Why Most Companies Get Brand Naming Wrong
Here’s the uncomfortable truth about brand naming: most companies don’t get it right because they’re naming for themselves, not their customers. Mike Carr’s latest insights on “Naming in an AI Age” cut through the naming noise and give you the real deal on what actually works.
After working with hundreds of companies on their brand naming strategy, the patterns are pretty clear. The businesses that nail their naming follow specific rules, while the ones that struggle make the same predictable mistakes over and over.
Key Points:
- Name for your customers, not yourself.
- Know what the name must do before using AI.
- Use AI for screening, but test and clear legally.
- Manage team feedback positively and sell the final name.
The Foundation: It’s Not About You
Stop Making It Personal
The biggest mistake in brand naming strategy? Thinking the name is for you. It’s not.
Your opinion matters way less than you think it does. The name is for your customers, your employees, your investors, your partners—basically everyone except the people sitting in the conference room debating it. This mindset shift alone will save you months of internal arguments and second-guessing.
We’ve seen CEOs fall in love with names that make perfect sense to them but completely miss the mark with their actual audience. Don’t be that CEO.
Ask One Simple Question
Instead of drowning in a checklist of naming requirements (short, memorable, easy to spell, conveys benefits, fits the brand, etc), ask yourself one question: What do you want the name to do?
One CEO gave us the best answer we’ve ever heard: “I want the name to sell itself.” Now that’s a tall order, but it gave us crystal clear direction. No committee-speak, no wishy-washy requirements—just one clear job for the name to do.
Getting AI Right (And Wrong)
Don’t Rush to the Robots
Yes, AI can help with brand naming strategy, but here’s where people get it wrong: they use it too early. Before you start prompting ChatGPT or Claude, you need to do the hard thinking first.
Who’s your real audience? What should the name actually accomplish? How will it need to evolve as you grow? Where will this name show up—trade shows, white papers, your website, business cards? Think through all of this before you touch AI.
When you skip the strategy and jump straight to AI, you’ll probably get a name that sounds pretty good right away. And you’ll stop there. Big mistake. That name might work today but fall apart when you try to expand or when you see it on a billboard.
Do your homework first, then give AI better prompts. You’ll get better names.
Where AI Actually Helps
AI isn’t useless for brand naming—you just need to know where it works. It’s surprisingly good at preliminary trademark screening, especially in the US. With the right prompts, it can flag obvious conflicts before you get too attached to a name.
It’s also decent at catching basic linguistic problems—does your name mean something embarrassing in Spanish? Is there slang you should know about? AI can catch some of this stuff, though it’s not perfect.
But here’s what AI can’t do: real name testing. There’s been talk about “synthetic respondents”—basically fake people created by AI to test your names. Don’t buy it. We haven’t seen it work, and frankly, it might never work properly.
Managing Your Team (And Their Opinions)
Kill the Negativity
When you’re presenting names to your team, you need to change how they think about the whole process. Tell them upfront: “You’re going to hate most of these names. That’s normal. Don’t worry about it.”
Instead of having them play critic, make them play advocate. If they had to argue for one or two of these names, which ones would they pick? Why might they work? This completely changes the energy in the room.
New names are fragile. They’re like babies with no context, no story, nothing wrapped around them yet.
Present Smart, Vote Smarter
When you present names, don’t let anyone react until they’ve seen everything. Show each name in context—on a building, on a business card, on your website. Let them see how the names come alive before they say a word.
Then have everyone vote independently and anonymously. No discussion first, no influencing each other. Just: “Here are your top 5, rank them.” You want their gut reaction before the politics kick in.
We use software for this, but paper works too. The key is getting that clean, unbiased first take from everyone before the meeting turns into a debate.
The Reality Check: Testing Names
Your Opinion Doesn’t Count
Remember how we said the name isn’t for you? This is where it really matters. If you’re a 55-year-old executive targeting Gen Z, your instincts about names are probably off the mark.
Real name testing isn’t about asking people which name they like best. It’s about watching how they actually react. There’s solid psychology behind this—Daniel Kahneman’s work on how we make quick decisions versus thoughtful ones.
The names that test well aren’t always the ones people say they prefer. Sometimes the name that gets the strongest reaction (even if it’s polarizing) ends up being the winner.
Don’t Skip the Professional Stuff
Legal Clearance Isn’t Optional
AI can give you a heads up about trademark issues, but it can’t replace a real trademark attorney. When you’ve narrowed down to your final choice, pay for proper legal clearance. It’s significantly cheaper than a lawsuit.
Same goes for linguistic clearance if you’re going international. You need actual people who live in those countries, not just AI translations. We use three linguists per country—people who grew up there, understand the culture, know what’s been in the news lately.
A name might be perfectly fine academically but sound like a recent scandal or joke to locals. AI won’t catch that kind of nuance.
Selling Your Final Choice
Make It Irresistible
When it’s time to present your final name choice, forget about being objective. This isn’t the time for “well, the data shows pros and cons.” You need to sell this name like your life depends on it.
