Checklist · Intellectual Property

Choosing a Mark the Law Will Protect: A Practical Checklist

Type
Checklist
Jurisdiction
National
Date
2025-08-14
Topic
Intellectual Property

A working checklist for choosing a brand name that the law will actually defend, organized around the distinctiveness spectrum that fixes both registrability and enforcement scope. It moves from confirming a trademark is the right tool, through generating and scoring candidates on the fanciful-to-generic scale, screening the Section 2 statutory bars for descriptive, geographic, surname, deceptive, and functional matter, and pressure-testing conceptual strength, commercial strength, and scope of protection. Each phase carries a plain-English rationale, trap warnings, and invented worked examples, closing with a scoring rubric, common mistakes, and a hand-off to clearance and filing. The through-line is that distinctiveness is decided at the naming whiteboard, not the examining attorney's desk. Educational, not legal advice.

Choosing a Mark the Law Will Protect: A Practical Checklist

Intellectual Property > Trademark Selection | Published 14 August 2025 - Updated 29 October 2025 | UpLaw - uplaw.us

The one-sentence version. Two bakeries open on the same block: one calls itself "The Bread Store," the other "Wondermill." Both sell excellent sourdough; only one owns a name a court would enforce. A trademark is not ownership of a word — it is a legally recognized link between a source and the goods or services it sells, backed by the law's interest in preventing consumer confusion. The strength of that link, and whether it exists at all, is set the moment the founder picks the name. This checklist helps you choose a name that registers cleanly, protects widely, and will not dissolve into the public domain. Pair it with Trademark Fundamentals: A Plain-English Primer and What the Law Will Protect: Substantive Standards for Trademarks.

Keywords: protectable trademark, distinctiveness spectrum, Abercrombie categories, secondary meaning, Section 2(e) bars, surname refusal, geographic descriptiveness, genericide, crowded field, conceptual vs. commercial strength, brand name selection, trade name vs. trademark.

Why this checklist exists

Most trademark disasters are not litigation failures. They are naming failures that surface years later, dressed up as litigation. A company spends a marketing budget teaching the public a name, then discovers the name was never strong enough to stop a competitor from adopting a near-copy — or worse, that the name was never protectable at all. By then the sunk cost is the brand itself, and the choice is between an expensive rebrand and an unwinnable fight.

The good news is that the single most important variable in that whole story is decided cheaply, early, and by you: the distinctiveness of the term you pick. Distinctiveness governs two things at once. It governs registrability — whether the U.S. Patent and Trademark Office (USPTO) will place your mark on the Principal Register without a fight — and it governs scope — how far your rights reach when you later need to enforce them. A conceptually strong mark is a fortress with a wide moat. A conceptually weak one is a fence made of string that any competitor can step over by arguing it was only describing its own goods.

This document treats naming as the legal decision it actually is. It is written as a checklist because that is the format that survives contact with a real branding meeting, but each phase is expanded into the doctrine, the leading authorities, and the edge cases a careful adviser would want at hand. Work the phases in order. Score candidates before you fall in love with one. And remember the governing idea throughout: you are forecasting every future fight the mark will ever be in. For the deeper doctrinal treatment behind these steps, see What Can Be a Trademark: The Subject Matter of Trademark Law.

Phase 1 — Confirm a trademark is the right tool, and what it can and cannot do

Before scoring names, confirm you are protecting the right thing with the right regime. A surprising number of naming problems are really category errors.

  • [ ] Confirm you are protecting a source identifier, not an invention or a creative work. Trademark law protects the brand signal — a word, logo, slogan, sound, color, or package design that tells buyers who stands behind the goods. It does not protect the underlying idea (that is patent territory) or the expressive content (that is copyright). If your value is in a mechanism or a formula, a mark will not carry it; see Copyright, Trademark, Patent, and Trade Secret Compared: A Practical Guide.
  • [ ] Separate your trade name from your trademark. Registering "Wondermill LLC" with the Secretary of State is a corporate act; it creates an entity, not a brand right. Trademark rights arise from use of the term to identify the source of goods or services in commerce, not from incorporation. The statute defines a trademark by that source-identifying function. See 15 U.S.C. § 1127.
  • [ ] Confirm the term will be used as a mark. A word buried in ad copy, or used only as the name of the product's category, is not functioning as a source identifier. The Supreme Court reminded everyone how central this is in Jack Daniel's Properties, Inc. v. VIP Products LLC, 599 U.S. 140 (2023): the threshold question is whether the designation is being used to identify source at all.
  • [ ] Decide whether now is the right time to invest. Distinctiveness is free to get right at the whiteboard and expensive to fix after launch. For the timing and priority calculus — including whether to file an intent-to-use application before you sell anything — see Timing Your Trademark Filing: A Practical Guide and When to Trademark Your Brand: An Overview of Priority, Timing, and Cost.

