AI Naming Strategies: Prompt Like a Pro for Better Brands

In this episode, Mike Carr shows how to become "AI native" in name development by using strategic, iterative prompting to create more relevant brand names. He shares a fun cat toy example and draws on insights from experts like Shelly Palmer, who suggests using a personal JSON file to guide AI. Key tips include clear direction, competitor and SWOT analysis, global awareness, long-term brand alignment, and data privacy.

Chapters

• 0:00 The Power of a Great Tagline
• 0:11 What Makes a Name ‘AI-Native’?
• 0:35 Strategy Before Prompts
• 1:15 The Trap of Lazy Prompting
• 2:20 Thought Leaders to Follow
• 2:55 Naming a Cat Toy with AI
• 4:15 The Golden Rule of AI Naming

Transcript

Mike Carr (00:11):

Last episode, we talked about the importance of slogans and taglines and how they can actually be your fountain of youth, your brands, and your names Fountain of Youth. So this week we're going to talk about something a little different. AI native name development. And everything we're talking about these days is ai because there are so many ways AI can improve the creative process, and there's so many ways it can degrade or ruin the creative process. So last week I talked about Andy Cress and Orbit Media's seven stage process for AI prompting. And if you miss that, you really need to go back and listen to that episode and follow Andy's seven stage process because pretty amazing and pretty cool. And I don't care how good you think you are at ai, I think you're going to learn a few things. Well this week, what do we talk about?

(00:59):

What do I mean by AI native, right? And that isn't about a strategy incorporated into ai, not just developing some cool, fun, exciting name that you think is pretty neat and that your team thinks pretty neat. But actually going back and looking at your strategy, looking at what's going to engage and resonate with your target and your audience, what is your long-term vision for your name and your brand? Building some of that into your prompting is going to definitely improve the output of what you get. Two other folks though that you really ought to think about are Shelly Palmer with Palmer Group. He has a daily blog posts he does about what's going on in ai. If you haven't followed him, I definitely would recommend him. And Andrew Ning, his last name is N the G. He used to teach at Stanford. He now has a company called Deep Learning ai, deep learning.ai.

(01:49):

And I'm going to start today with an example. He put out on creating a name. This is a name for a cat toy. And one of the points he makes, which I think is so important, is you have to guide the model. You have to guide the flavor of AI you're using, whether it's chat GPT or Google Advanced Gemini, or Gemini Advanced or rock or llama, or any of the other flavors that are out there. You have to sort of guide them to think through the answer. So he has a very simple three prompt example for this cat toy. So the first prompt, which is sort of interesting, pretty obvious though, is what are some fun, joyful words that relate to cats? Just words that are cat relevant but are sort of fun. So he came up with things like pur and whisker in feline, in pounce, in Meow.

(02:35):

Now, none of those necessarily would be great names for a toy, maybe pounce would be, but they're all pretty generic and be hard to really own and protect and build a brand around that. But the next prompt is sort of cool, and he just came up with this and lots of other ways to sort of generate the same kind of more ownable name. But his idea was rhyming names. So come up with names that rhyme for a toy based on those five words. So chatt PT came back with things like per twirl or whisker whisper or feline beeline or pounce Pounce. But my favorite is Meow. Wow, Meow. Wow. I think that's a great name for a toy, right? Yeah. Wow. Anyway, then the last prompt was add a fun relevant emoji next to each toy name. So it actually came up with some visuals.

(03:28):

So here you have three very simple prompts using chatt PT where you're refining each step and you're coming up with better names that sort of fit the playfulness and the whimsical of what you're after. And this is sort of what Andrew advocates in a lot of his courses. And this particular course is free. You can go to coursera.org, C-O-U-R-S-E-R a.org and just search for Andrew Ning. His last name is nng. And you'll find of course, generative AI for everyone. It's free. I highly recommend it. He talks about how be clear and specific in your prompting, and then as you start, you look at the result and it's probably not going to be ideal. So what he's advocating is don't spend a lot of time on the initial prompt. Go ahead and just try something, see what it comes back to you with, and then think about what's wrong with it, right?

(04:19):

Where'd you screw up? And then you refine it, you iterate. And as you go through this process multiple times you'll get better and better and pretty quickly you'll understand the kind of prompting that's going to work for the kind of name you're after or the kind of slogan you're after. Just be a little bit careful with confidential information. A lot of these LLMs will sort of gobble up your information and use them to learn, so you never know where that information may end up. And make sure you can trust the output, right? There's still some hallucination going on. It's gotten a lot better I think over the last few months, but every now and then it'll come back with something that's just bizarre. Now, Shelly Palmer, I mentioned with the Palmer Group, he said there's one file that if you're not using this file, it is key to leveling up your AI game.

