Checklist · Intellectual Property

Clearing a New Trademark: A Practical Checklist

Type
Checklist
Jurisdiction
National
Date
2025-08-16
Topic
Intellectual Property

This checklist walks a proposed brand name from a short list of candidates through a knockout pass, a full search across the federal register, state registers, common-law uses, domains, and social handles, a likelihood-of-confusion analysis under the DuPont and circuit factors, a risk triage, a documented clearance file, and the final go/no-go decision. It answers the two separate questions every clearance must resolve: is the name available, and is it protectable. It reflects the USPTO's replacement of the retired TESS system with the cloud-based Trademark Search tool, and explains why a documented clearance file is worth more after Romag Fasteners v. Fossil made willfulness optional to a profits award. Worked hypotheticals show how the analysis plays out for a coffee startup, a fintech, and a skincare brand. It is a research resource, educational and not legal advice; for a high-stakes launch, obtain a written clearance opinion from qualified counsel.

Clearing a New Trademark: A Practical Checklist

Intellectual Property > Trademark | Published 16 August 2025 - Updated 19 August 2025 | UpLaw - uplaw.us

Clearance is the investigative work of proving, before a business spends a dollar building on a name, that the name is both free to use and worth owning. Nothing in the law forces you to search — which is exactly why so many companies walk into rights that already exist. This checklist turns clearance into a repeatable workflow: scope, knockout, full search, analysis, documentation, decision.

Keywords: trademark clearance, knockout search, full trademark search, USPTO Trademark Search, likelihood of confusion, DuPont factors, common-law trademark rights, clearance opinion, willful infringement, good-faith adoption, intent-to-use priority, constructive use, coexistence agreement, domain name clearance, Nice classification.

A trademark is a promise about source. When a name earns customer trust, the goodwill riding on it can become the most valuable thing a company owns — and the most expensive thing to lose. Clearance is the diligence that decides, before the marketing budget is committed and the packaging is printed, whether a candidate name can carry that weight safely. Done well, it is cheap insurance. Skipped, it is a bet placed blind against every prior user in the country, most of whom appear in no database you have thought to check.

Two features of American trademark law make clearance both necessary and easy to get wrong. First, rights do not begin at the USPTO. Under the common law, protectable rights arise from use in commerce, not from registration, so the senior user who can sink your launch may hold no certificate and file no papers anywhere. See Unregistered Trademark Rights at Common Law: Priority, Geography, and the Tea Rose-Rectanus Doctrine. Second, infringement does not require that two marks be identical, or that the goods be the same. The test is likelihood of confusion — whether an ordinary consumer, encountering the junior mark in the marketplace, would likely be confused about source, sponsorship, or affiliation. A search that hunts only for exact matches on identical products answers the wrong question and manufactures a dangerous kind of confidence.

This resource operationalizes the five-movement framework set out in our full narrative treatment, Trademark Clearance Searching: A Comprehensive Guide: scope, knockout, full search, analysis, and the documented decision. Work it top to bottom, once per surviving candidate. It is educational, not legal advice, and it does not substitute for a written opinion from counsel on a high-stakes adoption.

The two questions every clearance must answer

Clearance is often described as a single inquiry — "is this name taken?" — but it is really two questions wearing one coat, and they fail independently.

Availability asks whether adopting the name would collide with someone else's prior rights. This is the likelihood-of-confusion question, and it is what most of the search work targets: prior registrations, prior applications, prior common-law users.

Protectability asks whether the law and the USPTO will let you own and defend the name once you have it. A term can be gloriously unclaimed and still worthless, because it is generic (no one may own it) or merely descriptive (no one may own it yet, absent secondary meaning). See What the Law Will Protect: Substantive Standards for Trademarks. The distinctiveness spectrum from Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4 (2d Cir. 1976) — generic, descriptive, suggestive, arbitrary, fanciful — sorts marks by how much protection they earn, and the Supreme Court's USPTO v. Booking.com B.V., 591 U.S. 549 (2020), is a reminder that even a term the public treats as generic can occasionally be rescued by consumer perception, but only rarely and at real cost.

A name that clears on availability but flunks on protectability is a trap: you can use it, but you cannot stop the copycat who follows you, and the money you pour into it builds a moat around a castle you do not own. Screen both questions before you fall in love. For picking a candidate that is worth clearing in the first place, start with Choosing a Mark the Law Will Protect: A Practical Checklist.

A word on tools: TESS is gone

Any guide, template, or well-meaning colleague still telling you to "search TESS" is pointing at a system that no longer exists. The USPTO retired the Trademark Electronic Search System and replaced it with a cloud-based Trademark Search tool at the agency's site. The interface, the query syntax, and the field structure changed; the underlying obligation did not. Two companion tools remain indispensable: the Trademark ID Manual, which shows how the office describes goods and services and which categories it treats as related, and Trademark Status and Document Retrieval (TSDR), which serves file histories and status. Database contents and interfaces change without notice, so confirm the current system before you rely on any search — a step this checklist repeats, because it is the step people skip.

