A logo is not an asset. Not yet.
The minute a designer flattens the final file, what you've made isn't property. It's a hypothesis — a bet that nobody, anywhere, in the same corner of commerce, planted that flag first. Most agencies never place the bet on purpose. They skip the question, hand over a gorgeous file, and call it ownership. It isn't. It's good typography sitting on top of an unanswered legal question: can the client actually own this, or did they just pay you to decorate someone else's rights?
I'm a trademark attorney. I've spent my career on the other side of that question — the side where the lookbook is already shot and the answer is no. So let me walk you through how ownership actually works, why beautiful names fail, and where, exactly, the floor gives out. Not to scare you. To make you better at the thing you already do.
Start with your best work.
A client hires you to build a clothing line, soup to nuts. You source the blanks — the heavyweight cotton that hangs right, not the scratchy stuff. You dial the wash, the fit, the palette, the feeling a person gets holding it. You book the studio. Models, lighting, a stylist, every piece shot like it's already iconic. You name it. You build the mark, the lockups, the brand book that makes the founder tear up a little. Months of work. The invoice clears.
Then someone runs the search, and the name's taken. Not by a competitor you've heard of — by some company three states over, registered in the same class of goods, doing something close enough that the law calls it a collision.
Now every shirt, every hangtag, every frame from that studio day is a beautiful corpse. Finished, flawless, and legally unburiable. The client can't build on it. You can't even run it as a portfolio piece, because it never launched. Your design wasn't wrong. Your sequence was. The cheapest step in the whole engagement got skipped, and it dragged the most expensive steps into the grave with it.
Here's the thing most creatives never get told, and it changes how you name forever.
You don't get a trademark by being first to think of a name. You don't get it by buying the domain, forming the LLC, or grabbing the Instagram handle. None of those are trademark rights. A Secretary of State will happily register an LLC name that infringes a federal mark — they don't check, that's not their job. The handle and the URL are just real estate. They tell you nothing about whether you can legally use the name in commerce without getting sued.
A trademark is something narrower and stranger: the right to use a specific mark, for specific goods or services, as a source identifier — a signal to buyers that this product comes from you and not someone else. That's the whole point of the system. It doesn't protect words. It protects the connection between a word and a source, inside a lane of commerce.
Which is why two businesses can own the same word. Delta Faucets and Delta Air Lines coexist because nobody's confusing a kitchen sink with a flight to Atlanta. Different lanes. The USPTO sorts goods and services into 45 of these lanes — the Nice classification system — and apparel lives in Class 25. Your client's rights live and die inside their class and the ones close enough to bleed into it. "Is the name available" is the wrong question. The right one is: "Is it available for this, against everything already sitting in and around this class?"
That's not a creative question. It's a legal one. And it's why a name can feel completely original and still be dead on arrival.
Now the part that'll actually make your naming sharper.
Not all marks are created equal. Trademark law ranks them on a spectrum of distinctiveness, and where a name lands decides whether it's protectable at all — before anyone else's rights even enter the picture.
Generic is the graveyard. You cannot trademark "Shirts" for shirts. Nobody gets to own the category's own name. Dead on arrival, every time.
Descriptive is the trap, because it's where founders instinctively run. "Cozy," "Pure," "Premium," "Organic" — words that describe what the thing is or does. Clients love them because they explain the product. The law treats them as weak-to-unprotectable, because you don't get a monopoly on describing your own goods. A descriptive mark only earns protection if it develops "secondary meaning" — years of use and money until the public hears the word and thinks of you first. Most brands die long before they get there. So when an agency hands a client a clean, descriptive, "instantly understandable" name, it often hands them the least ownable name in the room.
Then the good stuff:
Suggestive marks hint without describing — Netflix, Airbnb. They require a small leap of imagination, and the law rewards that leap with real protection.
Arbitrary marks are real words used somewhere they don't belong. Apple, for computers. The word's common; the use is unexpected, and that's what makes it strong.
Fanciful marks are invented from nothing — Kodak, Xerox, Verizon. The strongest marks on earth, because they exist only as brands.
Read that spectrum as a creative brief, not a law lecture. The names that clear, and the names worth fighting for, tend to live on the suggestive-to-fanciful end. The names that get rejected — by the USPTO or by reality — cluster down in descriptive and generic. Clearance isn't the enemy of good naming. It points you straight at it.
