Loading summary
Transcript67 lines
- [00:01]
A
I have some really exciting news. We just released our second annual AI Marketing Industry Report and it's free. You'll be surprised by how often marketers are using AI and the tools that they're using. This is a comprehensive study of more than 730 marketers, and it covers how marketers are applying AI to their work, the benefits of AI, the big concerns marketers have, and a whole lot more. Even though I'm the author of this research study, I can say with full confidence that this is the most comprehensive study of AI adoption among marketers that I have ever seen. Get your free copy now by visiting social mediaexaminer.com AI Report 25 this might be just what you need to get your boss or your clients moving along with those AI initiatives you want to start. Get it now@social mediaexaminer.com airport25 hey, it's Michael Stelzner here. After running Social Media marketing world for 12 years, I've noticed something fascinating about the marketers who attend. They usually fall into two camps. Camp number one is those who show up because they have to. Their results are declining. They're scrambling to catch up. And camp number two is those who show up because they want to. They're already achieving a level of success, but they know that change is coming and they want to go so much further. Both learn valuable insights, but their outcomes are completely different. As Emily Ray Shutti said, I came away with a million ideas and new creative ways to approach my work. My business has since doubled in size and revenue. The revolution is here. Instagram algorithms keep shifting, Facebook's ad costs keep rising, and the list goes on and on and on. What type of marketer will you be in 2026? Join thousands of marketers at Social Media Marketing World 2026 this April in Anaheim, California. Save big when you register today at Social MediaMarketingWorld.info.
- [02:02]
B
Welcome to the AI Explored podcast, helping you put AI to work. And now, here's your host, Michael Stelzner.
- [02:14]
A
Hello, hello, hello. Thank you so much for joining me for the AI Explored podcast brought to you by Social Media Examiner. I'm your host, Michael Stelzner, and this is the podcast for marketers, creators, and business owners who want to know how to put AI to work. Custom GPTs don't just save you time, they can actually make you money. People say that custom GPTs deliver generic results, but is that really true? In today's episode of the AI Explored podcast, we'll explore how to create custom GPTs that free your time and increase your sales. My special guest is an AI strategist who helps seasoned coaches, creators and service based businesses embrace AI to increase their revenue and free their time. She's known as the Custom GPT Gal. Her group coaching is the AI for Income accelerator. Wendy Breakstone, welcome to the show.
- [03:06]
B
Well, thanks for having me. I'm really excited to be here and have this conversation and go deep in custom GPTs today.
- [03:11]
A
I'm very excited as well. So let's start with where in the world did you get into AI?
- [03:16]
B
Oh my gosh. Well, I think I probably got into it it right around the same time that everybody else did, which was in November of 2022 when ChatGPT launched. And you know, at that moment, AI kind of became mainstream and it was very esoteric before that. So a lot of people jumped in right away and by the next day of ChatGPT coming out, they had courses and they were prompt engineers and they knew everything. And I took a little more time with it. I learned it, I explored it. But By April of 2023, I had started building co custom GPTs, but on third party platforms. I had realized that there was an opportunity for a program that I already had going, which was about audio summits, to help my clients get traction and get results faster. If I used this third party tool that clamped into ChatGPT and made these at the time, AI assistants. And so it started from there and really it took me a little, you know, it took several months for me to figure out where my place in AI is, because AI is. And when somebody says, oh, I'm an AI strategist, it's like, what does that mean exactly? Because there's so many paths and roads and it's like saying, I'm in marketing. What does that mean? And so I really wanted to make sure that I found my place and a place that I love and that I'm passionate about, that I can explore. And that really helps my clients, who are typically course creators and coaches and service providers, help them get traction better. Because like, at the end of the day I always feel like simple and uncomplicated wins. And you know, my background is in marketing project management for high tech companies and you know, for all different types of shapes and sizes of companies. And I've always gone back to simple and uncomplicated wins. And this really felt like the part of AI that matched my, like, mission and my goal in life as an online coach. Love it.
- [05:18]
A
Okay, so what are some of the biggest misconceptions that you see when it comes to people actually creating custom GPTs, because I, I would imagine that there's a lot of people that have tried it out maybe early and just said this thing doesn't work. I don't know what, what are some of the things you're seeing?
