The Naming FilesThe Five Types of Startup Names — And Which One to Pick
Cover artwork for The Five Types of Startup Names — And Which One to Pick
namingstartupsbrandingframeworkfounders

The Five Types of Startup Names — And Which One to Pick

A framework for understanding every startup name ever created, what each type costs you, and a clear recommendation for early-stage founders who don't have time for vague advice.

The Naming Desk·February 25, 2026·9 min read

Every startup name that has ever existed — every one, from the $3 trillion company to the failed MVP that shut down after eight months — falls into one of five structural categories. This isn't a marketing taxonomy invented by consultants. It's a description of the finite ways that language can be deployed to label a company.

Understanding the five types won't guarantee you a great name. What it will do is help you make a deliberate choice rather than a default one — and know exactly what you're trading against what when you commit.

Type 1: Invented Names

Definition: A name with no prior meaning in any language, created purely for the brand.

Examples: Kodak (George Eastman's fabrication, deliberately meaningless), Xerox (derived from "xeros," Greek for dry, but abstracted far enough to be functionally invented), Häagen-Dazs (invented by Reuben Mattus in the Bronx to sound Danish, which it doesn't, because Scandinavian languages don't work that way — but it worked anyway).

What you get: Maximum ownership. You can trademark an invented word completely because nobody else was using it before you. There's no prior meaning to conflict with, no competitor who can claim the same semantic territory, and no literal interpretation that will limit your growth. Kodak was not constrained by its name when it moved from cameras to film to photo processing to digital — the name was a blank canvas.

What it costs you: Everything in terms of brand building. Invented names require you to manufacture meaning from nothing, which is expensive in marketing terms and slow in brand-building terms. You cannot rely on the name to do semantic work for you. Your marketing has to do all of it. For a founder with limited resources and a tight timeline, this is a real cost.

Best for: Companies with significant marketing budgets, products with genuinely novel category definitions, or companies that expect to expand into multiple categories over time. Also excellent for companies that are targeting international markets from the start, since invented words travel better across language barriers than descriptive words do.


Type 2: Compound Names

Definition: Two or more existing words combined to create a new proper noun.

Examples: Facebook (face + book, a collegiate directory), YouTube (you + tube, a TV you create), Salesforce (sales + force, CRM with ambition).

What you get: The best balance of clarity and scalability in the naming world. Compound names import meaning from their component words without being trapped by either. "Facebook" borrowed the social intimacy of "face" and the information density of "book" — which is exactly what a social network is. But Facebook didn't have to stay a social network. The name had enough flexibility to encompass advertising, VR, payments, and a corporate identity as Meta when the company wanted to pivot at scale.

The scalability of compound names is often underrated. Facebook. YouTube. Salesforce. These companies grew in radically different directions from their initial products. The names accommodated that growth because the components were neutral enough to stretch.

What it costs you: Originality. Compound names have been the dominant naming strategy since the early internet, and the easiest, most obvious combinations have been claimed. Facebook was 2004; you cannot name your social network Facebook in 2024. And the farther you get from the obvious combinations, the more the name risks feeling forced — like the company worked backwards from the components rather than arriving at a natural brand.

Best for: This is the most broadly applicable naming strategy. It works for early-stage companies because the component words give you semantic footholds that help with discovery. It works for growth-stage companies because the names are scalable. If you're unsure what type of name to pursue, start here.


Type 3: Metaphor Names

Definition: A name borrowed from a person, place, object, or concept that aspires to a quality the brand wants to own.

Examples: Amazon (the largest river in the world — Bezos wanted scale before he had it), Apple (knowledge, creativity, simplicity — and also a fruit, which made it friendly and non-threatening in a category dominated by cold corporate names like IBM and Microsoft), Virgin (Richard Branson chose it because the company was starting fresh in every industry it entered — and because it was slightly provocative, which suited him).

What you get: Infinite conceptual headroom. A metaphor name is a bet on resonance rather than description. Amazon doesn't tell you it's an e-commerce company — it tells you something about the scale of the ambition. That means the name never expires, never becomes too narrow, never contradicts what the company grows into. Apple sold computers, then MP3 players, then phones, then watches, then streaming services. The name accommodated all of it because "apple" never meant "computer."

Metaphor names also give you rich brand territory. The visual identity, the voice, the culture — all of it can reference the source metaphor in ways that compound over time.

What it costs you: Execution dependency. A metaphor name only works if the brand execution is strong enough to draw the connection. Apple works because Apple's design is as beautiful as the company promises. Amazon works because Amazon's scale is genuinely Amazonian. If your product does not deliver on the quality or ambition implied by the metaphor, the name becomes ironic in the worst way.