Phase 1 — Scope the project before you search

  • [ ] Generate a short list of five or six candidates, never one. Founders arrive married to a single name; the knockout will kill some of them, and you want survivors, not a funeral.
  • [ ] For each candidate, define what you will actually sell — the goods and services and their natural, foreseeable extensions — because relatedness of the offerings, not the name in isolation, drives most conflicts.
  • [ ] Map the realistic geographic and channel footprint (nationwide e-commerce versus a single storefront; retail shelves versus app stores), since common-law rights are territorial and channels of trade are a confusion factor in their own right.
  • [ ] Fix your risk tolerance deliberately and in advance. A venture-backed national launch and a neighborhood bakery can look at the identical conflict and rationally reach opposite decisions.
  • [ ] Flag any registrability hazard on the face of the name — a surname, a geographic term, descriptive or laudatory wording, a short letter or number string — because these demand tailored searching and may be hard to protect even when available.

Why this phase matters. Everything downstream is only as good as the scope you set here. If you define the goods too narrowly, your search will miss the related products where confusion actually lives; if you define them too broadly, you will drown in false hits. The goods-and-services description is not paperwork you defer to the application stage — it is the lens through which every later hit is judged relevant or irrelevant. For the discipline of getting that description right, see Classifying Goods and Services for a Trademark Filing: A Practical Checklist and The 45 Nice Classes: How Trademark Classification Shapes Cost, Scope, and Conflict.

Scope also decides how much clearance is enough. Clearance reduces risk; it never zeroes it. What counts as "cleared enough" is a business judgment calibrated to what you stand to lose, and that judgment cannot be made after the fact — it has to be made now, on the record, so that the search you commission matches the exposure you face.

Worked example — scoping a coffee startup. Suppose a founder wants to launch a direct-to-consumer coffee subscription under the name LUMEN. Before touching a database, scope forces three decisions. What will you sell? Roasted coffee (a good, Class 30) and a subscription service that curates and ships it (a service, arguably Class 35 or 43) — and, foreseeably, mugs, cold brew, and a café down the line. Where? Nationwide, online, from day one, so common-law pockets anywhere in the country matter. Registrability? LUMEN is Latin for "light" and an SI unit of luminous flux; it is arbitrary as applied to coffee, which is good for protectability, but its very ordinariness means many owners in many fields may hold it — so relatedness will do the heavy lifting. That single scoping paragraph reshapes the entire search: it tells you to search Class 30 goods and Class 43 services, to comb common-law sources nationwide, and to expect and tolerate LUMEN registrations for unrelated goods like lighting or software.

Phase 2 — The knockout search

The knockout is triage, not diagnosis. It is a fast, cheap pass designed to eliminate obviously doomed candidates before anyone spends money on a full workup — half an hour per name, not a multi-day investigation.

  • [ ] Open the USPTO Trademark Search tool and run the exact mark against identical and closely related goods and services.
  • [ ] Search phonetic equivalents — LUMEN becomes LOOMEN, LUMENN, LUMYN — because consumers hear marks at least as often as they read them, and sound is a pillar of the similarity analysis.
  • [ ] Search visual and spelling variants, including deliberate misspellings and compressed or hyphenated forms (LUMENCO, LU-MEN, LUMENN).
  • [ ] Search conceptual equivalents — synonyms, translations, and the underlying idea, not just the letters (GLOW, LUMINA, ILLUME) — because the confusion test compares connotation and commercial impression, not spelling alone.
  • [ ] Use the Trademark ID Manual to see how the office would describe your goods and which adjacent categories it treats as related.
  • [ ] Run a quick .com and social-handle availability check; a legally available name you cannot deploy online is a weak business choice even if it is a clean legal one.
  • [ ] Knock out any candidate that surfaces an identical mark for identical goods, a highly similar mark for closely related goods, or an owner with a reputation for aggressive enforcement.

The traps that live in Phase 2. The cardinal error is searching exact matches only. Infringement turns on likelihood of confusion, and an exact-match search is structurally incapable of finding the near-matches that generate most disputes. If your query for LUMEN returns nothing, that is not clearance — it is the absence of the easiest possible hit, which is not the same as safety.

The second trap is the Live/Dead field, which reads like a verdict and is only a lead. A registration marked "dead" tells you the registration lapsed; it says nothing about whether the mark is still in use. Common-law rights survive an abandoned or cancelled registration for as long as the owner keeps using the mark in commerce, and a company that let its paperwork lapse may still be selling under the name and holding priority against you. Treat a dead registration as a reason to investigate, not a green light.