Why this matters. Founders routinely believe that forming a company under a name locks up the brand. It does not. The two systems answer different questions — "does this legal entity exist?" versus "who owns the right to use this name on these goods?" — and the answers can diverge badly. A company can hold a clean corporate registration for a name that is legally worthless as a mark because it is generic, and it can hold strong trademark rights in a name it never bothered to incorporate under. Sort the tool from the goal now, and the rest of the checklist has something real to work on.

Worked example (hypothetical). Imagine a founder, Priya, has invented a genuinely new kitchen gadget: a countertop device that cold-brews coffee in ninety seconds. She is tempted to pour everything into protecting "the name of the machine." But the machine is the thing that deserves patent attention; the brand on the box is the trademark. If she names the category-defining product "The Ninety-Second Cold Brewer," she has described her invention perfectly and protected her brand not at all — and she may even seed a future genericide problem, because the market will have no other word for the thing. Choosing "VELDT" for the same device separates the invention (patent), the manual and marketing copy (copyright), and the source signal on the box (trademark) into three regimes that each do their own job.

Phase 2 — Generate candidates and score them on the distinctiveness spectrum

This is the heart of the exercise. Every U.S. word-mark analysis runs through the classification the Second Circuit set out in Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4, 9 (2d Cir. 1976) (Friendly, J.), which sorts terms into five ascending categories of distinctiveness. Learn the categories cold; the entire law of mark strength hangs off them.

  • [ ] Brainstorm five or six candidates, not one. Clearance and the statutory bars will kill some of your favorites. Going in with a single beloved name is how founders end up rationalizing a weak choice.
  • [ ] Score each candidate on the spectrum:
    • [ ] Fanciful — invented words with no prior dictionary meaning, coined to be marks (KODAK, EXXON, XEROX). Strongest possible; inherently distinctive; the widest moat.
    • [ ] Arbitrary — real words with no relationship to the goods (APPLE for computers, CAMEL for cigarettes). Strong; inherently distinctive.
    • [ ] Suggestive — terms that hint at a quality but require a mental step to connect to the goods (MICROSOFT for software, COPPERTONE for suntan lotion, GREYHOUND for bus travel). Inherently distinctive, but sitting right on the most heavily litigated line in the field.
    • [ ] Descriptive — terms that directly convey a feature, quality, function, or characteristic (CREAMY for yogurt, SHARP for televisions). Not registrable on the Principal Register absent proof of secondary meaning under 15 U.S.C. § 1052(f).
    • [ ] Generic — the common name of the product category itself ("Coffee" for coffee, "Bread Store" for a bakery). Never protectable by anyone, no matter how much money is spent.
  • [ ] Favor the fanciful / arbitrary / strong-suggestive end unless you have a specific, well-understood reason to accept a weaker mark.

Why this matters. Fanciful, arbitrary, and suggestive marks are inherently distinctive — they identify a single source the instant they are used, so they can be registered on the Principal Register immediately and enjoy protection from day one. Descriptive terms are not inherently distinctive; they can only become protectable if and when consumers come to treat them as a brand rather than a description, which is the doctrine of secondary meaning (also called acquired distinctiveness). Generic terms can never get there, because a word that is the product cannot simultaneously point to one seller of it. The distinctiveness spectrum is not academic taxonomy; it is a pricing sheet for how much brand-building and litigation risk you are signing up for.

