You're six months in. The store converts, the ads are dialed, customers know your name. Then a certified letter shows up. Cease and desist: another company owns a mark close enough to yours that a customer might confuse the two, and they want you to stop using it. The name on your storefront, your packaging, your handles, and your ad account just turned from an asset into a liability.
Many of those conflicts were sitting in a public database the whole time, and a short search before launch can surface them. It's easy to skip that search. A founder checks whether the .com is free, sees that it is, and calls the name cleared, but a free domain says nothing about trademark risk.
This is a pre-launch review that catches what a domain check misses. Twelve checks across four areas: trademark, digital availability, distinctiveness, and documentation. On a single brand it takes roughly an hour and a half. Do it before you print anything, buy anything, or announce anything. Filing your own trademark is a separate decision for later. The review's only job is to help you decide whether that decision is worth exploring at all: it is not a legal clearance.
What a forced rebrand costs #
A rebrand forced by a conflict has costs spread across many places, which makes them easy to underestimate. You pay for redesigned assets and a new domain. You also lose the SEO equity built on the old name, the recognition customers already had, and the momentum of a launch you now have to run a second time.
The total depends on how far the brand has already spread, and dollar figures you see online for it are estimates, not measurements. The direction is what matters: rebuilding a live brand tends to cost far more than screening a draft one.
The screening side is close to free. The USPTO's Trademark Search costs nothing. Many visual search tools have free tiers. Domain and handle checkers are free. An hour and a half weighed against a rebuild that can take weeks is a lopsided trade.
Trademark coverage #
This is where the costly misses hide. Many founders check the brand name as a wordmark and stop there. That leaves three other surfaces where a conflict can quietly live.
1. Visual trademark search #
This check catches look-alikes the others can't. A visual search compares your logo against the imagery inside registered marks, so it can flag a similar design even when you have no idea what the other brand is called.
Since April 2, 2026, the USPTO's Trademark Search has offered a beta image search: log in to your USPTO.gov account, select "Image search," and upload the logo (USPTO: search system updates). The USPTO calls it "a supplemental strategy" and says design search codes "may deliver different results than the AI image search, so it's best to use both methods." Design-code and design-description searches are covered on its design code page (USPTO: design search codes). Running both, plus another visual search tool if you like, covers more ground than any one of them alone. Nice classes can help you organize and prioritize what to read first, but whether a look-alike is an actual conflict turns on the relatedness of the goods or services and the overall likelihood of confusion, not the class number alone. A quiet result here is a reason to keep going, not a clearance.
2. Wordmark search #
Search the USPTO's Trademark Search for your exact name. Two things should prompt a closer look: a live mark using the same name for related goods or services, and any pending application for a similar name. A famous registered mark whose owner is known to enforce is a reason to get professional advice before going further rather than something to design around on your own.
3. Sound-alikes and spelling variants #
The USPTO weighs how marks sound, as well as how they look and what they mean, when it judges likelihood of confusion, and its own example treats "T. Markey" and "Tee Marquee" as confusingly similar because they could be pronounced the same way, even though they're spelled differently (USPTO, likelihood of confusion). Trademark Search won't find every sound-alike for you. Its combined field covers USPTO-assigned pseudo marks for some spelling variants, and for phonetic equivalents and alternate spellings the USPTO points to regular-expression searches (USPTO: transitioning from TESS). Cover it by hand with wildcards and variant spellings: search the stem with a wildcard, then run the obvious misspellings, the plurals, the dropped vowels, and the swapped endings.
If your name uses an invented spelling like "Lyft" for "Lift" or "Tumblr" for "Tumbler," search the ordinary spelling too. Examiners may read it as the everyday word, and so will your customers.
4. Nice class crosscheck #
Nice classes organize a search; they don't by themselves determine whether two marks conflict. The same word can appear in different industries: Delta Air Lines and Delta Faucet each use "Delta," filed in different classes (Class 39 for transportation and Class 11 for plumbing fixtures; for example, USPTO Reg. Nos. 0654915 and 4638296), and the USPTO itself uses Delta faucets and Delta air transportation as an example of identical marks on unrelated goods (USPTO, likelihood of confusion). What lets marks coexist is that their goods and services aren't related enough to confuse customers, not the class number itself. Work out the one to three classes your brand will operate in to focus your searches, but judge any overlap by relatedness of the goods or services and likelihood of confusion (USPTO, likelihood of confusion): changing the class you file under does not make a conflict go away.