The third trap is scope creep in reverse — reading a hit that covers unrelated goods as fatal. LUMEN registered for track lighting is not, by itself, a bar to LUMEN for coffee, because confusion is unlikely between illumination hardware and roasted beans. Learning to distinguish the hit that matters from the hit that merely exists is the entire skill of clearance, and it is why the knockout ends not in a yes-or-no but in a shorter list of survivors worth the cost of a full search.

Phase 3 — The full search

The full search is where lazy clearance dies. A search confined to the federal register misses most of the universe: common-law marks appear in no registry, the fifty state registers stand separate from the federal one and from each other, and the entire digital layer — organic search results, marketplace listings, app stores, social handles — is invisible to formal trademark databases. Run the full search only on candidates that survived the knockout, and run all of it.

3.1 The federal register, rigorously

  • [ ] Re-examine the federal register across every phonetic, visual, and conceptual variant and across all related classes, not just the class you plan to file in.
  • [ ] Include pending applications, not only registrations. A competitor who filed last month but has not yet registered still holds a priority date, and that date may beat yours.
  • [ ] Pull the file history and set status alerts in TSDR for any application that could mature into a blocking registration.

Two doctrines make the federal search treacherous if you cut corners. The first is the anti-dissection rule: marks are compared in their entireties, as a hurried consumer encounters them, not filleted into pieces and matched fragment by fragment. Estate of P.D. Beckwith, Inc. v. Commissioner of Patents, 252 U.S. 538 (1920). A search that only chases your exact word string will miss composite marks in which your term is the dominant, memorable element. The second is relatedness: the register is organized by Nice class for administrative convenience, but confusion does not respect class lines. As the Federal Circuit put it in In re Majestic Distilling Co., 315 F.3d 1311 (Fed. Cir. 2003), there is no per se rule that goods in different classes cannot be confusingly related; the question is always whether consumers would assume a common source. DELTA famously coexists for airlines, faucets, and dental insurance precisely because those markets are so unrelated that no one is confused — but that coexistence is a conclusion about relatedness, not a rule you can assume.

3.2 Intent-to-use applications: the time bombs

  • [ ] Flag every intent-to-use (ITU) application filed under Section 1(b), 15 U.S.C. § 1051(b), that covers a similar mark for related goods.

An ITU application deserves its own line because it can defeat you even though the applicant has never sold a thing. Under the constructive-use provision, 15 U.S.C. § 1057(c), a filing on the Principal Register confers nationwide priority as of the filing date, contingent on the registration later issuing. A similar ITU application filed before your first use is therefore a live threat with a priority date that predates your rights — a time bomb whose fuse is the eventual registration. You will not find it by searching for marks "in use," because by definition it is not yet in use. This is why the full search must sweep pending applications and why founders who plan to rely on being first should consider filing their own ITU to lock in a constructive-use date. See Intent-to-Use Trademark Filings: A Practical Guide, Completing an Intent-to-Use Application: A Statement of Use Checklist, and Constructive Use and Nationwide Priority Under 15 U.S.C. § 1057(c): A Practical Guide.

3.3 State registers

  • [ ] Search state trademark registers, targeted to the states where you and your known competitors operate plus the large commercial states (California, New York, Texas, Florida, Illinois).

There is no unified national database of state registrations; each secretary of state maintains its own. State registrations rarely confer broad rights on their own, but they are excellent evidence of common-law use — a state filing is a public flag that someone is using a mark in that territory, and it often leads you to a common-law user you would otherwise never have found.

3.4 Common-law uses: the hard, essential part

  • [ ] Search common-law uses systematically: business directories, trade publications, industry association membership lists, news archives, and court and TTAB dockets.
  • [ ] Comb the open internet — organic and paid search results, marketplace listings (Amazon, Etsy, eBay), app stores, and reverse-image search for logos and trade dress.
  • [ ] Capture what you find; the web is volatile and today's hit may be tomorrow's dead link.

The unregistered marks most likely to ambush you live in no register at all. A regional chain that has sold under your candidate name for fifteen years in three states holds priority in its trading area under Hanover Star Milling Co. v. Metcalf, 240 U.S. 403 (1916), and United Drug Co. v. Theodore Rectanus Co., 248 U.S. 90 (1918) — the Tea Rose-Rectanus doctrine — even though it never filed anything. Federal registration can override that territorial common-law priority prospectively under the Dawn Donut framework, Dawn Donut Co. v. Hart's Food Stores, Inc., 267 F.2d 358 (2d Cir. 1959), but it does not extinguish the senior user's rights in the zone it already occupies. Understanding where a common-law user's rights reach, and where yours can grow around them, is its own discipline: see The Geographic Reach of Common-Law Trademark Rights: A Practical Strategy Guide and The Tea Rose-Rectanus Doctrine and Territorial Trademark Rights: A Doctrinal Guide.