The fanciful/arbitrary end: why coined and unrelated words win

Fanciful marks are made of nothing. KODAK meant nothing before George Eastman's company gave it meaning, which is precisely why it is unassailable — no competitor has any legitimate need to use it, and no consumer confusion story competes with the brand's own. Arbitrary marks borrow a real word but strand it far from its meaning: APPLE has everything to do with fruit and nothing to do with laptops, so the computer company owns the word in that field without denying grocers the use of "apple" for apples.

The practical payoff runs in both directions. At the USPTO, an inherently distinctive term sails past the descriptiveness refusal that sinks so many applications. In court, the same conceptual strength earns the mark a place at the strong end of the likelihood-of-confusion analysis, where courts give wide berth against later entrants. The trade-off is marketing legibility: a coined word teaches the buyer nothing about the product and must be built up with advertising. That cost is real, but it is a cost you choose and control, not a legal weakness baked into the asset.

The suggestive/descriptive line: the field's most consequential coin flip

The line between suggestive and descriptive decides whether a mark is protectable on arrival or must survive years of secondary-meaning proof first. Courts draw it with two recurring tools.

The imagination test asks how much mental effort a consumer must expend to move from the term to the product's characteristics. If the term conveys the characteristic immediately and directly, it is descriptive; if it requires "thought, imagination, and perception" to make the connection, it is suggestive. COPPERTONE requires a small imaginative leap from the word to "this will give you a copper tan"; CREAMY for yogurt requires none. The competitors' need test asks a policy question: do other honest sellers need this word to describe their own goods? If yes — if forbidding the word would handicap legitimate competitors — the term is descriptive and belongs to everyone. These tests appear throughout the case law and are distilled in the Restatement (Third) of Unfair Competition §§ 13–14 (1995) and 2 J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition ch. 11 (5th ed.).

Worked example (hypothetical). An oat-milk startup weighs five names for its first carton. "OATS" is generic — it is the product, unownable by anyone. "CREAMY OAT" is descriptive; it announces two features (texture and ingredient) and would draw a Section 2(e)(1) refusal, registrable only later if the company proves the public reads it as a brand. "MOOLESS" is suggestive: it takes a beat to get the joke — no cow, therefore dairy-free — and that mental step earns it inherent distinctiveness. "NIMBUS" is arbitrary as applied to oat milk; a real word about clouds, unrelated to breakfast, it registers cleanly with broad reach. "ZORVEX" is fanciful, a pure coinage, the strongest and the emptiest — it will need the most advertising and reward it with the widest protection. Ranked by legal strength the order is exactly the reverse of ranked by how much the name "sells the product." That inversion — the most on-the-nose name is the weakest asset — is the central tension of naming, and it is far cheaper to resolve here than after a rebrand.

Descriptive marks and the secondary-meaning bargain

Choosing a descriptive term is not automatically fatal, but it is a bargain with a bill attached. A descriptive mark can mature into a protectable one only by acquiring secondary meaning: proof that, in the minds of the relevant consumers, the term's primary significance has shifted from describing the goods to identifying their source. The statute lets the USPTO accept five years of substantially exclusive and continuous use as prima facie evidence of acquired distinctiveness, 15 U.S.C. § 1052(f), but that is a floor, not a guarantee, and a strongly descriptive term may need far more.

Courts assessing secondary meaning weigh a familiar cluster of factors: the amount and manner of advertising, sales volume, the length and exclusivity of use, unsolicited media coverage, consumer surveys, and evidence that competitors intentionally copied the term. No single factor controls, and direct evidence — a well-designed survey — usually carries the most weight. The Supreme Court's treatment of descriptive marks in Park 'N Fly, Inc. v. Dollar Park & Fly, Inc., 469 U.S. 189 (1985), confirms that a descriptive mark, once it clears this bar and registers, can become incontestable like any other. But building secondary meaning is a multi-year, evidence-hungry project, and until it succeeds the mark is exposed. If you must go descriptive, budget for the proof and consider the Supplemental Register as an interim home — see the discussion in Three Tiers of Trademark Protection.

The generic floor: the words no one can own

Generic terms are the common name of the genus of goods, and the Supreme Court has long defined a generic term as "one that refers to the genus of which the particular product is a species," Park 'N Fly, 469 U.S. at 194. A generic term is not weakly protected; it is unprotectable, forever, by anyone, because letting one seller own the category name would strip every competitor of the only word buyers use for the thing. Picking a generic name is the one naming decision the law will never rescue. It is discussed at length in What Can Be a Trademark and revisited under genericide in Phase 4.