Now the odds, because I don't deal in vibes.
According to IP practitioners analyzing USPTO data, roughly one in five U.S. trademark applications draws a Section 2(d) "likelihood of confusion" refusal — the examiner decides it's too close to a mark already registered. About 25% hit an office action of some kind. And when applicants fight a refusal on appeal, the Trademark Trial and Appeal Board affirmed those refusals around 90.9% of the time in 2020.
Sit with that last number. Once you're on the wrong side of a confusion call, the system is telling you, nine times out of ten, you lose the appeal too. There's no talking your way out after the fact. The leverage was all on the front end, in the search you didn't run.
And "likelihood of confusion" is broader than people expect. It isn't "is it the identical name." Examiners and courts weigh a whole set of factors — how similar the marks look, sound, and mean; how related the goods are; how the products travel to market; how strong the senior mark is. Close in sound counts. Close in commercial impression counts. "Lush" and "Lavish" in the same aisle can be a problem. Your client doesn't have to copy anyone to lose. They just have to land too close.
You might assume this only happens to people too cheap to hire lawyers. The opposite is true.
Gucci spent 2017 and 2018 firing cease-and-desist letters at Forever 21 over a few red-and-green and blue-red-blue stripes. It escalated into federal court and settled, quietly, on undisclosed terms. Two giants, war chests on both sides, full legal teams — and still a multi-year brawl over stripes.
Go bigger. The Dominican Republic paid roughly RD$34 million pesos to develop "Marca País," a national brand, and unveiled the logo at a government event. Within about 48 hours the internet flagged it as nearly identical to a 2014 design by a Russian artist, Ivan Bobrov — who confirmed he'd never sold it to anyone. The mark was abandoned. A sovereign nation, with a sovereign budget and presumably a building full of advisors, got caught flat because the work outran the clearance.
If Gucci and an actual government can get blindsided, the question isn't whether your client's shirt line is exposed. It's how exposed, and whether anyone checked.
Here's the structural failure, and it's not laziness. It's order of operations.
The creative process runs forward: discovery, strategy, naming, identity, rollout. The legal question gets quietly filed under "later — the client's problem." So the search, if it happens at all, happens at the end. Which is precisely when changing the name costs the most: the merch is shot, the deck is built, the founder is in love, the shelf agreement is signed. By then nobody wants to be the person who says the name has to go.
It should run the other way. A real clearance process runs alongside the creative, not after it. Early on, while names are still candidates, you run a knockout search — a fast pass to kill the obvious collisions before anyone gets attached. The two or three survivors get a deeper, comprehensive search across the relevant classes, common-law uses, and confusingly similar variants. Only then does a name graduate from "we like it" to "we can probably own it."
That single reorder is the whole game. It moves the expensive discovery — this name is taken — from after the studio day to before the naming even finishes. Same work. Wildly different cost.
This is the unglamorous part, and I'll keep it short because you didn't come here for a pitch.
Knockout searches, comprehensive clearance, reading a likelihood-of-confusion risk, and actually filing the application in the right classes — that's where a trademark attorney comes in. Not at the end, as cleanup. Early, as a filter that runs quietly behind your creative so the names you fall for are names that survive. I've done this with a handful of branding and marketing agencies, sitting in at the naming stage so the work that leaves the building is work the client can actually own. The mechanics — the search, the classes, the filing, the office actions that hit one in four applications — sit with the lawyer. The creative stays yours.
That's the entire role. A second set of eyes that reads the register while you read the moodboard.
Strip away the doctrine and it comes down to one idea: a brand isn't something you design. It's something you're allowed to own. The design is the easy half. The ownership — the class, the spectrum, the search, the confusion analysis — is the half that decides whether any of the design ever sees daylight.
The best agencies already sense this. They're not just studios; they're the first call a founder makes when they're building something real. Understanding how to trademark a logo — really, how to clear a name before it's chosen — is how you stop handing clients beautiful corpses and start handing them brands built to stand.
Design the thing. Just make sure it's theirs to build on.
Note: USPTO percentages are practitioner estimates based on USPTO data, not official USPTO-published rankings. The distinctiveness spectrum and confusion factors are well-settled trademark doctrine (verify exact phrasing against TMEP/case law before publishing). The clothing-line scenario is illustrative; the Gucci/Forever 21 and DR Marca País cases are verified.