3.5 Domains, social handles, and industry registries

  • [ ] Clear domain names: the exact .com first, then .net and .org, then relevant new gTLDs (.app, .health, .ai, .co) and any country-code TLDs for markets you will enter. Use WHOIS to see who holds a registered domain and whether it is used.
  • [ ] Clear social handles across the major platforms — an available legal name paired with taken handles everywhere forces awkward compromises later.
  • [ ] Check industry-specific registries where they exist: USAN/INN for pharmaceutical names, TTB for alcoholic beverages, FINRA and the SEC for financial services, and state licensing boards for regulated professions.

Domain and handle clearance is business diligence with legal teeth. A cybersquatter sitting on your .com is not merely an inconvenience; recovering it may require a UDRP proceeding or an action under the Anti-Cybersquatting Consumer Protection Act, 15 U.S.C. § 1125(d). If a domain conflict surfaces during clearance, price the recovery before you commit to the name. See Recovering a Domain Name Through the UDRP: A Practical Checklist and Top-Level Domains and Trademark Rights: An Overview.

3.6 The professional vendor search

  • [ ] For higher-stakes launches, commission a professional full search from a specialist vendor (Corsearch, CompuMark, and similar), which searches federal, state, common-law, and domain sources and returns a structured report.

A vendor search buys breadth and speed you cannot easily replicate — proprietary common-law databases, phonetic algorithms, and analyst review — with expedited turnaround available for a fee. It does not, however, buy judgment. The report is raw data; the analysis is still yours (or your counsel's) to perform. Keep that distinction in mind, because it becomes legally significant when the file is later scrutinized: the third-party search report is generally discoverable data, while the attorney's opinion built on it is what privilege protects. More on that architecture in Phase 6.

Phase 4 — Analysis: apply the legal standards

A search produces hits. Analysis decides which hits are threats. This is the intellectual core of clearance, and it runs each surfaced conflict through the likelihood-of-confusion framework.

  • [ ] Run each significant conflict through the likelihood-of-confusion factors — the DuPont factors the USPTO and Federal Circuit apply, In re E.I. du Pont de Nemours & Co., 476 F.2d 1357 (C.C.P.A. 1973), plus the circuit tests you may face in court: AMF Inc. v. Sleekcraft Boats, 599 F.2d 341 (9th Cir. 1979), and Polaroid Corp. v. Polarad Elecs. Corp., 287 F.2d 492 (2d Cir. 1961).
  • [ ] Weigh the two dominant factors hardest: similarity of the marks and relatedness of the goods.
  • [ ] Assess strength of each senior mark; the distinctiveness spectrum sets the width of its protective moat.
  • [ ] Triage every conflict into High, Moderate, or Lower risk, weighing not only legal merit but the cost of a fight you would ultimately win.
  • [ ] Investigate doubtful conflicts from public sources, staying clear of pretextual or direct-contact investigation that raises ethics problems.

For a factor-by-factor workflow you can run on each hit, use Working Up Likelihood of Confusion: A Practical Checklist; for the doctrine in narrative depth, see Likelihood of Confusion in Trademark Law: A Brand Owner's Guide.

4.1 Similarity of the marks

Marks are compared in their entireties — sight, sound, and meaning — as a consumer of ordinary care would encounter them in the actual marketplace, not side by side under a lawyer's magnifying glass. The Restatement collects the market realities that shape this judgment. Restatement (Third) of Unfair Competition §§ 20-21 (1995). Three sub-comparisons matter:

  • Sight — visual impression, including dominant elements, length, and structure.
  • Sound — phonetic similarity, because consumers ask for products aloud and hear advertising.
  • Meaning — connotation and commercial impression, including translations. Under the doctrine of foreign equivalents, a foreign-language mark is translated into English when the ordinary American purchaser would stop and translate it, and the translated meaning is then compared. Palm Bay Imps., Inc. v. Veuve Clicquot Ponsardin Maison Fondee En 1772, 396 F.3d 1369 (Fed. Cir. 2005).

Worked example — the translation trap. Imagine a skincare founder wants to sell a serum under SÖLV, chosen because it looks clean and Scandinavian. SÖLV is Danish and Norwegian for "silver." A knockout search for the letters S-Ö-L-V returns little. But under the doctrine of foreign equivalents, an examiner or a court may treat SÖLV as SILVER, and now the relevant search is for SILVER skincare marks — of which there are many. The visual novelty that made the name appealing is exactly what a purely orthographic search would let slip through. Conceptual searching, not spelling searching, catches it.

4.2 Relatedness of the goods and services

Confusion turns on relatedness in the consumer's mind, not on the class number. Goods need not be identical or even competitive; they need only be related enough that consumers would assume a common source or sponsorship. But relatedness is not infinite. Some pairings that seem close on the shelf require, in the Federal Circuit's phrase, "something more" than mere thematic overlap to support confusion — for instance, beer and restaurant services, where the court in In re Coors Brewing Co., 343 F.3d 1340 (Fed. Cir. 2003), rejected the assumption that a brewery and an eatery are automatically related. Relatedness is a factual judgment about marketplace reality, informed by how the goods are sold, to whom, and through what channels. Restatement (Third) of Unfair Competition § 23 (1995).