Phase 3 — Screen out the statutory bars and structurally weak categories

Even an inherently distinctive term can be refused if it falls into one of the statutory bars in Section 2 of the Lanham Act, 15 U.S.C. § 1052. Screening for these up front is far cheaper than arguing your way out of a refusal after you have filed and started building. Run every surviving candidate through this gauntlet. For the response side — what to do when a refusal lands anyway — see Overcoming a USPTO Trademark Office Action: A Practical Checklist.

  • [ ] Merely descriptive — § 2(e)(1). Would an examining attorney say the mark merely describes an ingredient, quality, function, feature, purpose, or use of the goods? If so, expect a refusal absent secondary meaning. This is the most common substantive refusal, and it is the direct downstream consequence of a Phase 2 scoring error.
  • [ ] Primarily geographically descriptive — § 2(e)(2). Does the term convey a primary and immediate association with a geographic place from which the goods actually come? If yes, it is treated like any descriptive term and needs secondary meaning.
  • [ ] Primarily geographically deceptively misdescriptive — § 2(e)(3). Does the term name a place the goods do not come from, in a way that would materially deceive a meaningful segment of buyers? After the trade-agreement amendments, this bar is effectively permanent — such marks cannot be saved by secondary meaning.
  • [ ] Primarily merely a surname — § 2(e)(4). Would consumers perceive the term primarily as a last name? Surname marks need acquired distinctiveness to register. See Registering Your Own Name as a Trademark: Surname Bars, Consent, and Secondary Meaning.
  • [ ] Functional matter — § 2(e)(5). Product or package features that are functional cannot be monopolized through trademark; that is the province of (time-limited) patents.
  • [ ] Deceptive, false-association, and prohibited-insignia matter — § 2(a). Screen for genuinely deceptive terms, false suggestions of connection with people or institutions, and government insignia. Note that the old disparagement and immoral/scandalous bars were struck down on First Amendment grounds in Matal v. Tam, 582 U.S. 218 (2017), and Iancu v. Brunetti, 588 U.S. 388 (2019); the remaining § 2(a) bars survive.
  • [ ] Consent for living persons — § 2(c). A mark that identifies a particular living individual needs that person's written consent. The Supreme Court upheld this "names clause" against a First Amendment challenge in Vidal v. Elster, 602 U.S. 286 (2024).
  • [ ] Laudatory and "informational" matter. Self-praise ("BEST," "PREMIUM," "#1") and ordinary informational slogans generally cannot function as marks — there is nothing source-identifying about telling everyone your product is good.
  • [ ] Crowded field. Check whether the term sits amid many third-party uses for similar goods. A crowded field thins everyone's protection under the sixth DuPont factor (discussed in Phase 4).

Why this matters. Each of these buckets converts a naming decision into a purchased refusal or a thin, hard-to-enforce right before you have filed a single form. The statutory bars are not obscure traps; they are predictable, and screening for them is the cheapest insurance in the entire branding budget.

Descriptiveness, again — the bar you already scored

Section 2(e)(1) is Phase 2 wearing an examiner's badge. If a candidate scored "descriptive" on the spectrum, this is where the USPTO says so out loud. The examiner will typically pull dictionary definitions, competitor usage, and the applicant's own marketing to show the term describes the goods. The defense is either secondary-meaning evidence under § 2(f) or an argument that the term is actually suggestive. Because that suggestive/descriptive line is genuinely subjective, reasonable examiners and applicants disagree constantly — which is exactly why you want to have won the argument at the whiteboard by picking a term that never invites it.

Geographic terms: place-of-origin traps in both directions

Geography cuts two ways. "SONOMA" on wine that actually comes from Sonoma County is geographically descriptive under § 2(e)(2) — the public associates the place with the goods, the goods come from there, and the term therefore needs secondary meaning. "HAVANA" on cigars rolled in Ohio is a different animal: the place is famous for the goods, the goods do not come from there, and buyers might reasonably be deceived. That is geographically deceptively misdescriptive matter under § 2(e)(3), and after the North American Free Trade Agreement implementing amendments folded a materiality requirement into the analysis, such marks are barred essentially permanently and cannot be rescued by acquired distinctiveness. The screening question is not just "is there a place name in here?" but "does the public link this place to these goods, and are the goods actually from there?"

