1. The label is the business
Walk the aisle of any store and run a quiet experiment: subtract the labels. What's left is fabric, fragrance, and formulation, most of it made in the same handful of factories, much of it nearly interchangeable. What you actually shop with, what you pay premiums for, trust with your skin, and gift to people you love, is the label. The brand is the product's memory: every good experience anyone has ever had with it, stored in a name.
That's not marketing poetry; it's the economic reality trademark law was built for. The Lanham Act's definition of use for goods is physical and specific: a mark is used in commerce on goods when it is "placed in any manner on the goods or their containers or the displays associated therewith or on the tags or labels affixed thereto," and the goods are sold or transported in commerce. 15 U.S.C. § 1127. The law literally lives on your hangtag. And once registered, that mark carries the presumption that changes every future fight: validity, ownership, and the exclusive right to use it for the registered goods, by statute. 15 U.S.C. § 1057(b).
Here's why this chapter comes first instead of the pricing. In a product business, every other asset is replaceable under pressure. Supplier raises prices? You switch suppliers. Design gets copied? You design the next drop. But if the name isn't yours, there is no workaround: you either fight for it or you re-label the entire operation, and re-labeling a physical-goods business means dead inventory, reprinted packaging, re-shot photography, and a customer base retrained from scratch. Founders who've been through it will tell you it's the most expensive thing that ever happened to them, and it was always preventable for a fraction of one production run.
One of the registrations in my own clients' portfolio makes the point better than I can: KREYÒL ESSENCE® anchors a beauty brand built on Haitian ingredients and a founder's story, federally registered, enforceable, licensable. The products are excellent. The protected name is what makes the products a company.
Subtract the labels and most products are interchangeable.
In a physical-product business, the brand is frequently worth more than everything else combined, because products can be copied and factories can be switched, but the name on the label is the only thing competitors cannot legally reproduce, once you own it. 15 U.S.C. § 1127 protects exactly this: the mark placed on goods, their containers, tags, and labels.
2. What you can own: name, logo, slogan, color, packaging
Most founders think trademark protection stops at the name. For a product business, the protectable surface is much larger, and the Supreme Court has personally walked most of it. Here's the inventory, easiest to hardest.
The name and the logo. The core, covered throughout this guide. Word mark first, always, because it protects the name in every font, color, and rendering; the logo follows as its own filing when budget allows, per the sequencing logic in The Lady Lawyer's Guide® to Trademark Registration.
The slogan. Registrable when it functions as a brand identifier rather than decoration, a distinction that matters enormously for apparel and gets its own chapter here (Chapter 4), because a slogan on a product label and a slogan splashed across a shirt front are, legally, two different animals.
The packaging. Here's one most founders don't know: distinctive product packaging, what the law calls trade dress, can be protected, and if it's inherently distinctive, it's protectable without proving the public already associates it with you. That's the holding of Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763 (1992). The unmistakable jar, the signature box, the bottle silhouette: if the look is distinctive and not functional, it can be yours.
The color. Yes, really. In Qualitex Co. v. Jacobson Products Co., 514 U.S. 159 (1995), the Supreme Court held that a color alone, there, the green-gold of dry-cleaning press pads, can be registered as a trademark once it has acquired secondary meaning, so long as the color isn't functional. This is advanced-portfolio territory, built over years of consistent use, but it's the ceiling worth knowing about: the most valuable product brands eventually own sensory real estate, not just words.
The hard boundary: product design. Now the carve-out that saves founders from an expensive misunderstanding. In Wal-Mart Stores, Inc. v. Samara Brothers, Inc., 529 U.S. 205 (2000), a case about copied children's clothing designs, the Supreme Court held that product design, the cut of the garment, the shape of the product itself, can never be inherently distinctive and always requires proof of secondary meaning before trade dress protection attaches. Translated for your business: trademark law will not stop a competitor from copying your silhouette or your formulation on day one. It protects the identity you sell under, and it protects designs only after the market has learned to read them as yours. (Design patents and copyright pick up some of what trademark leaves here, which is a strategy conversation, not a chapter.)
The practical takeaway: protect in order of leverage. Name first, logo second, slogan when it's doing brand work, packaging when it's distinctive, color when you've earned it. Each layer makes the knockoff's job harder, and Chapter 7 is about how much that matters.
Product design always needs secondary meaning — Wal-Mart v. Samara.