(05:01):

So the title of this blog post, which came out yesterday, and which I highly recommend you also take a look at the one file that will level up your generative AI game and how to write it. And so basically what it is is you're telling AI a little bit about yourself, your style, how you think and work. And then you can provide this file at the beginning of any session with AI or have it remember it since a lot of the AI flavors now have built-in memory, if you're paying the 20 bucks a month or whatever it is, so that before you start a session, it'll sort of mimic the way you think, the way you communicate and the results will be much more in line with something that's true to you in your authenticity. So you can think about this as an individual about a company, whatever it might be.

(05:46):

It's called a json JSON, and that stands for JavaScript Object Notation. Now, you don't have to write JavaScript. He gives you the actual JavaScript in this blog post, and then he shows you where you can fill in different things. So it is just something that even though you don't necessarily know any code you can still use to sort of set the stage for every one of your AI sessions. Some of the things that I thought were interesting in this JavaScript that you can think about is what's your name? What's your title? What's your role in your organization? And then a short description about your particular expertise. Where are you coming from? Are you an academic master's, PhD? What's your background? And then what voice do you want conveyed? And then what is your tone that you'd like to sort of be reflected in communications, perhaps in the name you want something that's very professional and very concise.

(06:38):

Or in some of the naming, we might substitute a brand archetype here. You want to be the hero. I want to be the magician. I want to be the guy that saves or the gal that saves the day. I want to be a super human kind of person. So whatever that persona is. And then what is your voice? You want to be direct. You want to be fact-driven, you want to be news or you want to be playful, you want to be fun, you want to be whimsical. Do you want to have a sense of humor? Is it optional? Is it something you want to see reflected in the name itself, in your messaging, in your slogan, in your tagline? And what is that self sense of humor? Is it rise? Is it self-deprecating or something else? And then what to avoid. I think this is very interesting.

(07:17):

So when you build this JSON about who you are and a little bit about yourself, the way you think, and you want to be communicate, what do you want to avoid? Do you want to avoid cliches? You just want to avoid filler, too much? Verbose speech or speculative language or just and dashes and things that slow down the road? Then who is the audience you're trying to appeal to? This is fascinating to me. Who do you want your communications to work for? Maybe you work with C-level execs, sophisticated audience, right? Senior managers of a company. Maybe it's Fortune 500, or maybe it's maybe it's a small company or maybe it's some other group. When you think about naming your brand managers, insights managers, folks that are in innovation, they're developing new products, mid-level directors that have a branding, marketing, creative, maybe a marcom background. So what is that audience?

(08:07):

What industries are they in? Do you specialize in a particular industry? In our case name, stormers. We're industry agnostic. We span all industries, which makes it so much fun, right? We get to work on a healthcare project one day, financial banking project, another way, an AI naming project the next day. It's just constantly new, exciting things with very sharp people in each of those industries. But anyway, you specify your industries in your interest, your own interest. Well, I'm interested in AI strategy. I'm interested in anything that's innovative, anything that's breakthrough, anything that's disruptive. I'm interested in growth. And then when it comes to your communications, what format I'd like nice tight paragraphs. I personally love things that are scannable. I have very little patience as my team knows for things that are too verbose or require a long read unless it's quiet and I really need to do some deep thinking.

(08:54):

So I like lots of bullet points. I like lots of infographics. And so you can even specify visuals. I like pictures. I want to limit pictures. I like infographics. I want to limit them. Any other ways to make your writing more scannable, easier to read, a more enjoyable read. And so the more clarity you provide, the more value AI will return. See, again, just like Andrew Shelly says, don't overthink it. Start prompting AI the right way. Build your JSON and see what happens. One of his tips, which I thought was interesting, is you can use AI to write your JSO to write the JavaScript code for you, whether it's chat, GPT or Claude Gemini. And then when it comes to naming, I want to leave you with these things because I think naming's a little bit unique. Some other things to think about that we always try to do for our clients.

(09:44):

And I'm not sure AI can do as good a job as perhaps humans in some of this, but it's getting better and better. Give it the competitors, right? Have it go out and scrape the web, the key competitors, what names are they using? Try to do a SWOT analysis if it can. Looking at the positioning, looking at the words on their website. Give me a SWOT analysis. What is each competitor's strengths, weaknesses, opportunities, threats. Where's the white space for us? Maybe feed it your brand strategy or your client's brand strategy and branding and naming guidelines that it needs to follow in coming up with names. And lastly, but perhaps most importantly is what's the vision for whatever it is you're naming? If it's a product, if it's a company, where are you going to go? Where's your client going to go? So the name's not too limiting.

(10:29):

So it has the robustness and the flexibility to span, not just what you're doing today and maybe what you're going to be doing in the future too. So those are some of the tips I wanted to leave you with today on using ai, becoming an AI native, using it the right way to really be much more effective in developing names and augmenting your team's capabilities, and then bringing in the experts that you guys have on your team or with your crew to really move the ball forward. And I'd love to hear from you guys. If you have questions, comments, if you'd just like to chat, please email me. It's mike@namesstormers.com, all one word with an S on end of it. Or just leave me a comment on this podcast or on this YouTube video, and I'll be sure and get back to you. Thank you so much. Have a great week.