Worked example (hypothetical). A furniture startup in Newark loves "BROOKLYN" for its industrial-style tables. Two questions decide the fate of that name. First, does the buying public associate Brooklyn with furniture such that the term is primarily geographic rather than, say, arbitrary or evocative-lifestyle? If Brooklyn carries only a vague aesthetic connotation and no goods/place association for furniture, the term may be treated as arbitrary or suggestive and sail through. Second, if there is a goods/place association and the tables are made elsewhere, the company risks a § 2(e)(3) deceptiveness problem if the "Brooklyn" cachet is material to buyers. The safe move is either to source honestly from the named place, to confirm no goods/place association exists, or to pick a name that does not gamble on either.

Surnames: shared last names and the personal-name problem

A term that consumers would perceive as "primarily merely a surname" is refused under § 2(e)(4) unless it has acquired distinctiveness. The USPTO weighs a familiar set of factors — how rare the surname is, whether anyone connected to the applicant bears it, whether the term has any recognized non-surname meaning, whether it has the "look and feel" of a surname, and (for stylized marks) whether the styling itself creates a separate commercial impression. These are often called the Benthin factors, after In re Benthin Management GmbH, 37 U.S.P.Q.2d 1332 (T.T.A.B. 1995). A rare surname that doubles as an ordinary word ("KING," "BIRD") is easier; a common, meaning-free surname ("MILLER," "OKAFOR") is harder and usually needs secondary meaning. Founders naming a company after themselves should expect this and plan for the proof. The full analysis, including the § 2(c) consent overlay and Section 2(f) strategy, lives in Registering Your Own Name as a Trademark.

Functional and laudatory matter: things that cannot be marks at all

Two categories fail not because they are weak but because they are structurally incapable of functioning as marks. Functional features — a product or package feature that is essential to the use or purpose of the article or that affects its cost or quality — are barred by § 2(e)(5) and by the functionality doctrine the Supreme Court applied in TrafFix Devices, Inc. v. Marketing Displays, Inc., 532 U.S. 23, 29 (2001), and Qualitex Co. v. Jacobson Products Co., 514 U.S. 159 (1995). Trademark cannot be used to seize a functional advantage that patent law would grant only temporarily; that is the channeling principle explored in Trade Dress Protection: Where Brand Identity Meets the Functionality Bar. Laudatory matter fails for a different reason: praise is not source identification. Everyone claims to be the best, so "the best" points to no one. That intuition is why marks built entirely of self-praise are so hard to register.

Phase 4 — Pressure-test conceptual strength, commercial strength, and scope

Surviving the statutory bars gets you a mark that can be protected. This phase asks how far the protection will reach, because the same distinctiveness that eases registration also widens the enforcement moat. Trademark practitioners separate two kinds of strength, and a careful naming decision accounts for both.

  • [ ] Rate conceptual (inherent) strength using the Phase 2 spectrum position. Fanciful and arbitrary marks start strong; suggestive marks are strong but nearer the descriptive edge; descriptive marks start weak.
  • [ ] Forecast commercial (marketplace) strength — how recognized the mark will realistically become. A conceptually modest mark backed by enormous, exclusive use can grow commercially strong; a conceptually brilliant mark that no one ever sees stays practically weak.
  • [ ] Test sight, sound, and meaning. Say the candidate out loud, imagine it misheard on a phone, picture it misspelled in a search bar. A hurried, imperfect-recall consumer is the audience the confusion analysis actually protects.
  • [ ] Check the crowded field. Count the third-party uses of the term (or its core element) for related goods. A term everyone is already using is weak no matter where it sits on the spectrum.
  • [ ] Map the scope you want to own. Sketch the natural extensions of the business — adjacent goods, related services, future product lines — and ask whether the mark can plausibly stretch across them.
  • [ ] Assess genericide risk. Is the candidate one careless step from becoming the category name? Plan usage discipline from day one.
  • [ ] Consider whether a distinctive logo or stylization can shore up a weak word without pretending it fixes the word itself.