- [05:35]
B
You know, I laugh because there's like a gazillion misconceptions, but let's just hit three really quick just to kind of keep this focus. So I think in the big picture, a lot of people hear custom GPTs and they AI to technology. And if people that feel fear of technology are going to naturally feel fear of any type of AI or building something that sounds techie. And so the first misconception that I hear a lot is I don't have that type of background or experience. I can't build custom GPTs, I'm not techie enough. And so really that comes down to the fact that you can use AI expertly in your business without being an AI expert, without having to teach AI, without feeling like you need to be the guru of AI. And so that's really what I try to impart on people, is that building custom GPTs actually isn't techie at all. It's actually about logic and strategy. And so where people say I'm not techie enough, I say let's think about the logic and make it step by step and simple and then go from there. So that's misconception number one, is it's just not that techy. Honestly, the second misconception and myth that I've heard a lot over the past couple years is that like, if I give my clients AI, are they going to need me? Are they going to need my services? And to that I always say, actually if you give them AI and we're going to talk deep about specifically custom GPTs, when you give them these tools, it draws them closer to you because you're giving them a deeper level of access24.7 that they didn't have before. And so instead of thinking they're not going to need me anymore, I always say they're actually going to want you more. And so that's really kind of like a mindset shift that I think a lot of people make as they start to dive into this world of transitioning their offers and some of the work that they do internally into AI powered work. The third one that I hear quite a bit, and this is where I think things get really, really interesting, is I always get generic responses from AI. So how are my custom GPTs going to be any differently. I don't want to release something that's going to make me look bad. And you know, that's interesting because first of all, whatever you put in, the level that you put into it, and the background of the custom GPT really dictates what you get out of it. And as we have this conversation today, I think people are going to come to understand that the GPT is like a second extension of you and your offer and your program and the work that you do. And so as you start to get better at putting in good stuff, you get good stuff out of it. And so like when I'm always thinking about proprietary methodologies and whatnot and how you can put those in, it really makes the outputs so much better. And that's not just with custom GPTs, honestly, that's with any AI that you're using.
- [08:40]
A
Awesome. Okay, so we're going to get deep today into creating custom GPTs. But before we go there, for those that are listening who maybe have not created really incredible custom GPT, what are the unlocks, what are the upside? What is the benefit if they apply what they're about to learn today?
- [09:01]
B
Yeah, well, you know, there's two sides of this. The first side is make more money and the second side is save more time. Right? Those are the two big things that you can always make more money, but you can't get time back. Right. Like time is constantly moving forward. And so when you have a tool like a custom GPT that can actually do both of those things for you, that's kind of like where you want to invest, right? That's where you want to maybe invest a little bit of money or invest a little bit of time to save time and money back in your wallet. Right? And so I always think about like the benefits of building these. Number one is it's going to like double your enrollment, your retention, your renewals and your referrals. Right? So the more people, if you are a coach, if you are a service based business, even if you, you know, work inside of a company, the better results you can get. Other people and those around you and your stakeholders, they are going to refer you. You're going to be the person that pops up in their head that says like, hey, so and so they delivered that fast or they got me the results fast. People don't care about the airplane, they care about the destination they're going to. Right. And so that's kind of like what the custom GPT is. But they also allow you to expand your offering very quickly. So you can spin up new income streams very quickly. You can expand your one to one work. If you are a service based provider or a consultant or you're, you're doing done for you work in some capacity, it's going to allow you to do more in less time. Which also means besides the fact that you're saving time, you can potentially take on more clients which will help you make more money. So it checks both of those boxes really, really well. And then finally, and something that I value a lot is it helps you stay competitive and relevant. And relevancy right now is a big deal. And if you want to stand out as a leader in your space, and I'm not talking about, you don't need to be like the guru of all gurus, but standing out as the leader, showing that you're innovative, showing that you're not the last to put custom GPTs or AI into your program, that says a lot about who you are and how you value other people's time as well. So those I think are some of the really like key benefits of the make more, more money, save more time balance.
- [11:22]
A
I love this and I know that a lot of people are resonating to everything that you just said and this next question is not something that I had planned to ask you, but I feel compelled to ask. Custom GPTs are built on ChatGPT and as of this recording, Google Gemini, to the best of my knowledge, does not allow you to create these shareable custom gems, which is their equivalent to custom GPT. And CLAUDE projects outside of a teams account also do not allow you to create these shareable projects like a custom GPT. But I can imagine a future where everything we're talking about eventually will apply to any of these large LLM models. Today it's all chatgpt, tomorrow could be Google or it could be Claude. Do you anticipate that likely being the case?
- [12:14]
B
Absolutely. And you know, Claude has this way now that you can build these kind of like apps, they call them artifacts. Yeah, artifacts. And I'm not a big cloud user because, you know, I started in chat GPT and like all my stuff is over there, so that's where I'm staying. But yes, all of these large language models, these LLMs are most likely and Claude is, you know, kind of picking up the pace on allowing people to create these apps. But personally I haven't gotten one to work awesomely over there yet. But it's coming. I spoke to somebody yesterday who has one, but there's also third party Platforms that I recommend and that I use frequently that allow you to build these types of AI assistants that it's still very easy to do. It's still not a quote, techie thing to do, but allows you a little bit more flexibility than even building on ChatGPT. Because when you start to think about these like, and I'm not talking about vibe coding here, I'm talking about specific platforms where you can build like a GPT that you can embed it right into a website or right into a course page or right into the back end of your internal management system. And so it's built in and you've created almost like a walled garden for your clients, for your members and for other team members to be able to use. So I think it's the way to go. It's so amazing.
