You want a name that fits on a business card, sounds sharp in a pitch, and doesn't require a pronunciation guide. Short startup names feel inevitable in hindsight. Stripe, Figma, Notion, Loom. Four to six letters, one or two syllables, impossible to forget.
Then you search and discover every obvious combination is parked, priced at $8,000, or owned by a domain squatter in Shenzhen. The 5-letter .com namespace has been strip-mined for two decades. That doesn't mean you're out of options. It means the obvious sources are dead and the interesting ones are hiding in plain sight.
This guide covers where to look, what to compromise on, and how to evaluate short brand names without falling into the trap of settling for something unpronounceable.
Why Short Names Still Matter
Short names reduce cognitive load. A founder says the name once, and the investor remembers it. A user types it in four keystrokes instead of fourteen. A designer fits it on a mobile nav bar without truncation.
But the real advantage is flexibility. A five-letter name can grow into an enterprise brand or stay scrappy. It doesn't box you into a category. Compare Stripe to OnlinePaymentSolutionsHQ. One works on a credit card reader. The other works nowhere.
A short name isn't about character count. It's about how many syllables it takes to say and how many attempts it takes to spell after hearing it once.
Figma is five letters and two syllables. Xqzt is four letters and zero syllables, because nobody can pronounce it. Length matters, but pronounceability matters more.
Where the Obvious Sources Fail
Most founders start with a thesaurus and a domain registrar. They type in spark.com, pulse.com, drift.com. All taken. They add a letter. sparky.com, pulsee.com, driftly.com. Taken or priced at five figures.
The pattern repeats until frustration sets in and they either:
- Add a suffix that makes the name worse (
getsparkly.com) - Switch to an obscure TLD (
.ly,.io,.co) that confuses users - Settle for a name they don't actually like
The problem isn't the search. It's the source. English dictionary words under seven letters are gone. That's a fact. But the English dictionary was never the only source of good names.
Source 1: Foreign Language Roots
Some of the best short brand names come from languages other than English. They're short, they're real words, and their .com domains often sit unregistered because English-speaking founders never think to look.
Examples of the pattern:
Vela(Spanish/Italian for "sail") — four lettersKuma(Japanese for "bear") — four lettersNido(Spanish/Italian for "nest") — four lettersSora(Japanese for "sky") — four lettersToro(Spanish/Italian for "bull") — four letters
The process is straightforward. Pick a concept relevant to your product. Translate it into five or six languages. Check the .com for each result. Many will be taken, but the hit rate is dramatically higher than English-only searches.
The risk is cultural. A word that sounds elegant in Japanese might mean something unfortunate in Portuguese. Run every candidate through a native speaker or at minimum a multi-language dictionary check.
Source 2: Word Fragments and Truncations
Not every good name is a complete word. Some of the most memorable short startup names are fragments that suggest a longer word without spelling it out.
Notion is a full word. Figma is not. Loom is a full word. Vercel is not. The fragments work because they're close enough to trigger recognition without being literal.
Techniques that produce usable fragments:
- Take a longer word and cut the ending:
catalyst→cataly,horizon→horiz - Take a longer word and cut the beginning:
element→ement,velocity→ocity - Merge two short words and drop letters:
code+pilot→codot,data+orbit→datorb
The output will be hit or miss. Most fragments sound broken. But when one works, it feels distinctive in a way dictionary words rarely do.
Fragments work best when they preserve the stressed syllable of the original word. Vercel keeps the emphasis of versatile. Figma keeps the shape of figure. If you can't hear the parent word in the fragment, neither will your users.
Source 3: The Two-Syllable Compound
Compounds are two short words pushed together. The trick is that the best ones aren't obvious. Facebook is obvious. Snapchat is obvious. Those are gone.
The non-obvious compounds are still available because they combine words from different domains:
Mailchimp— postal + animalPalantir— mythology + techShopify— commerce + suffixGrammarly— grammar + suffix
For 5 letter startup names specifically, the compound approach gets harder. Five letters doesn't leave room for two full words. But it leaves room for a word plus a fragment:
Paypal— word + fragment (6 letters, but the pattern holds)Dropbox— word + word (7 letters)Canva— fragment + vowel
The realistic play for five letters is a fragment plus a vowel or a fragment plus a single consonant. Loom is a full word. Luma is a fragment plus a vowel. Both work.
Evaluating Short Brand Names: A Framework
Finding candidates is step one. Filtering them is where most founders get stuck. They fall in love with a name that's hard to spell, hard to say, or legally risky.