Why this matters. Strength is not a compliment; it is a measurable input into every future dispute. A strong mark reaches across related goods and lets you stop near-copies early. A weak mark is hemmed in, and its owner spends the enforcement budget losing arguments about crowded fields and descriptive fair use. The strength you build in at naming is the strength you will spend in court.

Conceptual versus commercial strength: the two-axis picture

Think of mark strength as a grid. One axis is conceptual strength, fixed at birth by where the term lands on the Abercrombie spectrum. The other is commercial strength, earned over time through advertising, sales, exclusivity, and public recognition. A fanciful mark starts high on the first axis and can climb the second; a descriptive mark starts at the bottom of the first axis and can only reach protectability by climbing the second far enough to prove secondary meaning. Courts assessing likelihood of confusion look at both — the mark's inherent distinctiveness and its marketplace fame — because a mark's real-world reach is the product of the two. When you pick a conceptually strong name, you are buying a head start on the axis you cannot easily buy later. For how these strength findings play out in enforcement, see Likelihood of Confusion in Trademark Law: A Brand Owner's Guide.

The crowded field: why the sixth DuPont factor can gut a strong-looking mark

The confusion analysis at the USPTO runs through the factors of In re E.I. du Pont de Nemours & Co., 476 F.2d 1357 (C.C.P.A. 1973); its sixth factor asks about "the number and nature of similar marks in use on similar goods." When many third parties already use a term or a shared element, that element is diluted in the descriptive sense — consumers have learned to distinguish among the many users by looking at the other parts of each mark, so no single user gets to claim the shared element broadly. The Federal Circuit has repeatedly held that evidence of a crowded field narrows a mark's scope of protection; Juice Generation, Inc. v. GS Enterprises LLC, 794 F.3d 1334 (Fed. Cir. 2015), is a clean modern statement of the principle. The naming lesson is blunt: a term can look strong on the spectrum and still be commercially thin because everyone got there first.

Worked example (hypothetical). A hot-sauce founder falls for "DRAGON'S BREATH." It is arguably suggestive — heat, fire, a small imaginative leap — and would ordinarily score well. But a five-minute search of the marketplace turns up a dozen "DRAGON," "INFERNO," and "FIRE-BREATHER" sauces already on shelves. The field is crowded. Even if the founder registers, the mark's reach is thin: a court would let a competitor sell "DRAGON HEAT" because consumers in this category have been trained to distinguish spicy-mythology names by their trailing words. The same founder's coined alternative — "KZATL" — sits in an empty field and would own its lane completely. Spectrum position told half the story; the crowded-field check told the half that decides real disputes.

Genericide: the failure mode that grows with success

Here is the cruel irony of trademark strength: the one risk that increases as a brand wins is genericide, the death of a mark that becomes the public's ordinary word for the product itself. ESCALATOR, ASPIRIN, THERMOS, CELLOPHANE, and YO-YO were all once valid trademarks; each dissolved when consumers adopted it as the name of the thing rather than the brand of the thing. See Kellogg Co. v. National Biscuit Co., 305 U.S. 111 (1938) (shredded wheat); Bayer Co. v. United Drug Co., 272 F. 505 (S.D.N.Y. 1921) (L. Hand, J.) (aspirin); King-Seeley Thermos Co. v. Aladdin Industries, Inc., 321 F.2d 577 (2d Cir. 1963) (thermos).

Genericness turns on the primary significance test: to the relevant public, does the term principally name the product (generic, unprotectable) or the source (a valid mark)? The Lanham Act codifies this inquiry for cancellation at 15 U.S.C. § 1064(3), and evidence of how the public and the media use the term is central — generic media usage is treated as strong evidence of public perception, Murphy Door Bed Co. v. Interior Sleep Systems, Inc., 874 F.2d 95, 101 (2d Cir. 1989). Notably, a mark does not become generic merely because people use it as a verb: in Elliott v. Google, Inc., 860 F.3d 1151 (9th Cir. 2017), the court held GOOGLE survived even though people "google" things, because the public still understood the word to identify a particular search engine as a source.