- [13:35]
A
Love that. And folks, I've given you an incentive right now to stick around to the end because I'm going to ask Wendy what those platforms are when we get to the end.
- [13:42]
B
Yeah. Yeah. Okay.
- [13:43]
A
So, okay, moving on. Let's assume we're going to create custom GPTs. Is there any kind of simple foundational concepts that we either need to expand on that you've already mentioned a little bit or that we need to set the ground rules for before we get started?
- [13:57]
B
Well, before I dive into, like the big nuts and bolts, I want to kind of just tell you about my 2 of 3 rule. And my 2 of 3 rule is when I start a new project and I'm kind of a techie nerd kind of person, so I could waste a whole day just like noodling on something that isn't going to serve me at all. Right. It's just going to serve my passion. But think about this. The two of three rule says, can you check at least two of these three boxes? Will it make you money or save you money? Will it save you a significant amount of time? And does it scratch a creative itch? And so if you can check off two of those three boxes, it's most likely something that you want to dive into. Now, a lot of the times we follow our what scratches our creative itch. And that doesn't always save us time. It often forces us to spend time and it often doesn't make us the money. But I want to say that if you start building custom GPTs and you follow a system, you will save time and make money and you'll probably get the bug and start feeling really creative. It'll also check the third box too.
- [15:03]
A
Okay, so I like this. What I'm hearing you say is, hey, when you're about to start a project like this, you ideally click two of the three boxes. Is it going to either make you money or save you money? And obviously saving money, depending on what your role is, could be also saving time, right? Because if you are, for example, an entrepreneur or creative and your time equals your money and all of a sudden you have extra time, then it all of a sudden, hypothetically, can save you money or open up new room for you to make more money. And then on the time side of things, the productivity thing, I'm with you. A lot of these things are designed to simplify your productivity. But I also like the idea of, like, allowing you to go deeper into your craft or your creative itch, because this is something a lot of people don't think about. Like, I am a writer in addition to being the founder and CEO of my business, and I have found that a lot of these things I've been doing allows me to go way deeper into my creative craft, and that ultimately does make my company money. Right? It might take me more time, but it's the right kind of investment of my time because it results in more money. So I love this model. Okay, so now we're onto the next part, which is where the heck do we begin when it comes to creating custom GPTs?
- [16:16]
B
This is where I think a lot of people get, like, stuck, right? Like, all of a sudden they're like, they have to put their head back in the sand because they're just like, oh, no. I just. I checked out. No way. So I've tried to make it as simple for people as possible. So I always think about building custom GPTs as three different types that you need in your business. And I'm going to go into these. But for the sake of this, there's foundational GPTs, add on GPTs and standalone GPTs. And so if you start by thinking about foundational GPTs, this is really where you start to think about, like, if you have a course or a program, or if you provide a specific offer, or if you have a specific role in a company, what is the one foundational thing that needs to be done before everything else? So the use case that I always think about is if you do have a course or if you do onboard new clients, there's a big amount of work. I call that the module one work. I always joke and call it the module one work, because a lot of the times that work that we have people do which is foundational to them getting the end result, that transformation that they paid for, that they hire us to do. Right. If you don't do that foundational work, you're just not going to get the same results. You're not going to get to the end. And that's the big problem with like the whole online course creation industry right now, is nobody's getting to the end. This is going to help them get to the end. And so I always say, like, if you build just one GPT and you always have to start somewhere, so you start with one, make it a foundational GPT and this is really the one that's going to help yourself do the work for your clients faster. Or it's going to help your clients and students get through that module, one work so that they can move on and start to get the results. And it's usually that heavy work that we want them to go deep and spend so much time that it's just going to burn them out. So much foundational GPT can solve this problem for them so easily. Easily. And I always say, if you build nothing else, build that one and I guarantee your clients are going to get results from you faster. If you're delivering, your boss is going to think that you're just like the biggest rock star in the world because you're always on point and your students in your courses are going to get to that finish line faster, which means again, they're going to upsell, they're going to renew and they're going to refer like crazy.
- [18:36]
A
So I know you have an example that we're going to dig into and we're going to dig into all three of these folks. We've got foundational add on and standalone. And what I heard you say is that the foundational stuff is the core work that you do at your company or for your clients. That is almost like the lever that moves everything else. Right. It's the key thing that you do because some people who are listening are like, I do a lot of key things. So how do they decide which of the things that they do should be where they should begin? Do you have any tips on, on that or maybe that will reveal itself through your example?
