The Screening Sequence Before You Spend Any Money (And Why a Free Domain Proves Nothing)
There's a specific moment familiar to almost every founder: you think of a name, you check the domain, and it's free. Relief floods in. It feels like the search is over — you found it, nobody else wanted it, time to register everything and move on. That feeling is one of the most common and most expensive mistakes in early naming, because a free domain proves almost nothing about whether a name is actually safe to use in business.
What “the domain was free” actually tells you
A domain registrar checks exactly one thing: whether that specific string, on that specific extension, is currently registered with them. That's it. The registrar has no connection to any trademark register, any company registry, or any court — it is simply a database of taken strings, and an available domain only means nobody has claimed that string at that registrar yet. It says nothing about whether a business already operates under that name elsewhere, nothing about whether the name is trademarked in your industry, and nothing about whether using it would create a legal conflict the moment you launch. Treating “the domain was available” as evidence of anything beyond “the domain was available” is exactly the trap this article is about.
The same is true, for the same reason, of an unclaimed social media handle and a company-register search that comes back empty. Each of these checks a genuinely different, narrow system — a platform's username database, a state or national business registry — and none of those systems talks to a trademark office before handing out what you asked for. Three green checkmarks from three unrelated systems can sit right next to a real, active trademark conflict that none of them was ever built to detect.
The screening sequence, in the order that actually protects you
The fix isn't to distrust every tool — it's to understand what each step in a proper screening sequence actually does, and to run them in an order that spends the least money and time on the steps that turn out to matter least. Roughly, in order: run the mechanical self-checks on a candidate name; generate the domain and handle patterns you'd need to check; search plainly for existing use of the name in your market; check the real domains and handles with the actual registrar and platforms; run an actual trademark search; and, for anything you're seriously committing to, get a trademark attorney's opinion before you spend real money. The full reasoning behind each step — including exactly what it does and does not prove — is laid out in the name screening reference, which runs this exact sequence against several worked examples.
A worked example: two candidates, two very different pictures
Running the name-availability checklist against “Cedarly” — a candidate from this site's own naming-generation guide — returns a clean pass on all seven mechanical checks: comfortable length, safe characters, one easy word, about three syllables, a short seven-character domain slug, no generic-word match, and no repeated-letter runs. Running the same checklist against the word “Business” returns six passes and one fail — it's flagged as a plain generic term, hard to stand out with or protect distinctively. That contrast is genuinely useful information, and it's exactly the kind of self-inflicted problem this checklist exists to catch early, before you've spent any time or money on a name that was always going to struggle.
But look closely at what that contrast is actually telling you. A full pass for “Cedarly” means the string itself has none of the seven mechanical problems the checklist looks for. It does not mean the name is trademark-free, and it does not mean nobody else is using it. The domain name generator produces seven plausible-looking domain ideas for “Cedarly” — the bare name across four extensions, a get- and try- prefixed version, and an -hq suffix — every one of them flagged “valid” because every one of them is a technically well-formed domain label. “Valid” here is a statement about DNS syntax rules, not a statement that any of these domains are actually unregistered; no registrar was queried to produce that list.
Handles are the same trap wearing a different outfit
Everything said above about domains applies just as directly to social media handles, and it's worth calling out separately because the reasoning feels different even though it's identical. An unclaimed handle on a given platform means exactly one thing: nobody has registered that specific username on that specific platform. It is checked against that platform's own username database and nothing else — not a trademark register, not other platforms, not a company registry. A consistent, unclaimed handle across three or four platforms that matter to you is a genuinely nice practical outcome worth securing early, simply because usernames do get taken quickly once a business starts to get attention. But it belongs firmly in the same category as a free domain: real, useful, and completely silent on the legal question.
Why the legal question resists shortcuts
It's worth being explicit about why none of the checks above can stand in for a trademark search, because the reasoning isn't obvious if you haven't run into it before. Trademark rights, in many jurisdictions, can exist through genuine use in commerce even without any formal registration — so a clean search of a trademark register doesn't always tell the complete story either. Rights are specific to a class of goods or services, which means the identical name can be safely available in one industry and actively protected in another. And whether two names are “confusingly similar” enough to conflict is a legal judgement, weighed against how the marks look, sound, and are used, and against a body of case law — not something a checklist, a generator, or a plain search can determine. Rules and registers also differ by country: a name that's genuinely clear in one jurisdiction can be blocked in another, so operating across borders means repeating this whole sequence per jurisdiction that matters to you, not assuming a single clearance covers everywhere.
Money, in the order you're actually likely to spend it
The reason this sequence is framed around money, not just risk, is that founders rarely spend it all at once — it leaks out step by step, usually in an order that front-loads the cheap, exciting parts and defers the boring, protective ones. Registering a domain costs very little. Filing a DBA or forming an entity costs more, but is still a fairly contained, fixable expense. Ordering signage, packaging, and a first print run of business cards costs real money and is much more annoying to redo. Running paid marketing under a name, building an audience around it, and hiring people who introduce themselves with it every day is the most expensive to unwind of all, in money and in momentum. Doing the trademark search and getting a professional opinion before any of that spending happens, rather than after the signage is already ordered, is the entire point of putting these steps in order — not because the checklist steps don't matter, but because they're cheap enough to run early and thorough enough to catch the easy mistakes before the expensive ones are locked in.
What this sequence buys you, honestly
None of this is legal advice, and nothing in this sequence, run in full, constitutes a legal clearance either — that step still requires a trademark attorney reviewing your specific name, your specific classes of goods or services, and your specific jurisdictions. What the sequence does buy you is proportion: the cheap, fast, mechanical checks catch the obvious self-inflicted problems for free, in minutes, so that the expensive professional step — the one that actually determines whether a name is safe to use — only has to happen once, on a candidate that's already survived everything cheaper than it.
A free domain feels like an answer. It's really just one small, narrow fact in a much longer sequence. Treat it that way, work through the rest of the sequence in order, and you'll spend your money on verification before you spend it on anything you'd have to walk back.
Scaling the sequence to what's actually at stake
Not every name needs every step run at full intensity before you can move forward with reasonable confidence. A hobby project or a very small, local, low-visibility business might reasonably do the mechanical self-check, generate its domain and handle shortlist, and run a plain market search, then proceed while keeping an eye out for problems. A business you intend to scale, take outside investment for, or build real brand equity around deserves the full sequence run in earnest, ending in an actual trademark search and a professional opinion, before any meaningful money goes toward signage, packaging, or paid marketing. The sequence doesn't change based on the size of the business — only how far down it you can responsibly stop before spending money you'd hate to have wasted. When in doubt, run one more step than feels strictly necessary; it's almost always cheaper than the alternative. Treat every step in this sequence as a filter that gets more expensive and more authoritative as you move down it, and spend accordingly.