A product brand can protect far more than its name: the logo, the slogan, distinctive packaging (inherently protectable under Two Pesos), even a color that has come to mean you (Qualitex). The one carve-out: product design itself always requires proof of secondary meaning (Wal-Mart v. Samara Brothers). Build the portfolio in that order of difficulty.
3. Picking a name that survives the search
Every trademark lawyer in America learned the strength of brand names from a clothing case. Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4 (2d Cir. 1976), a fight over the word "Safari" on apparel, gave the law its enduring spectrum, and fifty years later it's still the first test your name has to pass.
Generic names, the word for the product itself, can never be owned. "Activewear" for activewear belongs to everyone, which means it belongs to no one. Descriptive names, "Soft Cotton Basics," "Hydrating Hair Oil," immediately describe the goods, and the law won't give one seller the plain words every competitor needs, not until years of heavy use teach the public the phrase means you specifically. If you're starting fresh, this lane is quicksand dressed as clarity. Suggestive, arbitrary, and fanciful names, the ones that hint without describing, borrow an unrelated word, or invent one outright, are where ownable product brands live, and the market agrees with the law here: the product names you can recall from across a room are almost never the descriptive ones.
For product founders specifically, two pressures make this chapter urgent in a way it isn't for service brands. First, your categories are the most crowded on the register. Apparel and beauty attract more filings from individuals and small companies than almost anything else, which means the name you love has more near-neighbors than a name in almost any other industry, and the legal test is not "identical." It's likelihood of confusion, judged on similarity of sound, appearance, and meaning for related goods under the factors from In re E.I. DuPont DeNemours & Co., 476 F.2d 1357 (C.C.P.A. 1973). In a crowded class, "close" is a conflict.
Second, your mistakes get printed. A service brand that discovers a conflict rebrands a website. A product brand that discovers one rebrands a warehouse: woven labels, hangtags, boxes, inserts, silk screens, UPC registrations, marketplace listings, and the photography that showed all of it. The same conflict costs a product founder a multiple of what it costs anyone else, which is why the sequencing rule of this entire practice applies to you with extra force: the search comes before the production run. Not after the sampling, not alongside the launch, before the first unit is printed.
A comprehensive search, federal, state, and common law, on your brand name is $679, with the written report in 3 business days and 30 minutes with me walking through exactly what it found and what I'd do. If the name is clear, you print with confidence. If it isn't, you just saved the cost of a warehouse full of someone else's trademark.
Find out if your brand name is clear.
Search My Brand Name — $679Product brand names live on the Abercrombie spectrum, from generic (unownable) through descriptive (weak) to suggestive, arbitrary, and fanciful (strong), and the case that named the spectrum was itself an apparel fight. Strength is half the question; availability is the other half, and in crowded categories like fashion and beauty, the search is where brands live or die before launch.
4. The ornamental trap: why clothing applications die
Now the chapter this guide exists for, because this is the refusal that kills more clothing-brand applications than any other, and almost nobody outside the profession has heard of it before the refusal letter arrives.
Here's the scenario, and if you're a founder, it will sound uncomfortably like the plan. You've got a phrase, your slogan, your statement, your movement. You print it big and proud across the front of the shirt, because that's the product: people buy the shirt to wear the phrase. You file for the trademark, submit a photo of the shirt as your proof of use, and months later the USPTO refuses registration because the use is "merely ornamental."
The logic, laid out in the examiner's manual at TMEP § 1202.03, goes to the heart of what a trademark is. A trademark answers the question "who made this?" Decoration answers the question "what does this look like?" When a phrase appears as a large graphic emblazoned across the front of a garment, the Office presumes consumers read it as the design of the shirt, the thing being bought, rather than as the brand identifying who made the shirt. And a phrase functioning as the product rather than the source isn't doing trademark work, no matter how much money it's making you. The doctrine's roots run back decades; in In re Olin Corp., 181 U.S.P.Q. 182 (T.T.A.B. 1973), the Board grappled with exactly this question for designs on t-shirts, and the framework that emerged still governs: placement, size, and context decide whether a use is ornamental or source-identifying.