- [19:10]
B
Yeah. So some of the tips I can give is that like, if you go back to some sort of form or template or swipe file for the same thing over and over and over again, chances are there's something very foundational in that work. Either it's you're going back and looking, you know, from a marketing perspective at your niche, at your ideal Client at your core, messaging, right. Or if you're some other type of coach or you're doing other types of work, it's that deep work if you're in the health related industry that could be going back and reviewing like the needs of that particular client, what do they want to eat, what do they don't want to eat, et cetera, et cetera. Right. And so the foundational work is really what drives everything else, right. If you work in a company and you're in a marketing department in a company, if you just onboard it, right. And they don't have like a way for you to get all of that foundational information, the messaging, the ideal client Personas, all of that stuff, you can't do your work properly. You're spinning your wheels down the road. And so I just love the idea of having these foundational GPTs. And I've seen many of my clients like launch a foundational GPT either for themselves to you in like an agency model or for their clients to use in a course or something like that. And right away it's, I mean I have a client who saved $5,000 just by building herself two foundational GPTs because it took over the work for one of her full time employees. Right. And it's so fast. And so I just feel like the foundational work that everybody wants to like slide over and glide over and put the blinders on for is really where we need to start. And the good news is the custom GPT make it really fast.
- [20:56]
A
So let's talk about an example here. I know you have one.
- [20:58]
B
Yeah, let's do that. So I have a client, her name is Brenna and Brenna is a pre launch specialist. So she's a marketing expert, a copywriter. I think you probably know her, Michael.
- [21:09]
A
Yeah. Can we say her full name? Just because she has been on my other show.
- [21:12]
B
Yeah, her name is Brenna McGowan and she has a program called the pre launch plan program. And her goal, she's a copywriter, she's this launch strategist. And what happens is, is that she believes just with all of her heart that if you nail your pre launch, the stuff you do before the doors even open to your offer, you are going to sell out. And so she has a very robust program that she's created over the years for this. But the problem is, is that it's a lot of work for her clients and she knew that. So last year Brenna came to me and she's like, how can I make this faster? And I said, well, what do you have them do in the first couple of weeks? And so what happened is, long story short is we ended up building two foundational GPTs, her particular program. And sometimes you'll realize that you need one, sometimes you need two. It doesn't really matter how many you need, but for sure you need one. Everybody needs at least one foundational GPT. So for Brenna, for instance, we built a foundational GPT called her Outstanding Offer Overview. And this helps her clients flesh out what their offer is and what the buyer beliefs need to be for people to buy that offer. They can't sell it if they don't know what their offer is. They can't sell it if they don't know deep, simply what their buyer beliefs are. And then secondarily, we built another foundational GPT because Brenna believes that her clients need to be talking with their ideal clients and with their existing clients. And so she has people do interviews. So we built a GPT that synthesizes the transcripts of the interviews. You basically drop in the transcript, and then what it does is it helps pull out all of the pains and desires and goals and misconceptions that those clients has in the real world voice the words that the client is using. So then what you can do is now they have the information, right? Now they know what their clients are thinking and feeling and want, and they know how to pair that with these buyer beliefs, right? And so foundationally, they have what they need to move on to the next section. I see a lot of people that come to me, a lot of the people that come to me are in kind of like the marketing and messaging, messaging and copywriting. And what I find fascinating is like right now I'm working with two private clients. They both need niche and ideal client GPTs. But each of these GPTs and every single one that I see built is radically different. Because when built right, they are built on your foundations and frameworks and methodologies and your experience and become that extension of your teaching. So it's very important that you make sure when you're building these that you put in all of that really deep information that you've spent time creating so that your clients get those results with these foundational GPTs.
- [24:04]
A
Have you ever attended a conference and came back with all sorts of ideas and you were fired up, but you struggled to get everyone else on your team to be on the same page? Well, here's what I noticed after a dozen years of social media marketing world, when one person attends, they get Great ideas. But when a team attends together, they create something much bigger. Think about it. One person learns about AI and marketing. A team builds out an AI strategy. One person discovers Instagram tactics. A team redesigns their entire content approach. Noah Stanley told us, quote, I will bring multiple members of my staff next year. The mindset, openness and amount of time the experts gave to us was head and shoulders above other conferences, unquote. When your team attends together, you can cross pollinate ideas in real time, divide and conquer to cover more ground and return with a unified action plan, not just a bunch of scattered notes that you never get around to. Plus, you've got a built in support system when you try new strategies and you're back at the office. The company's seeing the biggest results from Social Media Marketing world. They're the ones investing in team transformation. Bring your team to Anaheim this April and multiply your impact. Learn more at social media marketing world.info folks. This is going to be an unlock for some of you. Yes, Brenna created GPTs of her unique frameworks and methodologies and locked them behind a paywall for her customers. Right? And the benefit here is now she does not need to like answer a bazillion questions and like, her customers get instant access to these insights. And it's a benefit that is available as long as they remain a customer. Right. And for those of you that work inside a business like this allows maybe your staff, if you're like a director, marketing or whatever, to now all of a sudden, instead of ping you with a lot of questions, to have a really cool model that you have created that is simply available to them. And some people are going to ask, yeah, but isn't this all public? And well, why don't you explain there's different levels of sharing GPTs, right?