4.3 Strength of the senior mark

A strong mark casts a wide shadow; a weak one barely reaches beyond its own goods. Strength has two components: conceptual strength (where the mark sits on the Abercrombie spectrum — fanciful and arbitrary marks are strong, descriptive marks weak) and commercial strength (how much marketplace recognition the mark has actually acquired). A famous mark is entitled to broad protection precisely because its fame makes consumers more likely to assume that any similar mark, even on distant goods, is an extension of the famous brand. Recot, Inc. v. Becton, 214 F.3d 1322 (Fed. Cir. 2000). The flip side is that a crowded field of similar marks for similar goods weakens every mark in the field — if a dozen registrants already coexist using variants of your term, the senior user's ability to object to one more is diminished, which can turn a scary-looking hit into a manageable one.

4.4 Risk triage

Not every conflict is equal, and clearance is a resource-allocation exercise as much as a legal one. Sort each surviving conflict:

  • High risk — an identical or nearly identical mark for identical or closely related goods, held by an active user, especially one with a litigation habit. These usually end the inquiry for that candidate.
  • Moderate risk — genuine similarity but with distinguishing features (different channels, sophisticated purchasers, a weak or crowded senior mark, uncertain senior use). These are candidates for modification or negotiation.
  • Lower risk — remote similarity, unrelated goods, dead or dormant senior rights, or a senior user confined to a distant territory.

Triage weighs cost, not just merit. A conflict you would probably win still carries the price of an opposition before the Trademark Trial and Appeal Board or a federal suit, plus the delay and distraction. The purchaser-sophistication factor cuts here too: buyers of expensive, considered purchases (enterprise software, medical devices) are less easily confused than impulse buyers of cheap goods, which can lower the practical risk of an otherwise similar mark.

4.5 Investigate — carefully

When a conflict is doubtful — you cannot tell whether the senior user is still active, or how broadly it operates — investigate from public sources: the owner's website and social presence, the USPTO file history in TSDR, secretary-of-state records, and business databases like Dun & Bradstreet. Stay away from pretextual contact or ruses that pose as customers or investors; the goodwill you are trying to protect is not worth an ethics problem, and a clumsy investigation can itself become evidence of bad faith.

4.6 The two rules that make documentation valuable

Two legal realities give the analysis its stakes. First, doubts on likelihood of confusion are resolved against the junior user, who bears the duty to steer clear of a confusing mark. Bridgestone Ams. Tire Operations, LLC v. Federal Corp., 673 F.3d 1330 (Fed. Cir. 2012); Nina Ricci, S.A.R.L. v. E.T.F. Enters., Inc., 889 F.2d 1070 (Fed. Cir. 1989). You do not get the benefit of the doubt; the prior user does. Second — and this is the reason clearance documentation is worth more than it used to be — after Romag Fasteners, Inc. v. Fossil, Inc., 590 U.S. 212 (2020), a plaintiff no longer must prove willful infringement as a precondition to disgorging the infringer's profits under 15 U.S.C. § 1117(a). Willfulness remains highly relevant to whether a court awards profits and how much, but it is no longer a gate. That makes the good-faith record a clearance file produces more valuable, because it is now doing defensive work across a wider range of remedies. On how diligence blunts a willfulness charge, see Rebutting Willful Trademark Infringement: Clearance Searches, Opinions of Counsel, and USPTO Approval. And because a TTAB likelihood-of-confusion determination can carry preclusive weight in later litigation under B&B Hardware, Inc. v. Hargis Indus., Inc., 575 U.S. 138 (2015), the analysis you do now can echo for years.

Phase 5 — Confirm protectability of the survivor

Availability and protectability fail independently, so a candidate that clears Phase 4 still needs a protectability check before you commit.

  • [ ] Place the mark on the distinctiveness spectrum. Fanciful (KODAK), arbitrary (LUMEN for coffee), and suggestive (NETFLIX) marks are inherently distinctive and register on the Principal Register without more. Descriptive marks (a term that names a feature or quality) register only on a showing of secondary meaning under 15 U.S.C. § 1052(f); generic terms never register.
  • [ ] Screen for statutory bars in Section 2, 15 U.S.C. § 1052 — primarily descriptiveness and deceptive misdescriptiveness under § 1052(e), and the primarily-merely-a-surname and geographic bars.
  • [ ] If the name is a personal surname, expect a Section 2(e)(4) refusal absent acquired distinctiveness. See Registering Your Own Name as a Trademark: Surname Bars, Consent, and Secondary Meaning.
  • [ ] Decide whether a likely refusal is fatal or curable — a descriptiveness problem may be solved by time and use, a genericness problem never.