So how do real clothing brands register marks that also appear huge on the front of products? By using the mark the way the industry marks brands, and documenting it. Think about where you find the brand on any garment you own: the neck label. The woven side-seam tag. The hangtag. The printed interior. The packaging and the point-of-sale. Small, consistent placements in the locations consumers have been trained for a century to check when they want to know who made it. When your mark lives in those places, and your specimen shows it living there, the ornamental problem largely evaporates, and the giant front-of-shirt version becomes what it always was commercially: your brand, also enjoying itself as a design.
Three practical rules fall out of this, and they're worth implementing before you file rather than after you're refused. First, put your mark on the labels and tags from the first production run, even when the front-print is the star; the inside of the collar is where your trademark rights live. Second, when we file, the specimen shows the branded placement, not only the decorated one. Third, if your business genuinely is the phrase, if the slogan is the product, there are recognized paths (a registration for the brand under which the slogan-products are sold; secondary-source strategies), but they're precisely the kind of judgment call that distinguishes attorney filing from checkout-flow filing.
The ornamental refusal isn't a technicality. It's the law asking whether you built a brand or printed a phrase, and the answer is determined by decisions you make at the printer, months before any examiner sees the file. Make them on purpose.
Where you put the name decides whether you own it. TMEP § 1202.03.
The USPTO refuses apparel applications when the mark is used only as decoration, a large graphic across the chest of a shirt, rather than as a brand identifier, and TMEP § 1202.03 is the examiner's playbook for it. The fix is specimen strategy: the mark on the neck label, the hangtag, the packaging, the places the industry puts brands. Where you put the name decides whether you own it.
5. Classes: are you selling products, or running a store?
The classification system trips up product founders in a specific way: the line between selling products and providing the service of selling products. It sounds like philosophy; it's actually fee strategy. Here's the map.
Class 25: the goods on the body. Clothing, footwear, headwear. If you sell shirts, this is the class the shirts live in. It's one of the most heavily filed classes in the entire system, which is both a warning (crowded, per Chapter 3) and a validation (the market knows apparel brands are worth protecting).
Class 3: the goods on the skin. Cosmetics, skincare, haircare, fragrance. The beauty founder's home class. A brand selling both apparel and beauty, increasingly common in the creator economy, is a two-class filing by nature, and that's a budgeting fact better known before the strategy call than during it.
Class 35: the store, not the stock. Here's the distinction that confuses everyone. Retail store services, including online retail, are a Class 35 service, separate from the goods sold in the store. KREYÒL ESSENCE® illustrates the logic from my own clients' portfolio: registered in Class 35 for online retail store services featuring haircare, skincare, and related products. When your brand is the destination, the shop, the storefront, the curated line that includes others' goods or a broad assortment, Class 35 protects the store's name as a retail brand. Many product companies ultimately want both: Class 25 or 3 for what they make, Class 35 for the store they've become.
The supporting cast. Class 18 (bags and leather goods), Class 14 (jewelry), Class 21 (drinkware, a merch staple), Class 9 (the phone cases), Class 16 (printed goods and stickers). None of these is automatic; each is a $350 question answered by where your revenue actually is or genuinely will be.
Two principles govern the mapping, and they're the same ones from every guide in this library because they're true everywhere. Claim the business you have or genuinely intend, because overclaiming creates maintenance problems at renewal (your Section 8 declarations under 15 U.S.C. § 1058 require proving continued use class by class, per the owner's manual), and underclaiming leaves doors open behind you. And sequence by revenue: the class where the money is files first; the aspirational classes wait for the revenue to justify their fees, with intent-to-use filings available when a launch is genuinely imminent. Getting this mapping right is most of what the strategy call decides.
Each class is a separate $350 USPTO fee.
Class 25 covers the clothing you sell; Class 35 covers retail store services, including an online store, especially one selling others' goods alongside yours; Class 3 covers cosmetics and skincare. Most product founders need fewer classes than they fear and different ones than they guess, and each class is a separate $350 USPTO fee, so the mapping is strategy.
6. Specimens for goods: the strictest proof in trademark law
Every registration must eventually be backed by a specimen: real-world proof the mark is used in commerce. And here product founders face a stricter standard than any service brand in this library, because the statute itself draws the line. For services, advertising and marketing showing the mark can suffice. For goods, 15 U.S.C. § 1127 requires the mark "placed in any manner on the goods or their containers or the displays associated therewith or on the tags or labels affixed thereto," with the goods sold or transported in commerce. The ad for the shirt is not use on the shirt.
