How to Come Up With a Business Name That's Actually Available

Here's a pattern that plays out hundreds of times a day: a founder spends two weeks brainstorming, lands on the perfect name, checks the domain — taken. Falls back on their second favourite — also taken. Third choice — available as a domain, but the trademark is registered. Fourth choice — everything's clear, but the Instagram handle is squatted by a dormant account from 2013.

Most business naming advice skips the availability problem entirely. It tells you to be memorable, be distinctive, pass the "radio test" — all true — but leaves you to figure out availability on your own. This guide does both. Here's how to generate a strong name and verify you can actually use it, in the same process.

Why Most Names Fail the Availability Test

The English dictionary has around 170,000 words. The vast majority of common, pronounceable combinations of 1–2 English words as .com domains are already registered. This isn't hyperbole — domain registrars have been cataloguing squatters and speculators since the 1990s, and the good single-word .coms sold for millions years ago.

This means the naming game in 2026 isn't really about finding a great word — it's about finding a great name that clears the full availability stack:

1. Domain available (ideally .com, .io, or .ai)
2. No conflicting trademark in your category
3. Social media handles claimable (or close enough)
4. No established competitor with a confusingly similar name

The most useful shift you can make is to treat availability as a filter applied during brainstorming, not after. Generate names with availability in mind from the start — not as an afterthought that kills your favourite ideas.

The 6 Naming Archetypes That Work in 2026

Not all naming approaches are equally likely to produce available names. Here's what works and why:

The common thread across all six: they avoid describing exactly what the product does. Descriptive names ("FastShipping.com") are almost always taken, often trademarked by someone else, and limit your ability to expand. Evocative names leave room to grow.

A Practical Brainstorming Process

Don't start with names. Start with words. Give yourself 30 minutes and generate as many words as possible across these categories — without filtering:

Aim for 50–100 words before you start combining. Then work through the archetypes above: which words mash well together? Which could be slightly misspelled and still read correctly? Which metaphors from other fields apply?

Filter your list down to 10–15 candidates before touching a domain checker. At this stage you're filtering on feel, not availability.

The Availability Stack: Check in This Order

Once you have 10–15 names you like, run each through this stack in order. Stop at the first failure — don't invest more time in a name that's blocked:

1. Domain availability Check .com first, then .io and .ai. A taken domain isn't always a blocker (you can negotiate), but a taken .com with an active business behind it usually is.
2. Google search the name Search for "[name] + your industry". Look for established competitors, news articles, or products with the same or very similar name. Confusion in search results is a long-term problem.
3. USPTO trademark search (US) / IP Australia / UKIPO Search the relevant database for exact matches and similar marks in your Nice class (product category). A trademark in a totally unrelated category is usually fine — a trademark in your exact category is a dealbreaker.
4. Social media handle check Check Instagram, X/Twitter, TikTok, and LinkedIn. A dormant account with 0 posts isn't a blocker — an active brand account is. Use a tool like Namecheckr to scan all platforms at once.
5. Business name registry search In your country, check the companies house / secretary of state database. A registered business name in another state or country isn't always a blocker, but in your own jurisdiction it can be.

The Trademark Trap Most Founders Walk Into

The most expensive naming mistake isn't picking a bad name — it's building a brand on a name someone else has trademarked and not finding out until you've printed 10,000 business cards or raised a round.

Trademark searches feel tedious, but the logic is simple: if someone holds a registered trademark for your name in your product category, they have the legal right to force you to rebrand. Rebranding an established product costs far more than a trademark search costs now.

The "Radio Test" and Why It Predicts Domain Success

The radio test is simple: say your name out loud to someone who hasn't seen it written down. Can they spell it correctly? Can they find it by typing it?

This test predicts more than memorability — it predicts whether you'll lose direct traffic to mistyped variations, whether journalists will link to the right URL, and whether word-of-mouth converts into actual visits. Names that fail the radio test also tend to have more domain squatting problems because there are more plausible spelling variations.

Where AI Actually Helps (and Where It Doesn't)

AI tools are genuinely useful for generating word variations, finding synonyms, and exploring naming archetypes at scale. Ask a tool like ChatGPT or Claude to generate 50 portmanteau names combining two word lists, and you'll get combinations a human brainstorm would miss.

Where AI tools fall short: they generate names without checking availability. You still get a list of suggestions that are mostly taken. The round-trip of generate → check → regenerate → check is where most of the time goes.

This is what Domain-ate solves differently. Instead of generating a list and leaving you to check availability manually, it checks availability in real time as it generates — so every name you see is confirmed registerable before it appears on screen. The AI also asks about your business, audience, and tone before suggesting anything, so names are relevant rather than generic.

When to Walk Away From a Name You Love

Sometimes you'll find a name you genuinely love that fails one of the availability checks. Here's a quick framework for deciding whether to fight for it or move on:

• Domain taken, no active website: Consider making an offer. Many parked domains sell for $500–$2,000. If the name is strong enough, it's worth it.
• Domain taken, active competitor: Move on. Building brand equity on a name that's already associated with someone else in your space is an uphill fight you don't need.
• Trademark in an unrelated category: Usually fine. A trademark for "Orbit" in chewing gum doesn't block you registering "Orbit" as a project management tool.
• Trademark in your exact category: Walk away. The legal risk isn't worth it at any stage.
• Social handle squatted, 0 posts: Register everything else first, then reach out to claim it. Instagram and X both have processes for reclaiming inactive accounts associated with a registered trademark.

The Bottom Line

The best business names aren't found — they're engineered. You start with a clear sense of what you want the name to feel like, generate widely across the archetypes that tend to produce available names, and filter through the availability stack before you get attached to anything.

The founders who get stuck are the ones who fall in love with a name before checking whether they can use it. Run availability in parallel with brainstorming, not after — and you'll spend far less time in the painful cycle of loving a name you can't have.

If you want to shortcut the whole process, Domain-ate was built exactly for this: describe your business, and get available names that have already passed the domain availability check — no manual searching required.