You already have a name. You picked it fast, somewhere between the idea and the first order, and for a while it did the job. Now there is traction, a shelf crowded with competitors who sound a lot like you, and a low hum of doubt that the name is starting to cost you something. Or maybe the opposite is true. You have no name yet, a launch date on the calendar, and a browser open to a free generator that just handed you two hundred options you do not trust.
Either way, the choice usually gets framed as a money question. What a name should cost, how much to spend. That framing is the one that leads founders wrong, the reason so many overpay for help they do not need or skip the help they cannot afford to skip.
Fit matters more than price. There are three real ways to get a brand name, each built for a different problem, and the right one depends on where your brand is standing right now, not on what you can spend.
What you are actually choosing between
Strip away the price tags and the three options line up on one scale. That scale is judgment, not cost. It measures how much human thinking comes attached to the name, and how much risk somebody takes off your plate before handing it over.
At one end is a free AI generator. It gives you volume and little else. Type in a few words and it returns a long list of invented, brand-style names in seconds, then leaves the decisions that matter on your desk. It does not tell you whether the name is defensible, or whether it is even legally yours to use.
At the other end is a naming agency. It takes all of that off your plate. It researches the category, screens the linguistics, runs a trademark prescreen through counsel, and puts its name behind a recommendation. The name is the last and smallest thing it sells you. What you are really buying is the judgment and the months of work behind it.
In between sits a third option that almost no one will tell you about.
The free AI generator
Start with the tools at the bottom of the market, because for a lot of founders they are the right answer.
Namelix is the one most people land on first. Give it a business idea and a few keywords, pick a style and slide a dial for how literal or how random you want the results, and it generates short invented names built around your input. It is free and genuinely useful, and it markets itself on generating more original ideas than ChatGPT. Looka, Shopify, and a dozen others run the same play.
What these tools share is a hard edge to what they check. They confirm a domain is available. They do not confirm the name is yours to use. Shopify states on its own page that the names it produces are examples only and may already be used by other businesses, and it tells you to do your own due diligence or talk to a lawyer. NameSnack is blunter, saying outright that while it runs a domain search, it does not check against trademark databases, and that you should do that yourself or hire a lawyer if you intend to trademark the name. None of them claims a strategist read your shortlist, and none weighs how a name sounds against the rest of your category. They are not built to.
That is the trade. A generator hands you raw volume instantly and hands every judgment back to you. What it cannot tell you is whether a name will clear a trademark search, or hold up on a shelf full of competitors who sound like you. That judgment stays with you.
A generator hands back raw volume and leaves every judgment about it to you.
For a pre-revenue founder still circling the idea, the trade is fine and often smart. The cost of a name you later change is low, the money is better spent elsewhere, and a generator is a fast way to react to a hundred options and learn what you are drawn to. The danger is only in mistaking the list for a decision.
The naming agency
At the top of the market the model flips entirely. You are no longer buying a list. You are buying a firm's judgment and the apparatus behind it.
Lexicon is the firm most often named here, the shop behind Febreze, Sonos, Swiffer, and the Impossible Burger. Its own process page describes a multi-stage engagement run by a worldwide network of linguists, with structural linguistics, trademark evaluation through attorneys, and consumer research folded in, and it states plainly that a typical engagement takes eight to ten weeks. The firm has said it put more than $2 million and five years into its own sound-symbolism research.
That apparatus has a price. A single Lexicon naming project runs roughly $40,000 to $250,000. Step down to the working mid-market and a serious agency name still lands around $15,000 to $75,000 over six to ten weeks, going by the published tiers at firms like Tungsten and Tanj. Even at the transparent end, Eat My Words posts a menu that starts near $6,000 for a stripped-down package and runs to $49,000 for full-service work.
What the money buys is everything the generator skips. The agency does the category research, screens the names for how they sound and what they connote, and prescreens for trademark conflicts before it ever shows you a candidate. Igor, another well-known shop, lays its process out the same way, positioning, then competitive analysis, then name development, then a worldwide trademark prescreen before any name reaches you.
For the right brand it earns the price. The fit is a company creating a category, or running a rebrand where the name has to carry years of weight and a board will scrutinize every option. The misfit is the early DTC founder. For someone still proving the business, eight to ten weeks is a long hold on a launch, and a $40,000 line item assumes a marketing budget that is not in hand yet.
The naming service in the middle
Between a free list with no judgment and a $40,000 engagement with months of it, there is a wide gap, and it is where a lot of funded DTC founders find themselves. Search for naming help and the results mostly skip past them, the agency tier pitching budgets they cannot justify and the free generators handing over a list and walking off.
The middle option fills the gap between a free list and a six-figure engagement.