What passes, in practice: the product photograph showing the mark on the neck label or sewn tag. The hangtag. The packaging, the box, the jar, the tube, with the mark on it. The product insert. And the e-commerce product page, but only when it works as a true point-of-sale display: the mark appearing near a picture of the goods with the means to order them, your listing with the add-to-cart button, not your Instagram grid.
What fails, and fills refusal letters: the mockup or rendering that never existed as a sold unit. The digitally altered photo. The advertisement, lookbook, or press feature, promotion, not placement. The front-of-shirt-only photo, which walks straight back into Chapter 4's ornamental trap. And the specimen showing a different version of the mark than the one applied for, the logo when you filed the words, the old design after the rebrand, a mismatch that quietly undoes applications and, years later, renewals.
Two habits make this chapter cheap instead of expensive. First, build the branding into the physical product from the first run: labels, tags, and packaging carrying the mark, which, conveniently, is the same move that defeats the ornamental refusal, one production decision solving two legal problems. Second, archive as you go: dated photos of labeled products, packaging, and live listings, kept in a folder per mark. That archive is your specimen at filing, your evidence at the Section 8 renewal, and your proof in any dispute, the receipts habit from the monitoring and enforcement guide applied to physical goods.
And a candid note on timing, because product founders often launch before filing: if you're already selling, we file based on use with your real specimens. If you're pre-launch, we file intent-to-use, lock your priority date now, and submit proof when the first run ships, the mechanics covered in the registration guide. Either way, the specimen strategy is decided when we file, not improvised when the Office asks.
Goods demand the mark on the goods, tags, or labels. 15 U.S.C. § 1127.
For goods, the statute demands the mark be placed on the goods themselves, their containers, tags, labels, or point-of-sale displays, with the goods actually sold in commerce. 15 U.S.C. §§ 1051, 1127. Advertising alone fails. Mockups fail. The webpage passes only as a true point-of-sale display. Product founders have the strictest specimen rules in the system, and the cleanest way through is building the proof into the product from day one.
7. Knockoffs, marketplaces, and print-on-demand clones
Service brands get imitated. Product brands get copied, unit for unit, listing for listing, sometimes within days of a design taking off. If your brand sells physical goods, your infringement is the most literal kind the economy produces, and Congress, to its credit, armed product owners accordingly. Here's the arsenal, and what turns it on.
The marketplace layer, where most of the war is won. The platforms where product brands live and die, Amazon, Etsy, eBay, the print-on-demand houses, all operate brand-protection and infringement-report systems, and they are built around one document. Amazon's Brand Registry, the program that unlocks its strongest protection and listing-control tools, is organized around a registered trademark; the reporting lanes everywhere else move dramatically faster when a registration number rides along. For a product founder, this is the single most practical answer to "why register": it's the key that makes the takedown machinery work at the speed knockoffs appear. The clone listing reported with a registration attached tends to be a form and a few days. The same listing, reported by an unregistered brand, is a correspondence.
The counterfeiting hammer. When the infringement crosses from confusingly-similar into counterfeit, the fake labeled as you, the Lanham Act stops asking you to prove your losses. Under 15 U.S.C. § 1117(c), a brand owner can elect statutory damages of $1,000 to $200,000 per counterfeit mark per type of goods, rising to $2,000,000 per mark where the counterfeiting was willful. Congress wrote those numbers for exactly the merchandise economy you operate in.
The standard toolkit behind it. The § 1114 infringement claim with the § 1057(b) presumption doing the heavy lifting; injunctions with the statutory presumption of irreparable harm under § 1116(a); the infringer's profits and up to treble damages under § 1117(a); § 1125(d) and § 1117(d) for the fake "official" store domains, at $1,000 to $100,000 per domain. The full escalation ladder, takedown to letter to proceeding, is the subject of the monitoring and enforcement guide, and every rung of it applies here with extra force because the harm is so tangible: every counterfeit sale is a unit you didn't sell and a quality experience you didn't control, attached to your name.
And the watching, because clones don't announce themselves. Print-on-demand infringement in particular is a volume game: designs harvested by bots, listed across platforms, gone and relisted under new storefronts. The practical counter is the same recurring sweep this practice runs as a subscription: USPTO Watch at $49/month (daily register monitoring, including alerts when examiners cite your mark to block someone else's filing) and Nationwide Watch at $99/month, which adds the monthly nationwide sweep across marketplaces, business filings, domains, and social, misspellings and translations included, built for exactly the storefront-whack-a-mole product brands face. Every hit attorney-reviewed, a written report every month, and the flat $500 member rate on investigated cease and desist letters ($1,200 standalone) when a clone graduates from takedown to target.