- [26:12]
B
Yeah, for sure there is. So when you build a custom GPT Inside of ChatGPT, you've got a couple of different options. Number one is you can make it public to everybody everywhere. You can make it unique for just you. So the only me option. And then you can offer it to people who have the link. And once you they have the link, technically they can share the link, right? And so that does, you know, kind of give people a little bit of pause. But I always also say, well, you know, it's kind of like creating Google Docs, right? Like people can share your Google Docs if they wanted to as well. That said, this is where we start thinking about those third party tools where you can really lock them behind A paywall. Now, there is one thing that you just mentioned that I do want to quickly hit on because I think it's really important to get, especially for those people that are in corporate or that have teams and that want to build custom GPTs for their teams. And what I recommend is that you have a ChatGPT team account and that all of your people are in that team account because then you as the owner of it, have visibility and can see who's using your GPTs and for what they're using them for. And so you have that level of control when you have a team account, but that's really an internal thing. Now. I still build all my custom GPTs for the most part in ChatGPT because I like that experience for the end user the best. There's pros and cons to using ChatGPT versus using the third party platform that we're going to talk about at the very end if people stick around.
- [27:47]
A
Okay, so we hit the foundational thing. Let's move to the add on thing.
- [27:51]
B
Yeah, cool. So add on GPTs are kind of like the. Okay, what now? So a lot of people come to me and they say, I want to build a. I'll give you an example. I want to build a social media calendar. I want to help my clients write better social media or I want to write social media. And I say to them, okay, great, what are they going to put in? And then I get like the blank stare because they haven't done the foundational GPT. So an add on GPT takes the information from your foundational GPT and uses it to create a different set of information for the end user. So for instance, with Brenna, what we did is once her clients go through her two foundational GPTs, they now have a ton of really valuable foundational information about their messaging, about what their clients needs are, et cetera, et cetera. They take that information and they go and just drop it right into her social media email calendar creator.
- [28:54]
A
So we're talking copy and paste, presumably.
- [28:56]
B
We're talking copy and paste. I usually recommend that people like save the outputs into a workbook or some sort of a worksheet. Some people do, some people don't. But you know, if your ChatGPT sidebar looks like mine, it's a little bit of a hot mess sometimes. And honestly I save my outputs and I go back to my outputs and I use them over and over again in add on GPTs. But when you build an add on GPT, it knows that it's going to get information that you've created from another GPT. So they're tightly connected, they're like in tandem with each other. You can't use an add on if you haven't done the foundational work.
- [29:32]
A
Yeah, and by the way, these words that Wendy is using is a framework that Wendy designed. It's not necessarily like you're not going to find something called the add on or foundational GPT inside of custom GPTs. But conceptually the why behind this, which we really haven't addressed that much, is, and I'm going to hypothesize here, that we have some really useful information that has been produced with a custom GPT and we want to go deeper. And rather than just making the custom GPT able to do 10,000 things, it's better to create these little add on like sub GPTs almost for lack of better words, that are designed to do very specific things. That's coming from the other one. Is that the idea? Rather than have having one that does 10 things, have one that does the core things and then have a bunch of little ones. Is that kind of what we're really talking about here?
- [30:20]
B
Yeah, that's exactly it. Like I fundamentally believe that you don't want to like throw everything in the kitchen sink into one GPT. A lot of people do that and it's okay. But typically, you know, and the platforms continue to learn and get smarter and smarter so they can do more and more. But what I find is that, so it's the more it spits out, the more the end user also needs to take in. Often we're creating too much outputs for the end user to be able to use and to synthesize. It's creating so much information, it's too much information. When you have a step by step by step process and approach to any type of system, it's going to be easier for the end user. The end user could be you. You could be building GPTs for your own use in your role inside of a market department or inside of some sort of, you know, any type of a department as an office manager. You could be building custom GPTs for yourself to use after you onboard a new done for you client. Right. You're a foundational GPT for that could be something that takes in their onboarding form and then create some sort of a foundational report for you that then you can use to do the work that you're going to do for them. Ongoing and ongoing and ongoing. So, so it's not just about like how much can the GPT do? It's how much can the end user actually consume in one sitting. And oftentimes I find that if you're not careful, we get excited, we get a little carried away, and then we do want to shove too much in and it's just too overwhelming for the person that needs to use it.
- [31:56]
A
Okay, the third type of custom GPT that we're about to discuss here is the standalone. Why don't you explain what that is and why that is different than the others?