The protectability screen prevents a demoralizing sequence: you clear a name against all prior users, invest in it, file, and then draw a Section 2(e) refusal from the examining attorney that no amount of searching would have avoided. The distinctiveness spectrum and the color and functionality rules that flank it — Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763 (1992), and Qualitex Co. v. Jacobson Prods. Co., 514 U.S. 159 (1995) — are the substantive law behind that refusal, and a five-minute placement of your candidate on the spectrum tells you whether you are choosing a mark that will register cleanly or one that will fight. If a refusal is likely, the response toolkit lives at Overcoming a USPTO Trademark Office Action: A Practical Checklist.

Worked example — the descriptive fintech. A founder clears NORTHKEY for a payments API and finds no conflicts. Clean availability. But suppose the founder had instead chosen SECUREPAY. A search might show the exact term unregistered — because it is so descriptive of secure payments that prior users could not register it either. The absence of a registration is not a gift; it is a warning that the term is weak. SECUREPAY would draw a descriptiveness refusal, would be nearly impossible to police against competitors using similar descriptive language, and would leave the company with a name it spent money on but cannot defend. NORTHKEY — arbitrary as applied to payments — clears both questions. SECUREPAY clears only one, and the wrong one.

Phase 6 — Document the process

Documentation converts clearance work into clearance value. An undocumented search that happens to be thorough proves nothing later; a documented search founds a good-faith defense, supports a formal opinion, and satisfies the investors and insurers who will ask what diligence you did.

  • [ ] Record the date and scope of each search — which databases, which query variants, which classes — and capture screenshots of results, because database contents shift and may be unreproducible months later.
  • [ ] Write up the analysis of every significant hit, the risk categorization, and the recommendation, so the reasoning survives the people who did it.
  • [ ] Understand the privilege architecture. The third-party search report is generally discoverable raw data. The attorney's opinion built on it is the work product privilege protects. Discuss the handling of any negative finding with counsel before it is committed to writing.
  • [ ] For high-stakes adoptions, obtain a formal written clearance opinion from qualified counsel — an opinion that is logical, thorough, and genuinely relied upon, because a court weighing good faith will test exactly those qualities.

The privilege point deserves care, because it cuts both ways. A competent, genuinely-relied-upon opinion of counsel is the classic "shield of good faith" against a willfulness charge — but asserting advice-of-counsel as a defense generally waives privilege over the opinion. The document you may one day raise as a shield is a document the other side will then get to read, dissect, and use. That is not a reason to skip the opinion; it is a reason to make it honest and defensible, since a self-serving opinion that ignores the strongest conflicts will hurt more than help when it surfaces. A written clearance opinion also mirrors the structure counsel use in the patent world, where a freedom-to-operate opinion plays the same defensive role — a parallel explored in Clearing a Product for Launch: How to Conduct a Freedom-to-Operate Analysis.

Phase 7 — Decide: proceed, modify, clear the obstacle, or abandon

Clearance ends in a decision, and there are four of them.

Proceed

Modify

  • [ ] Modify when a conflict is problematic but not fatal. Add a distinctive element, adopt a special-form or logo mark, pair the term with a house mark, or narrow the goods — but narrow honestly, by describing what you truly sell, not by gaming a class number to dodge a hit you know is real.

Modification is underused. A conflict that looks fatal to the bare word mark often dissolves when the mark is presented as a distinctive composite, paired with an established house brand, or aimed at a genuinely different set of goods and buyers. The art is to change enough to defeat confusion without changing so much that you abandon the brand you wanted.

Clear the obstacle

  • [ ] Clear a blocking reference through a negotiated consent or coexistence agreement, a cancellation petition if the reference registration looks vulnerable to an abandonment or non-use challenge, or an outright purchase or license of the senior rights.

Each route has a companion resource. A coexistence agreement lets two users occupy adjacent space with agreed guardrails; a cancellation attacks a registration that no longer reflects real use — see Cancelling a Trademark Registration at the TTAB: A Practical Checklist. One caution runs through all of them: be prepared to walk away before you approach the senior user. Asking for consent, being refused, and then proceeding anyway can supply the very evidence of bad faith and knowledge that destroys the good-faith defense — the lesson of Stone Creek, Inc. v. Omnia Italian Design, Inc., 875 F.3d 426 (9th Cir. 2017), where a junior user's knowledge of the senior mark sank its Tea Rose-Rectanus good-faith claim. See Stone Creek v. Omnia: How Knowledge of a Senior Mark Defeats the Tea Rose-Rectanus Good-Faith Defense.

Abandon

  • [ ] Abandon when the conflict is an identical mark for identical goods, when the senior owner is a notorious enforcer, or when several independent conflicts stack up on one candidate.