Digital availability #
Trademark is about legal risk. Availability is about what you can actually hold. A name has to clear both before it's usable.
5. Domain availability #
For a US-targeted brand, many founders still prefer the .com, though alternatives like .co, .io, .store, and .shop work well in the right context: treat this as a preference, not a rule. Check availability first. If the .com is parked at a premium and you'd pay a significant sum to pry it loose (amounts vary widely), fold that cost into your launch math now. If a different company is actively using it, treat that as a flag worth investigating even when your fields don't overlap, because customers who search your name may land on someone else.
6. Social handle parity #
If social presence matters to you, consider locking down the platforms your audience actually uses: for many consumer brands that's some mix of Instagram, X, TikTok, YouTube, and LinkedIn, but the right set depends on your market. A tool like Namechk scans many at once. If a handle you need is held by a semi-active account, decide before launch whether you'll live with a suffix or change the name outright. Don't count on a platform releasing a handle from an existing account; plan around it rather than on an exception.
7. Cultural and language check #
Run your name through the languages of the markets you care about, then have a native speaker check slang and connotation rather than the dictionary line. Far better to notice this now than in a screenshot making the rounds later.
Brand distinctiveness #
Distinctiveness affects how protectable a mark is. Three things to look at before you commit to the name and the look.
8. Logo distinctiveness #
Distinctiveness is judged by the mark in relation to its goods and services, not by whether the typeface looks custom. A wordmark set in a common font like Helvetica, or a three-letter monogram, is not automatically unregistrable: the USPTO allows standard-character marks and does not require a proprietary font; "using standard characters protects the wording itself, regardless of how it is displayed" (USPTO, likelihood of confusion). What raises registration difficulty is a mark that's merely descriptive or generic for its goods, or one confusingly similar to an existing mark: separate questions from "does the type look plain."
As a creative gut-check (not a legal test, and not a USPTO threshold), try the five-brand exercise: go find five existing brands using a similar icon or style. If they surface in a couple of minutes, your design may be hard to make memorable and easy to confuse with others: a reason to push the concept further, not a prediction of refusal. For more on what affects registrability, see How to Check If a Logo Is Trademarked: A US Search Guide.
9. Color and font ownership #
Trade dress can reach past the logo. A single color can itself be a registered trademark once it has acquired distinctiveness through use (Qualitex Co. v. Jacobson Products Co., 514 U.S. 159 (1995)), which is how Tiffany blue, UPS brown, and T-Mobile magenta each became protected marks. Check that your hero color isn't a registered single-color mark in your field. Fonts work differently: typeface designs generally aren't copyrightable in the US (U.S. Copyright Office, Circular 33; 37 CFR § 202.1). And paying a freelancer for custom artwork, such as an icon or an illustration, doesn't automatically make you the copyright owner: a logo isn't one of the commissioned work-made-for-hire categories (U.S. Copyright Office, Circular 30), so ownership moves to you only through a signed written transfer (17 U.S.C. §204(a)). Put ownership in writing.
10. Pattern and illustration originality #
If your brand leans on hero illustrations or repeating patterns, reverse-image-search each one before launch. An illustration "inspired by" something you found online can draw a complaint, or a takedown notice, from the original artist once you're visible enough to be found. This also catches AI-assisted work that happens to reproduce an existing piece, which is a real risk even when the match is accidental (see Who Owns an AI-Generated Logo? US Copyright vs Trademark).
Documentation and defense #
The last two checks aren't about finding conflicts. They're about being ready for the day a copycat finds you.
11. Keep design history and use evidence separate #
Keep two different records, and don't confuse them. First, a design history: versioned logo files, a signed and dated brief, and creation dates on your assets. This shows when you made the artwork. Second, and only once you actually go to market, evidence of use in commerce: dated sales records, invoices, and photos of the mark on real goods or packaging. A timestamped Figma file or a mockup shows design activity, not commercial use, and a mockup, printer's proof, or digitally altered rendering is not a valid specimen if you later file a use-based application (USPTO, specimen refusal and how to overcome it). When it matters, "we've used this logo in commerce since March 2026" needs sales records and product photos behind it, not just a design file.