Create mockups. Show it on buildings, websites, business cards. Write an amazing tagline (AI is actually great for this). Build the story around it. Make it sparkle.
Why? Because new names are weak. They don’t have years of associations and success stories built up like established brands. Amazon got laughed at when it launched. So did Google and Apple. Your name needs all the help it can get.
The Bottom Line
Good brand naming strategy isn’t about finding the perfect name—it’s about following a process that gives you the best shot at success. Know who you’re naming for, be clear about what you want the name to do, use AI wisely (not early), manage your team’s psychology, test with real people, get professional clearance, and then sell the hell out of your final choice.
Most companies skip half these steps and wonder why their naming process turns into a nightmare. Don’t be most companies.
Let us run with your project
Ready to go?
Together, we will get you across the finish line.
Transcription
Mike Carr (00:01):
Here we are again this week, Naming in an AI age. We’re going to talk about 10 naming hacks. First hack is, or the first question to think about is who is the name for this is the most important question, and guess what? It’s probably not you. And this is the mindset change that’s so important at the outset of any naming project is think about the target, the customer, the employee, the investor, the partner. That’s who the name’s for not you. Your opinion may not be quite as important as you think it is, but anyway, talk about who the name is for. It sets the stage for the right mindset throughout the process and getting everybody thinking the right way. Hack number two, what do you want the name to do? One question, hard to answer. It’s not about all these other things. Well, it’s got to be short and it’s got to be easy to say and spell, and it’s got to relate to these 10 things.
(00:59):
Convey these benefits and fit our brand, fit our positioning. Simplify all down to one question. What do you want the name to do? Just think about an answer to that question. I will tell you an answer that A CEO gave me once that I thought was one of the best answers I ever got. Not easy to do what he wanted it to do, but his answer was, I want the name to sell itself. That was it. Now, that was a heavy lift for us to do, but it was an interesting answer to that question. So what do you want the name to do? Hack number three, don’t use AI too soon in the process. There are some great ways to use AI when it comes to naming, but what we’ve seen more and more of our clients do is use as a crutch too soon before you’ve really thought through what’s the strategy?
(01:52):
Who is the name really for what do we really want the name to do? What’s the runway we need the name to have over time? What we might grow into, what you might expand this name into ancillary or adjacent areas. How will the name be used in different venues on a booth, at a trade show, in a white paper, on a website? Think about all those things before you start using AI to develop names. They’ll inform your prompts. You’ll have a better prompt hierarchy. You’ll have a better thought out prompt. Otherwise, AI might come up with a name right off the bat that you think is pretty darn good. You may just stop. But if you haven’t prompted with all those strategic questions in mind, you’re going to end up with something that really isn’t going to fit the bill for you down the road.
(02:45):
So think about those types of things and work through your strategy before you get AI involved. Hack number four, this is really important when it comes to your team and any stakeholders that are going to be involved in the process. Take the negative off the table. That is don’t have people think about a name as a critic, we specifically tell folks when we’re presenting names, wear the advocate hat. You’re not going to like most of the names we present, don’t worry about it. And the rest of the names you may not even be that excited about, but there might be a few that are interesting. We’ll put that advocate hat on. And if you had to advocate for one or two of these names, which ones would they be? And then talk about why they might work. And by taking that negative off the table and by having them wear that positive advocate hat, you get them thinking the right way about these new names that are fragile.
(03:36):
They’re like these little babies. They have no context, no story, no anything really wrapped around them. Hack number five, present names, but don’t allow any initial reactions. So when we present names to clients or team of clients or a board, what we ask them to do is say, you can ask us questions, but we don’t want you to react until you’ve heard and seen all the names. And this makes a huge difference selling the names right? We’re wrapping as much sparkle and as much polish around each name as we possibly can. We’re dropping the names onto the side of a building, we’re onto a business card, the website, so they can see how the names come alive. And we want them to fully think through each name and see the context and the story wrapped around it before they react to the name. And then once we presented all the names, here’s hack number six, have them all vote independently.
(04:33):
We use a software tool to do this, but you can do it on pencil and paper. Software’s a little bit cooler. You can do it real time. Everybody can get on their phone and sort of do it independently. The key here is it’s done anonymously and no one’s reacted to any of the names or no one’s has spoken to which names they like or don’t like yet, right? So everyone’s seen the names and they hopefully haven’t communicated with one another. And now you’re asking everyone independently, what are your top five or what are your top three or what are your top eight? Give us those names in ranked order. And so you’re getting to that independent read from everybody in the room before there’s any discussion or there’s any bias introduced in the conversation, which is hugely important. And then what we’ll do is we’ll display up on the screen the ranked ordering of those names and we wait each vote.