Walking away during clearance costs only the time you spent. Walking away after launch costs the signage, the packaging, the domain, the search rankings, the printed collateral, and the customer recognition you paid to build — and then costs it all again to rebuild under a new name, possibly under the pressure of a cease-and-desist letter or an injunction. If a cease-and-desist letter is what prompted this analysis, work Answering a Trademark Cease-and-Desist Letter: A Practical Checklist in parallel.

Two decision hygiene rules

  • [ ] Re-run the knockout immediately before you commit serious brand spend. Clearance is a snapshot, and a competitor may have filed an ITU application in the weeks or months while you developed. A refresh search catches the application that landed after your first pass and before your first dollar.
  • [ ] For foreign expansion, clear country by country. Many jurisdictions are first-to-file, meaning the first to register wins regardless of prior use, so a name clean in the United States may already be registered — or squatted — abroad. The Madrid Protocol streamlines filing across member countries; it does not perform clearance, and it does not create rights in a country where a prior registrant already holds the mark.

Special situations

Rebrands and expansions. Clearing a new name for an existing company is not easier than clearing for a startup; it is harder, because the company's expansion into new goods or new territories can collide with users who were harmlessly remote before. Re-clear whenever you add a product line, enter a new region, or move into a new channel.

Acquisitions and licensing. If you are acquiring a brand rather than creating one, clearance folds into chain-of-title diligence: confirm the seller actually owns the goodwill, that prior assignments transferred goodwill rather than being bare assignments in gross, and that no lurking common-law user has priority in a key market. These questions live at the intersection of clearance and transactional diligence covered in The IP Clearance Toolkit: Securing Freedom to Operate Before You Launch.

House marks and sub-brands. A company with a strong house mark can lower clearance risk for a new product by presenting the new term as a sub-brand under the house mark (think of an umbrella brand followed by a product name). The composite is harder to confuse with a third party's standalone mark, and the house mark's strength does defensive work. This is a modification strategy available before you ever hit a conflict, and it is worth building into the candidate list in Phase 1.

Surveys. In a genuinely close case where the cost of being wrong is high, a properly designed consumer confusion survey can move the analysis from argument to evidence. Surveys are expensive and easy to botch, and a bad one is worse than none; if the stakes warrant one, design it to survive a Daubert challenge from the start. See Designing a Defensible Confusion Survey: A Practical Checklist.

An extended worked example: clearing LUMEN end to end

Return to the coffee founder and run the full workflow, to see how the phases interlock.

Phase 1 (scope). LUMEN, for roasted coffee and a nationwide subscription service, with foreseeable extensions into mugs, cold brew, and eventually a café. Arbitrary as to coffee, so protectability looks strong. Nationwide online footprint, so common-law pockets anywhere matter.

Phase 2 (knockout). The founder searches LUMEN and its phonetic and conceptual cousins. LUMEN turns up registered for LED lighting (Class 11), for a project-management SaaS (Class 9), and for a health-tech breath-analysis device (Class 10). None is coffee. A LUMINA registration exists for teas and herbal infusions — closer, and worth flagging. The candidate survives, but LUMINA-for-tea earns a full-search look.

Phase 3 (full search). The federal search confirms the lighting, software, and device registrations and turns up a pending ITU application for LUMEN covering "coffee and coffee-based beverages," filed four months ago by a company that has not launched. This is the time bomb: constructive use under Section 1057(c) would date that applicant's priority to a filing that predates the founder's first use. State registers show nothing new. Common-law searching finds a single café in Portland operating as "Lumen Coffee Bar" for six years, with no registration — a classic Tea Rose-Rectanus senior user with rights in its local trading area. Domain search shows lumencoffee.com is taken and parked.

Phase 4 (analysis). Two conflicts matter. The pending ITU for coffee is High risk — same mark, identical goods, and a priority date that likely beats the founder's. The Portland café is Moderate-to-High as to that market but geographically confined; under Dawn Donut, a later federal registration by the founder would not dislodge the café in the territory it already occupies, and the café could block the founder there. The LUMINA-for-tea registration is Lower risk after analysis — different mark, related but not identical goods, and tea is a distinct enough category that "something more" would be needed to show confusion.

Phase 5 (protectability). LUMEN for coffee is arbitrary and inherently distinctive; no Section 2 bar looms. Protectability is not the problem here — availability is.

Phases 6-7 (document and decide). The founder documents everything, then faces the ITU squarely. Options: (a) abandon LUMEN, the cleanest response to a same-mark, same-goods senior filing; (b) attempt to buy or take a coexistence from the ITU applicant, remembering that being refused and proceeding anyway would supply bad-faith evidence; or (c) modify — but modification does little against an identical-goods ITU, because the goods, not the mark's ornamentation, are the collision point. The disciplined decision is to abandon LUMEN for coffee and turn to the next survivor on the Phase 1 list, having spent only clearance time rather than a launched brand. That is the entire value proposition of clearance in one example: the ITU that would have surfaced as a cease-and-desist letter six months after launch instead surfaced as a line item in a search, and the cost of the collision was a few hours, not a rebrand.