The naming defenses are two. First, in selection: avoid a mark that is one small slip from the category term, and never adopt a coined name and then also use it as the only word for a brand-new product category — that is how a company teaches the market to genericize its own mark. Second, in usage: always use the mark as an adjective modifying a generic noun ("VELDT cold brewer," never "a veldt"), pair it with the generic name of the category, and police competitors and the press who use it generically. The genericide risk is the strongest argument for building a usage style guide before launch, discussed in Phase 5 and in The Life Cycle of a Trademark: How Brand Rights Arise, Strengthen, and Disappear.

Scope: naming for the business you will become

Distinctiveness widens more than the fight over your exact goods; it widens the radius of related goods you can plausibly reach. A famous, fanciful mark can stop confusingly similar uses across surprisingly distant categories; a weak, descriptive mark can barely defend its home turf. When you score candidates, look past the launch SKU to the natural extensions of the enterprise — the second product line, the services layer, the licensing deal three years out. A mark that can stretch across those extensions is worth more than one that fits only the first product. This is also where classification strategy begins; how you describe and class the goods shapes both cost and conflict, as explained in The 45 Nice Classes and Classifying Goods and Services for a Trademark Filing: A Practical Checklist.

A useful discipline: imagine the mark on the other side of a dispute. If you had to send a cease-and-desist letter over a near-copy, would the mark's strength carry the argument, or would the recipient's lawyer point instantly to a crowded field and a descriptive core? The distinctiveness that makes a mark easy to register is the same distinctiveness that makes it easy to enforce — and a name chosen for marketing punch alone often turns out to be a name you can neither register cleanly nor defend confidently.

Special-category marks: same doctrine, different application

Some candidates belong to categories that behave predictably differently from an ordinary coined word. The distinctiveness analysis does not change; its application does. Recognize the category and you can price the risk before you commit.

Personal names and surnames

Many people legitimately share a name, so a founder-name brand starts with a built-in weakness under § 2(e)(4) and needs both diligence (who else operates under this name?) and, usually, acquired distinctiveness to register. Personal-name marks can grow strong — plenty of famous brands are surnames — but they climb the commercial-strength axis to get there. Screen for others using the name in your field, plan the § 2(f) proof, and secure any § 2(c) consents if the name identifies a living person. The dedicated treatment is Registering Your Own Name as a Trademark.

Acronyms and short letter strings

Two- and three-letter marks live in crowded territory and turn heavily on the specific goods and the specific letters. An acronym is generally treated as descriptive if it would be understood as the equivalent of a descriptive phrase it stands for — "ISP" for internet service providers is no more registrable than "internet service provider" itself. Clear both the letters and their common expansions, and expect thin protection for short strings unless the letters are arbitrary in the field. A coined-sounding letter mark ("NRG" for an energy drink) may read as suggestive, but only if buyers hear it as a brand rather than as shorthand for "energy."

Slogans and taglines

A slogan built from descriptive or laudatory language may be unable to function as a mark at all — there is little to clear because there is little to protect. "SIMPLY THE BEST" for cleaning products is self-praise; "FRESH CLEAN EVERY TIME" describes an aspiration. A slogan can be protectable if it is distinctive and used as a source identifier rather than as ordinary informational or promotional matter, but the bar is real. If the tagline is doing marketing work, keep it; just do not mistake it for a defensible asset unless it clears the distinctiveness screen on its own.

The "generic.com" wrinkle

Adding ".com" to a generic term does not automatically save it, but it does not automatically doom it either. In USPTO v. Booking.com B.V., 591 U.S. 549 (2020), the Supreme Court rejected a per se rule that "generic + .com" is always generic, holding that the test is whether consumers actually perceive the whole term as a brand rather than as the category name. The upshot for naming: a "descriptive.com" or "generic.com" domain can sometimes be registrable, but only on a strong showing of consumer perception (typically survey evidence), and even then the resulting rights are narrow. Do not build a brand on a category-word domain expecting broad protection.