(05:19):
So first place vote, if there are five votes, might be five points. A fourth place vote might be four points. And so you’re getting names where maybe not everybody voted for it, but those that voted for it, it was their first or their second place kind of vote. And that’s a big deal. The seventh hack is use AI for preliminary vetting. AI can really do a pretty good job, especially if you’re in the US on initial trademark assessment. Now you have to prompt it the right way. And if you’re not sure how to do this, just give me a call or shoot me an email and I’ll be happy to get on the phone and sort of talk you through the process. But with right prompts, you can get a pretty darn good idea as to what trademark problems might you bump into, at least in the us.
(06:04):
And actually it’s getting a little bit better now. Some of the AI LLMs are getting a little bit better now in doing this outside the country, in the EU and other places like that. You can also use it for linguistic issues. Does the name mean anything offensive? Is there any slang? Meaning volgar, meaning it’s not quite as strong based upon our testing yet, but it’s interesting, it’s helpful. It’s not going to be as thorough as what you’re ultimately going to want to do, but it’s a great way to call out some names right off the bat. Now, name testing though is something that AI really can’t yet do. There’s been discussion about synthetic respondents. Do you know what a synthetic respondent is? What’s a synthetic respondent? It’s a fake person. Why would you use AI in synthetic respondents to test a name? Well, the school of thought is with enough data you can actually create personas that your typical target is let’s say a mom at home with a couple kids and she’s between 25 and 45 years old and she’s out there shopping for this family.
(07:06):
And so you can build these personas. And then the idea is with enough data, you can have AI mimic how they might react to a name, how they might evaluate a name. We haven’t seen that work yet. It may never work because the key to testing a name is not what name they say is their best name or their favorite name. It’s which name do they react to. And this gets into Daniel Kahneman’s book system Thinking Fast and Slow. Ray talks about system one and system two thinking it’s very academic, but there’s a lot of data and we are happy to talk to you. If you want to talk more about name testing that you can do yourself or platforms you can use or name testing we can help you with. Again, just give me a call or shoot me an email and be happy to talk to you about that.
(07:49):
But name testing is important. The reason it’s important, this goes back to the very first question, who are you targeting with the name? And it’s not probably you, which means your opinion on this name. As much as I hate to tell you this probably doesn’t matter a whole lot. Now, if you’re the CEO and you’re watching this, you’re going to say, oh my gosh, I’m the CEO of the company. Of course, my opinion matters. Well, if you can think like your target does great, but if you’re a 55-year-old white guy like I am or older like me, and you’re targeting a Genzer or a Gen Alpha or someone that’s a lot younger than you are, trust me, you’re probably not going to understand how they’re going to react to that name. So that’s why I think the testing can be such an important part. Hack number nine, final legal and linguistic clearance.
(08:38):
This is real important to do and it’s something you should have professionals do. Trademark attorney and IP attorney can do the final full searches. AI cannot do this. Do not rely upon AI to do this. It will not do as good a job as a trademark attorney. I’m beholden to any trademark attorneys, but we use ’em all the time and it’s an invaluable last step. And then also linguistic clearance. We use three linguists in country. They’re in each country we’re targeting that are fluent in that language. They’re grown up in that country that are exposed to the media. So even though the name may not be profane from an academic standpoint in Mandarin or in Hindi or in Farsi, but in that country, that name sounds like something that’s been in the press recently that would cause consternation or confusion or disconnect. You can only do that within country linguists.
(09:28):
You cannot do that with ai. I’m sorry. It just isn’t going to be as nuanced and understand sound and phonetics and confusion, like real boots on the ground folks. And then step number 10, when you pitch the name, once you’ve gone through all these other steps and when you’re pitching that name to the final decision makers or the naming committee or just two or three people in your company, you’ve got to sell it like your life depends on it. This is not a time to be objective and say, well, the data shows us this and it’s got these positives and these negatives. No, no, no. You have got to pitch it and you want to say everything you can. That’s great and exciting and challenging. You want to address the name up as I’ve already talked about and drop it on a side of a building with ai.
(10:17):
Put it on your website. You do mockups. Create an incredible tagline, which AI is very good at. Build the story around it and go through all that to really make a sparkle, to really make a shine. And the reason that’s so important is a new name is like an infant. It’s like a baby. It’s fragile, it’s delicate, and you can kill it so easily. It doesn’t have any of the story, any of the context, any of the muscle, any of the growth wrapped around it yet that really makes a name come alive. I mean, I can remember when Amazon came out and was laughed at. Google was a silly name. So was Apple. We’ve got some names on the table right now and Healthcraft and Vigo that with the right story can really come alive. You can go to our website and check those out if you want to sort of see what we’ve done with those names just as examples.
(11:02):
But without that, it is just going to fall flat. And so when you pitch that final name, and there are different ways to do this and there are different numbers of names to pitch and strategies on how many names you pitch and all that kind of stuff. And again, I can help you through that if you want, but just be sure and make it a sales pitch and give it your best. You get the buy-in you need, the excitement will follow. The passion will build what you’re after, right off the bat is an initial decision. So anyway, Ted Hex on naming and naming an ai AI age, when to use ai, when not to use ai. Have a great week.