- [32:06]
B
Yeah, so a standalone GPT actually does not require or and it's not connected to a foundational GPT. So I'll give you an example. We built a couple of standalones for Brenna and standalone GPTs are great for free. Lead magnets, low ticket offers, exercises in your curriculum or exercises that you do that aren't necessarily connected to that foundational work, but that require you to do them over and over again. Now I always like to caveat standalone GPTs because sometimes people get super excited and they build a standalone GPT that ends up cannibalizing something in their program or their offer that is core to it, that is foundational. So you don't want your standalones to be your foundational GPTs, but they do have a very powerful place in the ecosystem of your work that you do on repeat and of the work that you ask your clients or students to do. So I'll give you the example back to Brenna here. So we built a standalone calendar calculator for her. She wanted a new low ticket offer and a lead magnet that she could use when she does speaking engagements and events. And so it's not connected to her foundational GPTs, it's not connected to other parts. The people in her program can use this calendar creator, but she can also use it for a subset of people who may be interested and wondering and learning about her, her pre launch plan program, but then aren't ready to take the dive. So let me give you another example that I recently.
- [33:41]
A
Wait, wait, just real quick. Did you call it a calendar creator or a calculator? Because I want to make sure I understood what you were saying.
- [33:46]
B
So we built her a calendar creator which is an add on, and then we built her a calculator which is a standalone.
- [33:53]
A
So explain what that calculator thing does just so people understand what that means, because I'm curious about that.
- [33:58]
B
Yeah, so in this case, the calculator took like their launch date and open cart, closed cart, and it calculated backwards the five weeks that they would need and then mapped out based on some other like high level details that we put into it what they need to do based on when they want to open their cart and close their cart.
- [34:21]
A
So was this for her for Brenna's business or was it for her clients?
- [34:24]
B
She uses it as a low ticket offer and a lead magnet.
- [34:27]
A
Okay, keep going with your other examples.
- [34:29]
B
Okay, so here's another kind of really cool example. I have a client, she helps myofascial release therapists grow their businesses. So we built a foundational, we built an add on and we built a third one which is a rate calculator, which is a standalone. And this rate calculator they fill out like five things like how much do you want to make? How much are you spending, how many weeks per year do you want to work? How many clients per week do you want to see? And then it gave them the basis, it calculated the basis of what their package rate should be and a single therapy session should be. So it's completely standalone. People didn't need to do the foundational work, but they still absolutely need to do the rate calculator. So it's a really powerful tool inside of her program that helps her clients like see what they need to see in order to make the money they want to make, but it's not connected to that foundational GPT.
- [35:23]
A
Okay, so I think I'm tracking with you. I think what I'm hearing you say is that this standalone GPT is used by your clients as a way either to generate leads, like put your name on this thing and we will give you this cool custom GPT that hopefully moves them down the sequence to ultimately wanting to check out their other programs. Or it is a self liquidating offer where they sell it for, for a very small dollar amount. Right. And then they get access to this GPT which probably is recommending their other programs. Is that the kind of stuff we're talking about here when we say this kind of stuff?
- [36:03]
B
That's exactly, exactly right. And also standalone GPTs also have the place inside of internal work that you do and also inside of courses and programs and you know, any type of done for you, consulting that you also do. Don't discount them just because they stand alone. They're still extreme powerful tools that you'll have. And I have standalone GPTs that I use myself, that I've built just for me or just for clients that I use on repeat. I have one called Hooks Genius. And you give it a little bit of information and it writes a Bunch of your hooks for you. So I use that one on repeat all the time whenever I want to refresh or get some new like hooks ideas for an offer that I'm working on.
- [36:45]
A
Okay, cool. So we've set in motion three different kinds of GPTs. You can create foundational ones which are core to the work that you do. We've got add ons which are ancillary. If this was a visual diagram, they'd be the ones that jut out from the foundational one and those little circles around it. Right. And then we've got these little satellite ones that we call standalones which can be just kind of in their own little independent world, but serve an important purpose. So once we've identified some of these kinds of GPTs we want to create for our business, what's next?