Product brands face the most literal infringement in the economy: counterfeit goods and print-on-demand clones. Registration unlocks the tools sized for it: marketplace brand programs (Amazon's Brand Registry is built around a registered mark), § 1114 claims, statutory damages up to $2,000,000 per mark for willful counterfeiting under § 1117(c), and cybersquatting remedies. Unregistered product brands fight the same war with paperwork instead of weapons.
8. From first drop to product empire
Here's the pattern I've watched across a decade of product clients, and it's worth knowing at the first drop, because the founders who end up owning empires made the same early moves.
The line extends, and the portfolio extends with it. The t-shirt brand adds cut-and-sew. The haircare line adds skincare, Class 3 growing a second wing. The merch table becomes a storefront, and Class 35 enters the conversation. Each extension is a filing decision made when the revenue is real, the sequencing discipline from the owner's manual, and the founders who keep the map current never wake up to find a gap where a competitor built.
The collabs arrive, and they're licenses. The boutique that wants to stock you, the collaboration drop, the overseas distributor, every one of these is a trademark license whether anyone says the word, and the law requires that your licenses come with genuine quality control, because uncontrolled licensing can be treated as abandonment of the mark, the naked-licensing doctrine covered in the monitoring guide. For a product brand this is doubly practical: quality control clauses are also how you keep a licensee's factory from quietly becoming your counterfeit problem.
The exit conversation eventually comes. Product brands sell, to aggregators, to strategics, to the private-label giants. When that diligence folder opens, the questions are this guide's chapters read back to you: Is the name registered in the classes where the revenue is? Are the specimens and renewals clean? Is the enforcement file real? And the transfer itself runs on one statutory rule worth knowing a decade early: under 15 U.S.C. § 1060, a trademark is assignable only with the goodwill of the business it symbolizes, a bare sale of the name risks invalidity, which is why these deals are papered by people who do this and priced off registrations rather than vibes.
Every one of those future moments is cheaper because of a filing made now. If a collab, a distribution deal, or a bigger conversation is already on your table, the IP clauses are a thirty-minute read with me before anything gets signed: a consultation, flat $200.
Product design always needs secondary meaning — Wal-Mart v. Samara.
The product brands that become companies treat the trademark portfolio as infrastructure: new marks for new lines, Class 35 when the store becomes the brand, licensing with real quality control, and assignments that travel with goodwill (15 U.S.C. § 1060). The registration you file this year is the asset the acquisition values later.
9. What protection costs, stated upfront
The same structure as everything in this practice: flat, staged, with a real decision between every step, and sized against what's actually at risk, which for you is measured in inventory.
Step 1: Comprehensive Search & Attorney Review, $679. Federal, state, and common-law search on your brand name, a written report in 3 business days, and 30 minutes on the phone with me: what's clear, what's crowded in your class, and what I'd file. This is the only step you're committing to today, and for a product founder it has a deadline logic all its own: it belongs before the production run, because it's the last moment a conflict costs nothing but a decision.
Step 2: Application Strategy, Drafting & Submission, $1,800 flat. Only if the search says go. The ownership decision, the class map from Chapter 5, the specimen strategy from Chapters 4 and 6, planned at filing rather than improvised at refusal, the drafting, the submission, and tracking through registration.
Step 3: USPTO filing fee, $350 per class. Paid to the government at submission. A two-class brand, apparel plus the online store, or beauty plus apparel, budgets $700 here, which is a number better met in this paragraph than at the invoice.
One class, start to filed: $2,829. Registration takes a year or more, your priority runs from the day we file, and in the meantime the ™ goes on everything while the application works. Measured against your actual downside, a warehouse of relabeled inventory, a marketplace takedown you can't win, a knockout punch from a brand that filed first, it's the least expensive insurance a physical-product business can buy, and the only kind that appreciates.
$679 for the comprehensive search and attorney consultation. If the name clears and you proceed, $1,800 flat for application strategy, drafting, and filing, plus the USPTO's $350 per class. One class, start to filed: $2,829, staged across three decision points, and almost always less than one production run of a name you'd have to abandon.