Here's a scoring table that forces objectivity:
| Criterion | Weight | What to Check |
|---|---|---|
| Pronounceability | 30% | Can a stranger say it after reading it once? |
| Spellability | 25% | Can a stranger spell it after hearing it once? |
| Domain availability | 20% | Is the .com available or affordable? |
| Trademark risk | 15% | Any existing marks in your category? |
| Emotional fit | 10% | Does it feel right for the product? |
Score each candidate 1–5 per criterion, multiply by weight, and total. A name scoring below 3.5 overall is a liability, no matter how much you like it.
The pronounceability test is brutal but necessary. Say the name to three people who know nothing about your startup. Ask them to spell it. If more than one person gets it wrong, the name has a spelling problem. If people hesitate before saying it, it has a pronunciation problem.
What to Do When the .com Is Taken
The .com is still the default. Users type it automatically. Investors expect it. But a taken .com doesn't have to kill a great name.
The options, in order of preference:
- Buy the domain. If it's listed for under $2,000 and the name is genuinely strong, pay it. That's a rounding error compared to the cost of rebranding later.
- Use a prefix or suffix.
getluma.com,tryluma.com,lumaapp.com. This works but adds friction. Users forget the prefix. - Use an alternate TLD.
.co,.io,.so,.aiif relevant. The risk is that users type.comout of habit and land on someone else's site. - Change the name. If the
.comis taken and the owner wants $50,000, the name isn't available. Move on.
Never build a brand on a domain you don't control. If the .com belongs to someone else, you're building equity in their asset. Every user who types the wrong TLD goes to them, not you.
A Realistic Search Workflow
Here's a sequence that produces better results than random registrar searches:
Define your concept cluster. Write down 10–15 words that capture your product's essence. Not synonyms. Concepts. If you're building a project management tool, your cluster might include: flow, build, stack, team, ship, track, plan, move, clear, light.
Translate the cluster. Run each word through Google Translate into Japanese, Korean, Spanish, Italian, Swedish, and Turkish. Collect the short results. Check
.comavailability for each.Generate fragments. Take each cluster word and truncate it in both directions.
flow→flo,ow.build→bui,ild. Combine fragments with vowels. Check availability.Run the scoring table. Filter candidates through the five criteria. Kill anything below 3.5.
Check trademark databases. Run the top three candidates through USPTO and EUIPO searches. This takes an hour and prevents a lawsuit.
Test with real people. Say the finalists to five strangers. Ask them to spell them. Watch their faces.
The whole process takes a weekend. That's less time than most founders spend agonizing over a name they'll regret in six months.
The 5-Letter Reality Check
Here's the honest truth about 5 letter startup names: most of the good ones are gone. Stripe, Slack, Figma, Notion, Loom, Canva, Airtable. All taken. All built into valuable brands.
But "most" isn't "all." New five-letter names enter the pool every year through:
- Foreign language words that English speakers haven't discovered
- Fragments that no one has combined yet
- Made-up words that happen to sound good
The supply isn't infinite, but it's not zero either. The founders who find good short brand names are the ones who stop searching the same tired sources and start doing structured, creative work.
The name you want isn't sitting on a registrar's "premium domains" page. It's hiding in a language you don't speak, a fragment you haven't combined, or a compound you haven't considered. Go look there.
FAQ
How short should a startup name be?
Aim for 4–7 letters and 1–3 syllables. Shorter is better for memorability, but don't sacrifice pronounceability for character count. A six-letter name that's easy to say beats a four-letter name that isn't. The real test is whether someone can say it after reading it once and spell it after hearing it once.
Are 5 letter startup names still available?
Yes, but not in the obvious places. English dictionary words of five letters are effectively gone from the .com namespace. What remains available are foreign language words, word fragments, and made-up combinations that happen to work. The hit rate is lower than it was ten years ago, but structured searching still turns up usable options.
Is a .com domain still necessary?
It's strongly preferred but not strictly required. If your product is developer-focused, .io, .dev, or .so can work. If it's AI-focused, .ai is increasingly accepted. But if your audience is mainstream consumers, the .com matters more. The risk with alternate TLDs is that users type .com out of habit and land on someone else's site.
How much should I pay for a taken domain?
For a strong name that fits your brand, $1,000–$3,000 is reasonable. Above $5,000, you should question whether the name is worth it or whether you're rationalizing a sunk cost. Above $10,000, you're almost always better off choosing a different name. The domain is one input to your brand, not the brand itself.
What makes a short name memorable?
Memorability comes from three things: distinctiveness, pronounceability, and emotional resonance. A name like Loom is memorable because it's a real word with a clear image attached. A name like Figma is memorable because it's distinctive without being hard to say. If your name sounds like a dozen other startups in your category, it won't stick no matter how short it is.