After enough naming projects you stop seeing unique mistakes and start seeing the same eight, wearing different clothes. None of them come from carelessness — they come from smart founders optimizing for the wrong moment. Here they are, with the reasoning and the fix.
1. Naming for the founding feature, not the company
The classic. You are building an invoicing tool, so you name it Invoicely. Two years later you also do payments, expenses, and payroll, and the name is a cage. Descriptive names feel safe because they explain themselves on day one, but they explain the wrong thing by day one thousand. Mailchimp grew from email into a whole marketing platform precisely because the name never promised only email.
Fix: name the company you intend to become, not the feature you are shipping first. A slightly abstract name you grow into beats a precise one you outgrow.
2. The committee name
Four founders, four favorite words, and the "compromise" is Nexvantage or Synergix — a name no single person would have chosen, assembled from the parts nobody objected to. Committee names are recognizable by their smoothness: they offend no one and excite no one. They are the beige of branding.
Fix: let one person own the naming decision, with input from the rest. Naming by consensus reliably produces the least distinctive option in the room, because distinctiveness is exactly the thing someone always objects to.
3. Invented words no one can say
Coined names are powerful — Kodak, Xerox, Spotify — but there is a failure mode where the invention outruns pronounceability. Hand someone Qhyllo or Xprym and they cannot say it, spell it after hearing it, or find it again. A name that dies on first contact with a human voice is not distinctive, it is inaccessible.
Fix: invent, but keep it sayable on the first try. The test is the radio test — read it aloud once and ask someone to type it. If they hesitate, it is too clever.
4. Dropping vowels a decade too late
Flickr made vowel-dropping look clever in 2004. In 2026 a name like Snpr or Grwth reads as a name that could not afford its own domain, because that is usually why it exists. The style now signals lateness rather than wit, and it forces you to spell the name out every time you say it.
Fix: if the .com of the real word is gone, treat that as a domain problem on its own terms — an alternate extension or a different word — not a reason to amputate a vowel. There is a whole companion piece on naming when the .com is taken.
5. Ignoring trademark classes until incorporation
Founders check the domain and the Instagram handle and consider the name cleared. Then a lawyer, months in, finds a registered mark in their exact class, and the rebrand costs more than the seed round's entire legal budget. A domain being free tells you nothing about whether you can legally use the name in your category.
Fix: run at least a basic trademark search in your actual class early, before the name is on a pitch deck and a sign. You are not clearing it perfectly — you are catching the obvious collisions while changing course is still cheap.
6. Falling for the name before checking the domain
The emotional order is backwards. You fall in love with Halcyon, build a week of mental momentum, and only then discover the .com sold for forty thousand dollars and every reasonable alternative is taken. Now every option feels like a downgrade from a name you never actually could have had.
Fix: check availability while you are still generating — a startup name generator that verifies domains live keeps you from bonding with a name you can't have. Falling for a name you can register is far cheaper than falling for one you can't.
7. Names that can't survive a pivot
Related to the feature trap, but about direction rather than scope. GroceryRun is a great name until you become a general delivery company, and Cloudbooks is perfect until you move off the cloud narrative entirely. Names welded to a category, a technology, or a moment become liabilities the day the strategy moves — and startup strategy moves constantly.
Fix: pressure-test a finalist against your own roadmap. Ask what happens to the name if the boldest item on it comes true. If the name breaks, it is too literal.
8. Sound-alike collisions with an incumbent
You name your fintech Stripey or Pled, and every conversation opens with "like Stripe?" or "is this Plaid?" You have handed a larger competitor free real estate in your customer's head. Sounding adjacent to a leader feels like borrowing credibility; it actually confirms that you are the echo, not the source.
Fix: say your shortlist out loud next to the three biggest names in your category. If any of yours makes someone think of a competitor, cut it. You want to be remembered as yourself, not as a near-miss for someone bigger.