Common mistakes

  • Searching too narrowly — exact matches only, one class only, one spelling only. This is the single most common and most dangerous error, because a clean narrow search feels like clearance and is not.
  • Getting the goods analysis backward — treating every similar mark as fatal (paralysis) or every non-identical mark as safe (recklessness). Relatedness is a spectrum, judged by marketplace reality.
  • Over-reading dead registrations — a lapsed registration is not a lapsed mark; the owner may still be using it with live common-law rights.
  • Underweighting common-law searching — the unregistered users most likely to ambush you appear in no register and only surface through directories, dockets, and the open web.
  • Skipping the refresh — failing to re-run the knockout right before major spend, and missing the ITU that landed during development.
  • Consulting counsel too late — after the cheap fixes (modify, pick another candidate) are gone and only the expensive ones (litigate, rebrand) remain.
  • Confusing availability with protectability — clearing a descriptive or generic term against prior users and mistaking the empty register for a green light, when the emptiness is a warning that the term cannot be owned.

Primary authority

Statutes (Lanham Act, Title 15, U.S. Code). § 1051 (application and filing bases, including § 1(b) intent-to-use); § 1052 (grounds for refusal, including § 2(d) likelihood of confusion, § 2(e) descriptiveness and surname bars, and § 2(f) acquired distinctiveness); § 1057(c) (constructive use and nationwide priority); § 1064 (cancellation, including for genericness and abandonment); § 1072 (registration as constructive notice); §§ 1114 and 1125(a) (infringement of registered and unregistered marks); § 1125(c) (dilution); § 1125(d) (cybersquatting, ACPA); § 1117(a) (monetary remedies).

Cases. In re E.I. du Pont de Nemours & Co., 476 F.2d 1357 (C.C.P.A. 1973) (confusion factors); AMF Inc. v. Sleekcraft Boats, 599 F.2d 341 (9th Cir. 1979); Polaroid Corp. v. Polarad Elecs. Corp., 287 F.2d 492 (2d Cir. 1961); Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4 (2d Cir. 1976) (distinctiveness spectrum); Estate of P.D. Beckwith, Inc. v. Commissioner of Patents, 252 U.S. 538 (1920) (anti-dissection); Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763 (1992); Qualitex Co. v. Jacobson Prods. Co., 514 U.S. 159 (1995); Hanover Star Milling Co. v. Metcalf, 240 U.S. 403 (1916) and United Drug Co. v. Theodore Rectanus Co., 248 U.S. 90 (1918) (territorial common-law rights); Dawn Donut Co. v. Hart's Food Stores, Inc., 267 F.2d 358 (2d Cir. 1959); Stone Creek, Inc. v. Omnia Italian Design, Inc., 875 F.3d 426 (9th Cir. 2017); In re Majestic Distilling Co., 315 F.3d 1311 (Fed. Cir. 2003) and In re Coors Brewing Co., 343 F.3d 1340 (Fed. Cir. 2003) (relatedness); Recot, Inc. v. Becton, 214 F.3d 1322 (Fed. Cir. 2000) (fame); Palm Bay Imps., Inc. v. Veuve Clicquot Ponsardin, 396 F.3d 1369 (Fed. Cir. 2005) (foreign equivalents); Bridgestone Ams. Tire Operations, LLC v. Federal Corp., 673 F.3d 1330 (Fed. Cir. 2012) and Nina Ricci, S.A.R.L. v. E.T.F. Enters., Inc., 889 F.2d 1070 (Fed. Cir. 1989) (doubts against the junior user); B&B Hardware, Inc. v. Hargis Indus., Inc., 575 U.S. 138 (2015) (issue preclusion); USPTO v. Booking.com B.V., 591 U.S. 549 (2020) (genericness); Romag Fasteners, Inc. v. Fossil, Inc., 590 U.S. 212 (2020) (willfulness and profits).

Secondary and agency authority. Restatement (Third) of Unfair Competition §§ 20-23 (1995); J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition; USPTO, Trademark Manual of Examining Procedure (TMEP); USPTO Trademark Search tool and Trademark ID Manual; Trademark Status and Document Retrieval (TSDR); WIPO Madrid Protocol; ICANN Uniform Domain-Name Dispute-Resolution Policy (UDRP). Database contents, tools, and fee structures change; confirm the current systems before relying on any search.

Related UpLaw Resources

Related Guides

Related Checklists

Related Articles

Related Toolkits

This checklist is a research resource, educational and not legal advice. Clearance involves subjective, fact-specific judgments; for a high-stakes adoption, obtain a written clearance opinion from qualified counsel.