How to Name Your Business in 2026 (Without Getting Sued or Stuck With a Bad Domain)

Picking a business name feels like a creative exercise. It is mostly a legal and financial one. The cost of getting a name wrong is not “we have to think of another one.” The cost is somewhere between $5,000 (a forced rebrand at month 6, redoing logos, business cards, and Stripe metadata) and $50,000+ (a cease-and-desist letter from a trademark holder you did not know existed, plus a forced rebrand under deadline pressure).

This is a checklist you can run in 60–90 minutes that catches the expensive mistakes before you commit. It is not legal advice — for the final clearance you want a trademark attorney — but it filters out the 80% of names that should never have been on your list.

Why naming is a financial decision, not a creative one

Think of every brand asset as a sunk cost compounding from day one of launch. The longer you operate under a name, the more it costs to leave it:

• Domain authority. Every backlink you earn points at one domain. A rebrand restarts that compounding.
• Stripe / payment metadata. Charges show up under your business name on customer statements. Customers who do not recognize the new name file disputes.
• Trademark precedent. The longer someone else has used a similar name, the harder your defense becomes if it goes legal — the standard is “likelihood of confusion in commerce,” and time-in-market is a major factor.
• Search-engine equity. Your brand-modified queries (your name + everything) become a meaningful share of organic traffic by year two. A rebrand resets that to zero.
• Customer recognition. Every email, invoice, support reply, and social post has been signed in the old name. People remember the old one.

Use the business valuation calculator to estimate what those assets are worth at your stage. For a $300K-ARR SaaS, the brand equity component alone is often $50K–$150K of the multiplier — meaning a forced rebrand can wipe out a year of growth.

The point: spending an extra 90 minutes upfront to validate the name is the cheapest insurance policy in the entire startup playbook.

Step 1 — Brainstorm wider than you think you need to

Aim for a longlist of 30–50 candidate names. Most founders stop at 5 and pick the least-bad. The reason for the longer list is not that 50 are usable — it is that the ones that survive Steps 2 through 5 are usually not the names you would have ranked highest on creative merit.

Generation tactics that work:

• Describe the product in a sentence. Pull out the 3 most distinctive nouns and verbs. Combine them.
• Look at related words in Latin, Greek, Old English. They often produce shorter, more available candidates.
• Pick a 2-syllable invented word with strong consonants (b, k, p, t). They register cleanly verbally and are usually available across registries and domains.
• Pair a real word with a short modifier (“Stripe” → payment infrastructure; “Linear” → project management). The modifier-pair pattern still has a lot of available combinations.

Avoid:

• Common dictionary words alone. “Pulse,” “Spark,” “Boost” are all in 200+ trademark conflicts each. Pick a less-saturated word or modify it.
• Hyphens, numbers, or unusual spellings. They will haunt you every time you say the name out loud or read it back to a customer over the phone.
• Geography you might outgrow. “Berlin Pricing Co.” is fine until you sell in the US.
• Words that translate badly. “Nova” in Portuguese means “doesn’t go” if you read it as “no va.” Do a 30-second check on translations into your top 3 target markets.

Step 2 — Run a basic trademark search (3 registries, free)

This is the step founders skip most often, and it is the one with the highest financial risk if you skip it. You do not need an attorney for the first-pass check — the official registries are searchable for free, and most disqualifying conflicts show up in the first 60 seconds.

For each candidate name, search:

• USPTO TESS (United States): tmsearch.uspto.gov — free public database. Search for exact match and “sounds-like” variants in your industry’s class (e.g. class 9 for software, class 35 for advertising/business services, class 41 for education).
• EUIPO eSearch (European Union): euipo.europa.eu/eSearch — covers all EU member states with a single registration.
• UK IPO (United Kingdom): trademarks.ipo.gov.uk — separate from EUIPO since Brexit. Required if you’ll do business in the UK.

Three categories of result:

1. No matches in your class. Green. Move to Step 3.
2. Matches in your class. Red. Even if the existing mark is a small operator, the cost of a conflict letter is high. Drop the name.
3. Matches in adjacent or unrelated classes. Yellow. May be defensible, but get a trademark attorney to opine before committing.