Logos, stylization, and trade dress

A distinctive logo or stylized presentation can add source-identifying character to an otherwise weak word — but it protects the design, not the word underneath. A stylized "CREAMY" logo may register for its design elements while leaving the word "creamy" free for competitors. Do not treat a logo as a cure for a weak word or, worse, as a substitute for clearing the word at all. Where the shape or look of the product or packaging is itself the brand signal, you are in trade-dress territory, which carries its own inherent-distinctiveness rules (product design can never be inherently distinctive and always needs secondary meaning per Wal-Mart Stores, Inc. v. Samara Bros., Inc., 529 U.S. 205, 210–13 (2000), while distinctive packaging can be inherently distinctive under Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763 (1992)). See Trade Dress Protection: Where Brand Identity Meets the Functionality Bar.

Non-traditional and digital-first marks

Color, sound, scent, and now virtual-goods marks all run through the same distinctiveness and functionality lenses. A single color can be a mark only on secondary meaning and only if non-functional (Qualitex, 514 U.S. at 163–66). For brands launching in games, metaverse storefronts, or NFT drops, the source-use question from Jack Daniel's becomes central, and classification gets its own complications — see Trademark Doctrine for Virtual Goods and NFTs: The MetaBirkins Landscape. The doctrine is the same; the specimens and the perception evidence are different.

Phase 5 — Hand off to clearance, filing, and disciplined use

Distinctiveness answers whether a mark is protectable. It does not answer whether the mark is available — that is, whether someone else already owns rights that would block you. Those are separate questions, and you need a "yes" to both.

Why this matters. The most common way a distinctive name fails is not at the examiner's desk but at the priority line: someone used it first for related goods. Distinctiveness gets you a mark worth owning; clearance tells you whether you can own it; disciplined use keeps it alive. The whole arc — selection, clearance, filing, maintenance — is mapped in the From Brand Idea to Registered Mark: A Trademark Registration Toolkit and the Federal Trademark Registration Lifecycle checklist collection.

A quick scoring rubric

When you have to compare candidates fast, score each on five dimensions and total them. It will not replace legal judgment, but it forces the right conversation.

  • Spectrum position (0–4). Fanciful/arbitrary = 4; strong suggestive = 3; weak suggestive = 2; descriptive = 1; generic = 0. A zero is disqualifying.
  • Statutory-bar exposure (0–2). No § 2 issues = 2; a curable issue (surname, geographic-descriptive) = 1; a hard bar (deceptive, functional, geographically deceptively misdescriptive) = 0.
  • Crowded-field score (0–2). Empty field = 2; moderate third-party use = 1; saturated field = 0.
  • Scope and stretch (0–2). Reaches across natural extensions = 2; fits only the launch product = 1; awkward even for the first product = 0.
  • Genericide resilience (0–1). Comfortably distinct from the category term = 1; one careless step away = 0.

A candidate scoring 9–11 is a fortress worth clearing. A 5–8 is workable but wants a hard look at its weakest axis. Anything below 5 should go back to the whiteboard before it consumes a clearance budget.

Common mistakes

  • Choosing a descriptive name because it "explains the product," then discovering it is barely protectable and hemmed in by every competitor's right to describe their own goods the same way. The name that sells hardest is usually the name that defends weakest.
  • Confusing a trade name with trademark rights. Incorporating "Wondermill LLC" secures an entity, not a brand. Source-identifying use secures the mark. See 15 U.S.C. § 1127.
  • Ignoring the crowded field. A term can look strong on the spectrum and still be commercially thin because a dozen competitors got to the shared element first.
  • Using the mark as a noun or a verb, or launching a coined name as the only word for a brand-new product category, and thereby teaching the market to genericize it.
  • Treating a logo as a substitute for clearing the underlying word, or as a cure for a word that is descriptive or already taken.
  • Betting the brand on a "generic.com" domain and expecting Booking.com to deliver broad protection it rarely gives.
  • Skipping clearance because the name "feels available." Availability is a priority question that only a search answers; distinctiveness cannot substitute for it.
  • Falling in love with one name before scoring five. Single-candidate naming breeds rationalization, and rationalization is how weak marks get chosen.

Primary authority

Related UpLaw Resources

Related Checklists

Related Guides and Articles

Related Toolkits

This checklist is educational and not legal advice. Distinctiveness and registrability are fact-specific, and the suggestive/descriptive line in particular turns on how real consumers perceive a term. For a high-stakes naming decision, consult a qualified trademark attorney and commission a proper clearance search before you commit a marketing budget.