Best for: Companies with a clear aspirational identity and the budget to execute it. Not recommended for pre-launch products without product-market fit, because you don't yet know what quality you're aspiring to represent.


Type 4: Clipped Names

Definition: A name created by shortening a longer, more descriptive phrase — often an acronym or compression.

Examples: Intel (Integrated Electronics), FedEx (Federal Express), Reddit (Read It — users "read it" before submitting links), Yelp (reportedly a contraction of "yellow pages help," though the founders dispute this).

What you get: Smart compression. Clipped names retain the meaning of their source phrases while becoming pronounceable, typeable, and memorable. Intel tells you immediately: integrated electronics. But "Intel" fits on a chip. "Integrated Electronics" does not. FedEx implies the full promise of the brand — federal reliability, express speed — in five characters. The clipping creates efficiency, and efficiency in naming is underrated.

Clipped names also give you a story, which is a brand asset. When someone asks "what does Intel stand for?" and you explain it, you've created a memory hook. The etymology becomes a piece of brand lore.

What it costs you: Front-loaded explanation. Every clipped name creates a moment where someone asks what it stands for, or worse, mispronounces it because they don't know the source phrase. This isn't fatal — "Intel" has been explained a billion times and it clearly worked — but it's a tax on early brand communications. Every introduction is slightly more complicated than it would be with a transparent name.

Best for: Technical or professional categories where the source phrase has genuine meaning and the clipping creates elegance. Also works well for companies that are comfortable with the explanation-as-brand-story dynamic and can use it in marketing.


Type 5: Founder and Place Names

Definition: A company named after its founder(s) or a specific place with genuine significance to the company.

Examples: Dell (Michael Dell, who started the company in his University of Texas dorm room), Ford (Henry Ford, who built the first moving assembly line), Patagonia (Yvon Chouinard named it after the region of South America he explored as a climber before founding the company).

What you get: Authenticity that cannot be manufactured. In a marketing environment increasingly saturated with "brand storytelling," a founder-named company has something genuinely irreplaceable: a human being who is accountable for the product. Dell isn't a made-up word. Dell is Michael Dell. That personal accountability, real or perceived, shapes how customers relate to the brand.

Place names work similarly — but only when the connection is genuine. Patagonia works because Chouinard actually went to Patagonia, actually fell in love with it, actually built a company around that experience. A startup named "Patagonia" by founders who've never left Brooklyn would just be a lie. The authenticity only exists when the connection exists.

What it costs you: Scale and exits. Founder names are personal in ways that complicate institutional growth. When the founder leaves, is celebrated, or becomes controversial, the name carries the history. Dell survived Michael Dell stepping down and returning. Ford survived Henry Ford's documented awfulness. But the name requires a founder who is either sufficiently admirable or sufficiently anonymous to stay out of the way. The exit challenge is real too: it's harder to sell "Dell" without Michael Dell than it is to sell "Nexify" without whoever built it.

Best for: Founders who are themselves the brand — whose personal reputation, expertise, or story is central to the value proposition. Also appropriate for craft, heritage, or artisan categories where human accountability is the differentiator. Not recommended for founders who expect to exit within five years.


The Clear Recommendation for Early-Stage Founders

Don't bury this in qualifications. Here it is.

If you are pre-Series A, start with Type 2 (Compound Names).

Here's why: compound names give you semantic clarity without locking you in. They're the easiest to trademark, the easiest to explain, and the most scalable of the five types. They work before you have brand authority because the component words do semantic work for you. They work after you have brand authority because they're not too literal to stretch.

If you want to be more ambitious and have the marketing resources to execute it, pursue a Type 3 (Metaphor) with conviction. The upside is higher — infinite headroom, rich brand territory — but the execution requirement is higher too.

Avoid Type 1 (Invented) unless you're well-funded and building a genuinely novel category. The blank-canvas advantage is real, but the marketing cost is steep and the timeline is long.

Use Type 4 (Clipped) only if you have a specific source phrase with genuine meaning and the shortening creates genuine elegance. Don't clip for clipping's sake.

Consider Type 5 (Founder/Place) only if you are the brand — if your personal identity is genuinely the differentiator and you're planning to be the public face of the company for the foreseeable future.

One last thing: whatever type you pick, check availability before you decide you love it. The best name in the wrong category is the wrong name.

Done reading. Ready to ship?

Name your startup →

Try kickass.name free →