- [37:16]
B
Yeah, so the first thing that you always want to do is, and oftentimes in all honesty, you kind of hit the nail on the head right there. As you said, once we identify, sometimes that is the hardest part of the whole thing is getting out of your own way and identifying a few that you want to start building. And a lot of the times, you know, like we get caught up in our brain and we think about the 50 that we want to build or where they go the best. And the beautiful thing about them is you can update them. They're living, they're breathing, basically you can update them on repeat. It's not a one and done done thing, although once you nail it, you get to use it on repeat. And that's one of the things that I love about them is that you kind of can build it once you fine tune it, you tweak it, you make sure that it's doing what you want it to do and then you get to use it on repeat, which is where that simplicity and uncomplication in your business, like helps everything. Right? Helps you make more money and saves more time. That said, spending a little bit of time on the front end to figure out what is that first GPT if I build nothing else. And that is often the hardest part for people to clear their mind and think, what do I want to build first? What is going to make the most impact for myself as the user or for my end users to build first. And so that's first and foremost is you got to identify, kind of got to audit your own work a little bit sometimes or your course sometimes it's hard to do because you don't want to change what you've spent, spent, you know, the past 10 years building or the past, you know, year building. And it just requires a little bit of like, finessing to be able to incorporate that GPT and let the GPT do the work for you. I find that sometimes people are still asking the GPT to do 20% of the work when it honestly could a lot of the times do the 80% and then you just follow up and do the 20%. So it's a little bit of a mindset shift that people need to get the their arms around if they haven't started to go down this path yet.
- [39:15]
A
Okay. So it seems to me if I don't have a GPT and I want to create a GPT, what I'm hearing you say is focus on the big problem that you feel like ChatGPT can help you solve here and try to make it so that it can do 80% of the work.
- [39:33]
B
That's what I think.
- [39:34]
A
I would imagine we're trying to get somebody who uses this to have some sort of a wish, right? Are we looking for some sort of like a wow response from the people that use it if it's not us? I mean, if so, tell me a little bit more about that.
- [39:47]
B
So you said something early on in this conversation that I wanted to circle back to. But first of all, I will say that, yes, the idea is, is that you want your clients to show up and be moving forward. And so what you had said is if you're hearing the same questions or answering the same questions over and over again, that's a great starting point. Right? And so as you're evaluating your program, your offer, the work that you do, right? The place where you're maybe even your boss has said, like, can you make that, can you do that quicker? Why does it take you a week to do that? It's like, because people think at a very slow pace compared to AI. That's just the reality of it. And so if we can think about those places where we're hearing those same questions over and over, or you have calls, or your boss is saying to you, like, get it to me faster, right? That's where you need to build your first GPT. That's the place.
- [40:41]
A
And if you work for me, unfortunately, you hear that a lot.
- [40:45]
B
You need GPTs.
- [40:46]
A
Sorry, staff. Okay, so let's talk about some tips on actually building out the GPTs. I'm sure you have a lot.
- [40:52]
B
I do have a lot. But I'll just go quickly through what I call the stars framework. And the stars are if you think about, like when you Sit down to build, the first thing you really need to think about is that strategy. What do you want to solve? What do you want to accomplish? And then take that deep look into your own work, your own offers, and again, where people are getting stuck. The best place, if you can clear a road and get people out of their frustration, whether it's your students and clients or boss or whatever, Michael, get him out of his frustration, then you're winning. And then yes, people are going to be like, wow, how'd you do that so fast? Wow, I would have never thought of saying it that way. Wow, I would have never thought about doing XYZ instead of abc. Right. That's what the GPT is going to do for you. I always find that inside of the responses is the answer. Right. It may not be the exact way that you wanted to say it. And also it's our responsibility to, you know, if it can get you that 80% of the way there, it's your responsibility to take it another, you know, 10 to 20%. Like you want to get to that good 90%. Right. It should get you to 80%, get to 90, 95%, you're golden. Right? So you think about strategy. This is back to that STARS framework. So the S is strategy. And then you move into the training docs. And this training docs is really where you can start to go back and think about like if I'm getting generic responses, my training docs, the training that I gave it, the information I gave the GPT isn't good enough and it's the builder's fault, it's not the GPT's fault usually. So training docs super important in GPTs, they're called knowledge docs. So always think about like what type of knowledge and examples do I want to give my GPT? And then you move into the A, which is actions. And actions are the things that you tell the GPT to do and you want to be specific and you want to talk to the GPT like it's a five year old and be that kind of pedantic with it and say like first you're going to do this, then you're going to do that. Like if you were teaching a 5 year old how to bake cookies, you wouldn't just say you need flour and chocolate chips and some butter, you would be very, very specific about it. And that's what you need to do. Because a recipe is part ingredients and part directions. Right? And that's what these GPTs need as well. So if you think about the training docs or the knowledge docs being some of the ingredients. The actions are what they need to do, right? How to do it, what's the temperature it needs to be set at, et cetera, et cetera. And then you move into the responses and this is where you're starting to look at how is the GPT responding? What is it giving back? Do you like the output? Is the output not great? Is it inconsistent? Do you want it in an outline? Is it using too many emojis? Like all those things that GPT is known for, those are the types of things that you can control in the actions to make sure that your responses are really dang good. And then finally we get to the final S and stars and it's structure. So you start with strategy and you end with structure. Okay? These are the bookends of every good custom GPT and structure is how you follow along. And so all of your GPTs have a very specific structure because if they're all willy nilly and kind of here and there and you just kind of of flow with it and yes, chatgpt to build it for you, it's going to be really hard to figure out where to fine tune it and where to do the tweaking. So that's the STARS framework and if you kind of follow that and you think about those things as you move through, you're going to be golden.
