How to Name a Brand: The Five Tests a Name Has to Pass

A company name is the most permanent decision in branding, and the one founders take least seriously. Here is how to name a brand, and the five tests a name has to clear.

When my son was born, my wife weighed the usual questions: meaning, family history, cultural resonance. I brought one most parents never consider. I wanted the first name and the middle name to be five letters each.

His name is David Johan. Five and five. Five-letter names sit at the edge of what the eye reads as a single shape, the way you read a logo. Longer, and the brain breaks it into syllables. Shorter, and it loses presence. The name works in two languages, and I bought his domain the day we chose it.

I ran a naming firm's process on my own kid. Picture what I think about the name a founder gives a decade of work in a tired weekend session.

A name outlasts everything else you build. Logos get redrawn. Taglines get rewritten. Sites get relaunched. It rides through every chapter, and it is the hardest thing to swap once the company has weight.

The most-used word your company owns

The name shows up on every invoice, every email signature, every job listing, every product page, every search result. Win, and it gets spoken and typed billions of times. It is also a trademark, defensible property, but only if you chose something distinct enough to defend. The trademark office and the customer's memory ask the same thing: could this be confused with something already out there?

Here is the test I use. Read your name aloud, inside a sentence, to a stranger. If they write it down right the first time and find your site with one search, it works. If they ask you to spell it or repeat it, the name leaks energy every time someone reaches for you.

That leak compounds. Misspelled press. Customers who type a variant and land on a stranger's homepage. Trademark conflicts you trip over three years in, when the name is too expensive to change.

A name is infrastructure.

The five tests every brand name has to pass

Most naming frameworks run fifteen or twenty criteria long, too many to hold and too soft to apply. So I cut a tighter set at the studio, refined across hundreds of candidates. Every name we recommend clears all five.

• Distinctiveness. Could the name be confused with anything in or near your category? If so, your marketing budget builds equity that drifts to a competitor.
• Ownability. Can you actually hold it: the domain, the handles, the trademark, the search results? A name you cannot own is one you spend years apologizing for.
• Scalability. Does it still fit when you are ten times bigger, running three product lines, on two more continents? Names that hug the current product trap you. Names that stay generic vanish.
• Pronounceability. Can a stranger say it right the first time, and spell it after hearing it, in your customer's language and a couple more? Most billion-dollar consumer brands run one or two syllables for a reason.
• Narrative load. Does the name carry meaning or mood that helps the brand work? The strongest ones pre-load the customer with an idea before a dollar of marketing has run.

Four out of five is a near miss. Five out of five is the name.

Where good names come from

Three sources, in order of value.

• Your customers' own words. Read your sales transcripts, reviews, and support tickets. Some of those words, recombined, become candidates. The customer half-names the brand for you, and most founders never read their own evidence.
• Adjacent vocabulary. The obvious category words are taken. In cannabis, "high," "smoke," "green," and "bud" are mostly claimed. The good names live one step off the obvious. Knack holds no cannabis word. Liquid Death holds no water word.
• Constraints over freedom. Open sessions fail: five hundred candidates, no filter. Set hard limits up front: syllable count, must work in Spanish and English, no animals, no Latin roots, a real word or something close. Then generate inside the fence.

Run the brand naming exercise this week

Block two hours. Print your last fifty reviews, tickets, or sales calls, and read them with a highlighter. Mark every word a customer uses to describe what you do or how you make them feel. Ignore names. Collect words.

Two hours later you have a vocabulary, maybe thirty words, maybe a hundred. Cluster by theme. Then write five constraints: how many syllables, what languages it must survive, what it cannot be, what it must imply.

Now generate for an hour. Aim for fifty candidates, bad ones welcome. Volume is the point. Run them through the five tests. Most die at distinctiveness or ownability. Take three survivors to your team and your customers before you commit.

This will not replace a naming firm at the scale that warrants one. It replaces the version most founders run: three people guessing in a Slack channel.

Four names from my studio

Four projects, four principles.

• Knack. A cannabis brand in New York. Four letters, percussive, idiom-loaded. The name carries a built-in sentence, "you've got a knack for this," that sells before any marketing runs. Lesson: narrative load is free advertising every time the name gets said.
• Onspire. A healthcare marketing agency named after four companies merged into one. A portmanteau of "on" and "inspire," built to read as one word that signals momentum. Lesson: a name can be corporate diplomacy, making a merger feel like a future.
• Melo. A fermented hand soap meant to feel calm and clean. It reads as "mellow" without spelling it, and says easily in any language. The same operator owns Risewell, "rise" plus "well." Lesson: suggestive names beat descriptive ones, and good naming logic travels a portfolio.
• Mortality Index. A data project we built that maps what we fear against what kills us. The name is the proposition: a Bloomberg terminal for death, and the product delivers exactly that. Lesson: sometimes the strongest name states the category so plainly the customer cannot look away.

Naming is a strategic act in disguise

The five tests are mine, but the lineage is real: Marty Neumeier's SUFP test, Steve Manning's typology at Igor, David Placek's sound symbolism at Lexicon (Pentium, BlackBerry), and Ries and Trout's strategic frame. All four agree that naming is strategy.

Naming breaks in three ways. Founder vanity, which traps the company inside the founder's biography. Description-as-name, "Premier Marketing Group," which commodifies the brand on day one. The unpronounceable invented word, where someone confused making up a word with picking a good one. All three forget the same fact: the name has to work in the world, for everyone who owes the founder nothing.

The names that win do strategic work while the customer believes they are only remembering a word.

Choose it like you would name a child

Run the five tests. Do the customer-language exercise. Generate inside constraints. And if you are tempted to skip it because you already half-like a name, sit with the math: fixing a name after the company has grown runs three to twenty times the cost of getting it right now. Every founder I have walked through a rebrand wishes they had taken the name seriously the first time.

I gave my son a name that is easy to say, balanced across two languages, and ownable as a domain. Five letters and five letters, the discipline I would bring to a billion-dollar company. He will carry it for life, and I am glad I took it seriously on the first try.

This is the written version of Position to Win, Episode 5.