three times actual damages, and attorney fees in exceptional cases. Adopting a mark after notice of a conflict is the kind of fact that pushes a court toward the high end.
Going further: How to lock your brand across trademark and domain, step by step .
This page is general legal information, not legal advice, and it does not create an attorney-client relationship.
The Seventh Circuit froze a good-faith Illinois junior user inside a 20-mile Mattoon enclave and gave the federal registrant the rest of the state.
The Ninth Circuit held that a junior user who knows of a senior user's mark cannot claim the remote good-faith defense, deepening a circuit split.
A unanimous Supreme Court held that whether a later mark may 'tack' onto an earlier mark's priority date is a question for the jury, locating the decisive moment in a clearance dispute in the fact-finder's assessment of consumer perception.
How the Supreme Court protected a good-faith Louisville druggist against a senior 'Rex' user from Massachusetts, establishing that common-law trademark rights are territorial, not national.
The Federal Circuit's analogous-use decision held that promotional activity can establish trademark priority before sales, but only on proof that it reached a substantial portion of the relevant consuming public, not merely that the user intended an association.
The Seventh Circuit's ZAZU decision held that token sales and an intent to register cannot establish trademark priority: only genuine market use that links the mark to a source in consumers' minds will do.
The Second Circuit's foundational concurrent-use decision held that a federal registrant cannot enjoin a good-faith remote junior user until it is likely to expand into that user's trading area, a rule that still shapes every trademark clearance opinion.