A naming-decision service fills that gap by splitting the difference in how it works. AI generates wide, the way a generator does, so you are not paying a person by the hour to brainstorm. Then a human strategist runs a kill pass over the output, cutting against the brief and against the names already on your shelf, weighing how each one actually sounds. What survives is a short list a person stands behind rather than a raw dump. Then evidence goes to work on the finalists, a trademark screen and, at the top tier, a panel of real buyers. You walk away with a verdict and the reasoning behind it, not three hundred options and a shrug.
A cheap version of this already exists, the email-only outfits that will sell you a short list of names for under $250 with no trademark work and no buyer input. That clears the floor. The harder slot to fill pairs the human judgment with real evidence, and that is the lane BrandNames occupies, in three one-time tiers, each worked by a person and each including a strategist call.
The entry tier, Sprint ($999), hands back a hand-narrowed short list, a kill pass on what got cut and why, a conviction memo on the top pick, and a positioning summary, inside five business days. Clearance ($1,299) adds a founder-grade knockout search across trademark, adjacent classes, domains, and social handles, with a preliminary risk summary on the finalists. That search reduces risk and points you at the names worth a lawyer's time, and it is not a legal opinion, so an attorney is still required for that. Validation ($1,499) puts the finalists in front of a panel of thirty to fifty buyers screened to your category, adds a visual placement check and a language and culture screen, and delivers the full strategy deck.
None of this is an agency, and it does not pretend to be. There is no eight-week discovery phase and no five-figure invoice. What it borrows from the agency is the part that matters to a founder making a call they have to defend, a person reading the names and evidence standing behind the pick, at the price of a logo rather than a rebrand.
The three side by side
Read down the column that matches where you are, not the one with the smallest number at the top.
| AI generator | Naming service | Naming agency | |
|---|---|---|---|
| Typical price | Free | $999 to $1,499 | $15,000 to $250,000 |
| Time to a decision | Seconds | Days | 6 to 10 weeks |
| Who makes the call | You, alone | A strategist, with you | The firm |
| Trademark work | None, domain only | Knockout search, attorney still required | Full prescreen through counsel |
| What you walk away with | A long list | A verdict and the reasoning | A researched recommendation |
| Best fit | Pre-revenue, exploring | Funded, defending a crowded shelf | Category creation, enterprise budget |
The middle column is the one the search results hide, the option for the founder who has outgrown the free list but cannot justify the firm.
Which one fits your stage
Match the option to where you are, not to the price you would prefer to pay.
You are pre-revenue, or close to it, and still testing whether the idea has legs. Use a free generator. The cost of a name you swap later is low, the money is better spent on the product, and a generator is a fast way to find what you are drawn to before anything is at stake. Paying a strategist now buys conviction you do not need yet. This is the one stage where the cheapest option is also the right one, and an honest guide should say so plainly.
You are creating a category, or naming something a board and a legal team will pick apart, and the budget and the calendar can absorb it. Hire an agency. The eight-to-ten-week process and the five-figure invoice are the cost of getting a foundation right when a rename later would cost far more, and at that altitude the research and the trademark rigor earn their price.
You have a real business, somewhere past the scramble of launch, and the name itself is now the problem. It is hard to defend, hard to trademark, or it vanishes on a shelf full of competitors who sound the same. This is the founder the middle was built for. You do not need another three hundred raw ideas, and you cannot wait two months or spend what a full agency charges. You need the choice de-risked, a person cutting the list against your real category, and evidence behind the name you carry forward. That is a naming-decision service, and for a funded DTC founder it is usually the fit.
One cost sits outside all three and lands on every one of them. Protecting a name is a separate bill. A federal trademark runs about $350 to $850 and a state registration $50 to $800, on top of whatever you paid to find the name. A fuller map of every price, from free to six figures, lives in the companion piece on what brand naming costs.
Frequently asked questions
Are AI brand name generators good enough? For producing volume to react to, yes, and they are free. For a decision you can stake a relaunch on, no. They check domain availability, not trademark or category fit. NameSnack says outright that it does not check trademark databases, and no generator claims a strategist has read your shortlist. The list is a starting point, not an answer.
What is a brand naming service, and how is it different from an agency? A naming service is a productized engagement. AI generates wide, a strategist cuts the list by hand and stands behind what survives, and a trademark screen and a buyer panel de-risk the pick. An agency is a bespoke multi-week engagement priced for an enterprise budget. The service hands you a verdict in days. The agency hands you a full research program over months.
Is the cost of naming the same as the cost of trademarking? No, and treating them as one number is a common mistake. Naming is what you pay to find the name. A federal trademark is a separate $350 to $850, and a state business-name registration $50 to $800. A generator checks neither, so that bill lands on you no matter which option you choose.
BrandNames is a naming-decision service for founders, built for the gap between a free generator and a six-figure agency, with a strategist on every project. See how the three tiers work.