Three regions catches roughly 80% of the conflicts you’d face as an early-stage company doing business in English-speaking markets plus the EU. If you’re targeting Asia (especially Japan, China, South Korea) you need additional regional searches — the registries there are not in English and are harder to navigate without help.

Step 3 — Check domain availability (the right way)

Most founders check .com only and stop. That’s a mistake. The pattern that survives in 2026 looks like:

• .com available, exact match. The strongest signal. Get it.
• .com parked / for sale at a reasonable price ($1K–$10K). Still acquirable. Budget for it.
• .com is a real business in another industry. Workable but not ideal — you’ll lose direct-navigation traffic and your customers will mistype.
• .com is a competitor. Kill the name. You will spend years explaining you’re not them.

Beyond .com, also confirm:

• .co, .io, .ai — secondary defensible domains. If .com is taken but the rest are clean, you can still launch on .co or .io and acquire .com later.
• Country TLDs in your launch markets — .de, .uk, .fr if relevant. Cheap to register, expensive to lose to a squatter.
• The exact spelling on common typos. Register the obvious typos to prevent phishing.

A common rule: if the .com is unavailable AND no clean alternative exists for under $5K acquisition cost, the name is not viable. Do not launch on a third-tier TLD with a competitor sitting on .com.

Step 4 — Lock down social handles (all of them, on day one)

Every platform has handle squatters. The day you decide on a name, register the handle on the platforms that matter for your audience — even ones you don’t plan to use yet. This is a 15-minute task that prevents a real headache 6 months later.

Minimum platforms in 2026:

• X / Twitter
• LinkedIn (company page)
• GitHub (organization)
• Instagram
• YouTube
• TikTok (if you have any chance of using video)
• Reddit (the subreddit name, even if you never moderate one)
• Bluesky and Threads (post-Twitter alternatives gaining founder mindshare)

If 3+ of these are taken in your name, that is a signal the name is too generic. Reconsider.

Step 5 — Run the linguistic and brand-feel checks

Before committing, do these final checks:

• Say it on the phone. Call someone, pronounce the name, ask them to spell it back. If they get it wrong twice, the name is too hard to dictate.
• Type it on a phone keyboard. Are the letters far apart? Names that are hard to type lose people who would have searched for you.
• Look for negative associations. Search the name in news from the last 5 years. If the first page of results is a scandal or lawsuit at a similarly-named company, your customers will see that too.
• Check pronunciation in your top 3 target markets. A name that means nothing in English may have an obvious meaning in Spanish, German, or Mandarin.
• Read it as a verb. “Let me Stripe that” works. “Let me [your name] that” should not be embarrassing.

If a name passes all 5 steps, you have a name worth committing to. If 3+ candidates pass, pick the shortest one — fewer characters means fewer typos, easier verbal sharing, cleaner logos, faster recognition.

A faster version of this — built by us

Disclosure: Nombrio is built by sydacos GmbH — the same team behind EconKit.

Steps 2 through 4 are the slow parts. A trademark attorney typically charges $2,000–$4,000 to do a clearance search across the major registries plus domain and social availability — and they take 1–3 weeks to come back with results. For early-stage founders running multiple candidate names, the math doesn’t work; you can’t pay $20K to clear 5 names before you’ve even validated the idea.

We built Nombrio to compress that work into a single report covering USPTO, EUIPO, and UK IPO conflicts, plus domain availability across major TLDs and social handle status. It can also generate name candidates for Step 1, but the trademark + availability check is the real value.

Single-report pricing starts at $9.99 — good enough for first-pass screening that filters 80% of bad candidates. For the final commitment, still get an attorney to do a formal clearance opinion. Nombrio is the cheap fast filter you run on the longlist before you ever pay for legal hours.

What to do if you’ve already named it (and want to know if it’s safe)

If you’re past the naming stage and now wondering whether your name is defensible, run the same Step 2 checks. The cost of finding out you have a problem now and pivoting is roughly an order of magnitude lower than the cost of getting a conflict letter from someone else’s lawyer in year two.

A name is one of the few startup decisions where the cheap, boring, slightly paranoid path is also the high-EV path. Spend the 90 minutes. Write the longlist. Run the registries. Pick the name that survives the gauntlet — not the one you fell in love with on day one.