12. Set up post-launch monitoring #
This is the one many people forget. You launch, you grow, and one day you find someone selling under your name in another market, sometimes with a registration already in hand. Getting ahead of it is cheaper than reacting late. A commercial trademark-watch service can monitor new filings in the markets you care about, and Google Alerts on your name is free. Note the limits of USPTO tools here: the USPTO's TSDR lets you track the status of your own application or registration (USPTO TSDR), but it isn't a service that notifies you of every new third-party filing in your class. Watching your own file is not the same as watching a whole class. Finding a competitor years in, holding rights you can no longer easily challenge, is the costly outcome you're trying to avoid.
After the review #
If the review turns up nothing alarming, that's a reason to keep moving, not a legal all-clear: only an attorney can give you a clearance opinion. The harder questions then move up: file it yourself or hire an attorney, file on which basis, one class or several. For reference, the USPTO base application fee is $350 per class (USPTO fee information), plus surcharges that depend on how you complete the application. On timing and basis, the rule is straightforward: you may file an intent-to-use application under §1(b) when you have a bona fide intention to use the mark, or a use-based application under §1(a) only once you are actually using the mark in commerce (USPTO, intent-to-use applications).
If the review turns up an issue (say a mark in a possibly related field, or a pending application that's close) treat it as something to investigate, and consider professional advice, before you commit. A close conflict for related goods, or an exact wordmark already held by an active enforcer, is a strong reason to pause and reconsider the name now rather than after you've built on it. Changing a name before launch costs one redesign cycle; changing it after can cost much of what you built on top of it.
This is general information, not legal advice, and it's no substitute for an attorney when your situation calls for one. A pre-launch review can reduce the risk of the most expensive mistake an indie founder can make with a brand (renaming it after the world already learned the old name) but it can't guarantee the name is free to use.
Related reading #
- How to Check If a Logo Is Trademarked: A US Search Guide
- Who Owns an AI-Generated Logo? US Copyright vs Trademark
- Trademark Search Tools Compared: Coverage, Features & Costs (2026)
Frequently Asked Questions #
How long does the full checklist actually take? #
For one brand name and logo, the full 12-check sweep takes roughly 60 to 90 minutes once you're comfortable with the USPTO's Trademark Search. First-timers should budget two to three hours for the learning curve. If you launch brands regularly, later runs go faster once the workflow is familiar. These are rough estimates, not guarantees.
Can I skip the visual search if I only have a wordmark? #
For a text-only brand, such as a SaaS launching with no icon, the visual check is lower priority. But if you plan to add a logo within the first year, run it now anyway. Pivoting the name later because your planned logo conflicts is the worst possible time for a rebrand.
What if my brand name is taken in one country only? #
It depends on the country and your launch market. If you're US-focused and the conflict sits in a country you'll never enter, you can set it aside for now. If you expect to expand internationally before long, treat it as an issue to investigate and look into Madrid Protocol or national filings in your target countries, ideally with counsel.
Do I need a lawyer to run this checklist? #
No. The screening itself is built for founder DIY and doesn't require credentials. Filing is different: the downstream decisions (filing strategy, responding to an opposition, planning international expansion) benefit from counsel. And if your permanent legal residence or principal place of business is outside the United States, the USPTO requires you to be represented by a US-licensed attorney for all trademark matters, so self-filing is not an option in that case (USPTO foreign-applicant rule).
Should I file the trademark before or after launching? #
Either can be correct, depending on your facts. If you have a bona fide intention to use the mark but aren't using it yet, you can file an intent-to-use application under §1(b). Once you are actually using the mark in commerce, you can file a use-based application under §1(a), and only then, because a use-based filing requires genuine use (not token use) and a valid specimen (USPTO, intent-to-use applications; USPTO, specimen refusal). There's no fixed schedule every founder follows; match the basis to your actual situation.
What if a social handle I want is taken by a dormant account? #
In practice, treat a dormant handle as taken. Some platforms have programs that release inactive handles, but they are limited and not something to plan a launch around. Assume the handle is unavailable rather than betting on an exception.
