10 myths about business and product naming

The verdicts at a glance

The assumption	Verdict
A name should describe the business	Mostly wrong, and legally the weakest option
Invented words are risky	Wrong. They are the strongest marks and the costliest to explain
A free domain means a usable name	Wrong. A registration grants no trademark rights
A name that passed the checks is protected automatically	Wrong. Checks do not create or confirm trademark rights
Which name sounds better is pure taste	Wrong. Sound carries associations that replicate across cultures
You will know the right name when you hear it	Wrong, and it is why naming meetings stall
A workshop is how you find the name	Half right. It produces candidates, not a decision
Let customers vote on the shortlist	Wrong. A vote measures familiarity
The name can be translated for other markets	Wrong. Translation is not the test that matters
The name is the brand	Wrong. It is the container the brand goes into

Myth 1: "A name should say what we do"

This is the most common instinct and the most expensive. Trademark offices rank marks on a distinctiveness spectrum, and the United States Patent and Trademark Office puts descriptive marks in the weak half, registrable only in limited circumstances such as acquiring distinctiveness through years of use. Competitors can describe themselves the same way, so a descriptive name buys comprehension and pays for it in both differentiation and legal protection.

Myth 2: "Made-up words are too risky"

The same spectrum puts fanciful marks, meaning invented words, at the strong end. A coined name is the easiest thing to register and the easiest domain to secure. Its cost is real but it is a marketing cost, not a legal one: an invented word means nothing until you teach it. Founders routinely reverse these two risks and choose the weak option for safety.

Myth 3: "The domain was free, so the name is ours"

The USPTO states it flatly: registering a domain with a registrar gives you no trademark rights, and you may later be required to surrender it if it infringes someone else. A registrar sells you an address, not a position. Domain strategy and legal availability are two questions, and answering the cheap one tells you nothing about the expensive one.

Myth 4: "It passed the checks, so it is protected"

NamingStudio.ai checks name suggestions for basic availability and for room to stand apart from competitors, and runs a cultural check, a language check and a relevance check within the target audience. Trademark rights and register entries cannot be checked automatically. We therefore strongly recommend that you check the suggested names yourself for existing rights and for their eligibility for trademark protection, and, if needed, register and record the name as a trademark. NamingStudio.ai cannot do this automatically either. It is a manual process, and you are responsible for it as a user of the platform.

Myth 5: "Which name sounds better is just taste"

Sound is the one layer with real experimental evidence behind it. The bouba/kiki effect, where people reliably match rounded sounds to rounded shapes and sharp sounds to jagged ones, replicates across cultures and writing systems. Phonaesthetics is not a law you can apply mechanically, and it does shift between languages. It is also not nothing, which is what "just taste" implies.

Myth 6: "We will know the right name when we hear it"

You will know when you like one. That is a different sentence. Without written criteria, the decision has no evidence attached, so the loudest or most senior person in the room becomes the deciding mechanism. The professional move is unglamorous: write down what the name has to achieve before generating anything, in the naming brief.

Myth 7: "A workshop with the team will find it"

A workshop reliably produces candidates and reliably fails to produce a decision. It also produces the wrong shape of argument, because everyone present is defending a favourite. Volume is what makes elimination mean something: a shortlist drawn from twenty candidates is a shortlist of the least bad, and a longlist handed over intact is the work passed back to you.

Myth 8: "Let the customers vote"

Ask people to pick a favourite from a list of unfamiliar words and they pick the one that feels most like names they already know. A vote measures familiarity, and a new name is unfamiliar by definition, so the method quietly selects for convention. Structured reaction from a defined audience is genuinely useful. See persona validation for what that looks like when it is done properly.

Myth 9: "We will translate it for the other markets"

A name is not translated, it is re-examined. The question is not what the word means in the dictionary of each market but what it carries: taboo readings, religious associations, numbers and colours that mean something local, and collisions with brands already there. Cultural screening is a separate stage from language checking for exactly that reason, and it has to run before anyone has a favourite.

Myth 10: "Get the name right and the brand follows"

A name is a container. It sets a ceiling on how easy your brand is to talk about and it cannot fill itself with meaning. This is the myth worth holding on to in the gentlest form, because the opposite error is just as common: treating the name as interchangeable, then discovering that distinctiveness is measured against whoever is already standing in your category.

So what is the actual advantage agencies have?

Not better taste, and not a secret word list. The advantage is procedural, and it is almost boring when you see it written down.

They read the category before inventing anything, which is what converts a preference into an argument that survives a meeting. They write the judging criteria first, so the conversation is about evidence rather than who feels strongest. They name the creative directions before the words, so a client can reject a whole direction cheaply instead of voting on individual candidates. They generate far more than they will ever show, because elimination needs something to eliminate. And they run the screening during the cut rather than after it, while a problem is still cheap to find.

What the client actually leaves with is not the word. It is the reasoning they will need in the next room, when someone senior asks why this name. That is the transferable asset, and it is the thing a list of suggestions cannot be.

Why this stayed behind a paywall

Published rates for a full naming engagement run from around $15,000 for a single naming subject to $150,000 and beyond at enterprise scope. The method was never the secret. The hours were what kept it inside companies that could fund them.

That is the part worth automating, and it is what NamingStudio.ai does: the category read first, candidates built against named directions, screened across languages and cultures, then cut hard, with the reasoning attached to every name that survives. If you are naming something right now, look at how the process runs and judge it against this list.