Brand Naming

When to use this

Use when a name must be created, chosen, or defended. Use before committing to trademark spend, so that legal effort goes toward candidates that survive behavioral testing.

Do not use it as a substitute for trademark, linguistic, or regulatory clearance. This skill produces creative screening and behavioral evidence. It does not produce clearance, and saying otherwise exposes the client to real cost.

Inputs

The chosen territory and the position it expresses. Also: the markets and languages the name must work in, the product architecture it must extend across, the audiences who will say it aloud, existing names in the portfolio it must sit beside, and any constraints already known — established marks, rejected directions, regulatory limits.

Ask the user for the markets and the product roadmap explicitly. A name evaluated for one market and one product is not evaluated.

Mission

Create names that work in mouths, in messages, in interfaces, in conversation, in search boxes, in sales calls, and in a product architecture that does not exist yet.

The best name is not the one with the best origin story. It is the one whose real-world behavior supports the strategy when nobody is there to explain it.

Method

1. Write the naming brief. Strategic role, audience, markets, tone, expansion needs, constraints. Establish what the name has to do before generating anything.
2. Define naming territories by mechanism, not by word list: descriptive, suggestive, metaphorical, invented, compound, symbolic, vernacular, technical, or culturally grounded. Different mechanisms carry different risks and different ceilings.
3. Generate broadly within each territory. Volume here is legitimate; it is cheap and the yield rate is genuinely low.
4. Cull for obvious collisions and structural weakness before investing attention.
5. Say them aloud. In sentences. On a phone call. Left on a voicemail. Names fail in speech far more often than on a page, and speech is where most of them first travel.
6. Run the behavioral tests in the section below.
7. Shortlist with reasoning, and record what each finalist risks.
8. Register the required professional checks and route them to the right authority.

Behavioral tests

Run these on every finalist. They are the evidence; everything else is opinion.

• First-hearing recall. Say it once in a sentence. Ask for it back an hour later.
• Spelling from sound. Say it. Ask someone to type it. Count the variants you get.
• Search from memory. Would someone who heard it once find the company? Search it and see what they would land on instead.
• Descriptor pairing. Does it survive next to the category word without becoming a sentence?
• Sentence usage. Can a customer use it as a verb, a subject, an object? Does it sound natural in a complaint as well as in a compliment?
• Social introduction. Would someone say it in a meeting without flinching or explaining it?
• Product extension. Add three plausible future products. Does the architecture hold?
• Language and cultural screen. In every target market, by someone who speaks the language. Your own reading of a language you do not speak is not a screen.

Strong versus merely clever

Test the name without its backstory. If it collapses, you have a story, not a name.

Failure modes

• Presenting availability, clearance, or safety as established when nothing was checked.
• Rejecting a name because it feels unfamiliar. Unfamiliar is what a new name is.
• Falling for the name that reads beautifully but cannot be spelled from hearing.
• Ignoring the second and third products the company will ship.
• Scoring candidates numerically and choosing the highest total, which reliably selects the inoffensive middle.
• Screening a language you do not speak.
• Letting the founder's favorite skip the behavioral tests.

Never claim without evidence

Never state that a name is trademark-available, legally clear, culturally safe, linguistically inert, or unregistered in any market. You have not checked, and these are the claims with direct financial and legal consequence.

Where a check has been run — a domain lookup, a search of a public trademark register, a native speaker's read — record exactly what was checked, in which jurisdiction or language, on what date, and by whom. A public register search is a screen, not an opinion on registrability.

Route trademark clearance to counsel, linguistic screening to native speakers in-market, and regulatory naming constraints to whoever governs the category.

Output

Produce naming-report.md using the structure in references/naming-report.md.

Verification

• Run at least one blind spoken test and one written recall test with people outside the project.
• Search each finalist in the real markets and channels the brand will use, and record what a customer would actually find.
• Read the risk register back and confirm every unverified claim is labeled as unverified with a named owner for the check.
• Counterfactual pass. For each recommendation: if it were removed, what meaningful outcome becomes less likely? No credible answer means cut it, downgrade it, or reframe it.
• Fresh-eyes pass. Read the output as someone who must act on it and was not in the process. Cut process debris, duplicated findings, unearned certainty, and any sentence that impresses without changing a decision.

Complete when

The shortlist contains genuinely different mechanisms, each finalist has behavioral evidence rather than advocacy, and every unresolved legal, linguistic, or market risk is visible with an owner attached.

Handoff

Verbal Identity inherits the name's rhythm, register, and the sentences it lives inside — a name that resists certain constructions constrains the voice, and that constraint should be recorded here. Identity System inherits its length, letterforms, and lockup demands. Brand Ship will check that the professional clearances registered here were actually completed before launch, and will block on the ones that were not.