- [44:22]
A
Okay folks, the thing that some of you have been waiting for, which is the third party platforms for custom GPTs, which is not part of our script. So earlier and multiple times throughout the interview, Wendy mentioned that there are tools that you can, and for lack of better words, hide your custom GPTs behind so you can create the equivalent of apps and bots and all that stuff. So we would love to hear your preferred or favorite tools that you use for this kind of stuff.
- [44:50]
B
Yeah, so, and again, this is like building a GPT in another program and then being able to embed it into a website or a lesson or whatever. Right? Put it behind some sort of paywall or email wall. So the first one I use is called Rad Genius. R A D Genius. I think it's RadGenius.dev and I love this one because it's very simple to use, it's very, very user friendly and it's very, very affordable. And I love the people that own it. I know the people that own it, I work closely with them and I kind of recommend it to everybody. So that's Rad Genius. The second one that I real Quick.
- [45:28]
A
What do you love about it? What is it? What does it do that you love?
- [45:31]
B
Well, it's uncomplicated. And again, if you go back to this, you want to use AI expertly but not have to be an AI genius. That's the tool you want to use because it's very similar to building in ChatGPT and it doesn't have a ton of bells and whistles. They're adding a few more features here and there that are really powerful, but they're very specific about how they're adding the features. And also, just as a caveat, it only works with the OpenAI models. So the ChatGPT models, 4, 041 those models, perfect. So that's the first one, and it's super affordable. It's $7 a month, so there's no. Nobody can say to that. Right. It's super good. The second one that I like a lot, that I actually started building with before Chat GPT even launched. Custom GPTs is called Pickaxe. And Pickaxe, I just got an email right before this podcast, as a matter of fact, that they just raised another round of funding and they've added in more features and Pickax, a very robust platform, and they've made some significant changes over the past year. And some of the changes have made it a little more complicated than I think it needs to be. And so for a lot of my clients who want to be able to have that walled garden, have control over their GPTs, create powerful AI assistants or they work in a corporation, they want that control. They want to embed it right into their, like, intranet or whatever, right. They're going to to choose something that's like a rad genius. Some people that are a little more advanced are going to choose something like a Pickaxe. So those are the two that I have used and that I recommend. There is a third real quick.
- [47:11]
A
It's Pick P I C K A X E co Just in case anybody wants to check it out. And what's the third one?
- [47:17]
B
Pickaxe. It's super fun. Both of these platforms are not going anywhere. The third one that I've heard a lot about, and it's a little more spendy, is called Formwise. And so a lot of people are. Have been testing this one. I haven't tested it. I had a trial account. It's great, but it's in a different. It's in a little bit of a different price point, and I don't think it's necessary right now.
- [47:41]
A
Yeah. And that's formwise AI. Okay, well, Wendy, this has been extremely insightful. And now people, I'm confident, have a lot of ideas of how they could use GPTs, custom GPTs, in a way they've never done it before, but I know we've simply scratched the surface of that mind of yours. So if people want to connect with you on the socials, is there a preferred platform? And if they want to work with you, where do you want to send them?
- [48:09]
B
Yeah. So typically I spend most of my time in my free community that's over on school and it's called AI Powered Business with Wendy. But if people want to kind of go deeper and learn more, I have a free master class and they can sign up for it over@wendybreakstone.com and that way you can sign up and you can join me live for my next live masterclass. I host them once or twice a month and it's gonna help you go deeper into this conversation that we've had here today and learn more about, like, is this for me? Is it time? I would say it's probably time for most everybody to get going on it and that's where they can hear about me. Thank you.
- [48:48]
A
Thank you so much, Wendy, for sharing your insights with us today. I'm very excited. I've got all sorts of ideas running through my brain now.
- [48:54]
B
Thank you for having me.
- [48:56]
A
Hey, if you missed anything, we took all the notes for you over@social mediaexaminer.com A75. Be sure to follow this show on your favorite podcasting app. And if you've been a listener for a little while, I would love a review on whatever app you're listening to this on. And also do let your friends know about the show. You can find me on Facebook, LinkedIn and or Apple. And do check out our other shows, the Social Media Marketing Podcast and the Social Media Marketing Talk show. This brings us to the end of the AI Explored podcast. I am your host, Michael Stelzner. I'll be back with you next week. I hope you make the best out of your day and may AI help you become more successful.
- [49:33]
B
The AI Explored Podcast is a production of Social Media Examiner.
- [49:41]
A
This is the year to finally come to Social Media Marketing World 2026. Grab your tickets right now by visiting socialmediamarketingworld.in FOX.