The Hidden Cost Inside Your AI Tool

[00:00:00] If AI is part of what you deliver, it's part of your business model. So decide on purpose. Welcome to the next evolution of the podcast, Entrepreneur School (In the AI Era). We are here to figure out how to integrate AI into the business you've worked so hard to build in a way that still feels like you. I'm your host, Kelly Sinclair, award-winning marketer turned AI platform co-founder.

[00:00:27] AI has fundamentally changed how I thought about work, systems, and what's actually possible. This show is about navigating that new reality together. It's going to be a wild ride, my friend, but I truly believe there's a way to leverage AI in a way that's intentional, human-centered, and aligned. It's an ongoing evolution, so let's explore, because AI may just be the unlock you

[00:00:54] need to achieve the life-first business you truly desire. Every time someone uses an AI tool that you built, somebody pays for it. Not you, maybe, not yet, but somebody. So if you are a creator, a coach, or a course builder, and you're thinking about putting AI inside what you sell, that one fact might change your business model. And most of us haven't done the math. I mean, I got a PR degree so

[00:01:22] that I wouldn't have to do math, but this week, my world flipped upside down. And I'm not going to give you the answers today, but I am going to give you the questions. And these are the ones that I was thinking about if I was building a new course, or a program, or thinking about how to transition that into this AI era. These are the questions that you need to ask around where AI fits into how you deliver

[00:01:49] that, because they might save you a lot of money. So let me give you a little backstory. Recently, we hosted a live AI product build lab, and that triggered a lot of questions from a really engaged audience, which I absolutely loved. And we made a lot of big business decisions at Wave this week

[00:02:13] about our pricing structure. I was building pricing calculators like a maniac for our website. And you can go there now and see exactly how AI use and Wave and everything fits in to your offers if you want to actually play with the tools that I build. And all of this came from me thinking through a big question about our company, which is what does it actually look like to be the tech engine behind

[00:02:41] someone's expertise? Because Wave is not just a software that you log into, we're actually part of how a creator delivers their work and makes their money, which means we need to know how that charge fits into how they make money. Because if those two things clash, then nobody wins. And I want to be really intimately aware of all the different business models and structures so I can make kind of the right recommendations. And that was kind of what was funny about the live event

[00:03:10] last week. Because somewhere in between me doing demos and teaching people how to use quad MCP connector to build tools and push them to Wave, I also took on a new role, which was business model strategist. And by the way, if you haven't yet registered for the AI product build lab, there is a coupon code in my show notes. Use the coupon code podcast one and you can get in for a dollar. You will get

[00:03:40] the last one sent to you right away and you'll be invited to come to our next one live, which is happening monthly at this point. So the conversations with our build lab participants all kind of kept going to the same place. They're talking about the kind of tool they build. They would ask, how do I price this? Can I just give people lifetime access to my thing forever?

[00:04:04] And we kept circling around something that I don't think has surfaced enough yet, but it's about to, especially with the cancellation of custom GPTs. And that is AI has a cost. Yeah, I know. So let's get the foundation down because everything else is building on this. Anytime a tool calls AI to do something, that is called an inference cost. So every message, every response, every automation,

[00:04:32] every time AI reads a file and answers back, it costs real money. And someone pays for it somewhere, no matter what platform you're using. And also this is the thing that is very unpredictable because the models themselves in the labs are increasing the cost per inference. And that's why your cloud usage bill changes every month and all of those kinds of things. And I won't get into all

[00:04:57] the details because it is quite complicated. And a lot of us are kind of missing this because especially if you've been building custom GPTs, you never saw that bill. Your user was using their own chat GPT subscription. So they were technically paying their $20 a month or whatever to open AI. And then your GPT was running on top of that. So the cost was passed on to them without anyone noticing.

[00:05:22] It was basically invisible. And also why custom GPTs were a great choice to build in the first place because they kind of, they felt free and you didn't have to worry about this cost of what would happen if people like used it heavily over time. But when you're building a tool that people access directly and that tool needs AI to run, that cost lands on the tool, which means it lands on you

[00:05:48] or the platform you're building or your pricing or whatever. So it's kind of like a utility meter. The tool runs, the meter ticks, and your clients aren't thinking about that meter because they don't really necessarily know what that means to you, but you have to keep your eye on what that looks like. So that is, that's the first thing to consider. It's like, okay, no matter what, you're paying for

[00:06:15] some subscription to some platform that has AI use credits or whatever component of it and your multiple software subscriptions every month are adding up to a certain cost. It is hard to predict. And also one of the main things that I wanted to try and solve with our pricing structure with Wave is to create a predictable pricing model for you so you could understand

[00:06:43] what it was going to cost your business to use our tool to power your expertise with our platform. So let's just start with a question. How long should someone have access? I have feelings. I always have feelings. So lifetime access was a thing. For years, it was a selling point.

[00:07:10] It's in the sales page bullets. It feels generous. It feels like value if you can give somebody lifetime access to your product. But let's be honest, nobody was coming back for a lifetime to use those things anyway. You know it. I know it. Most people using the things in the first few weeks or they never use it at all. There's been studies done on course completion rates. They are abysmal. And ultimately, lifetime access

[00:07:34] is more of a sales tactic than a requirement because it was giving people permission to say, I'll get to it later. And then that later never came. So now if you add AI to the picture, your tool runs on AI, lifetime access means a lifetime of meter ticking. You've taken one payment and you've signed up for unlimited costs. And that math gets ugly pretty fast. So the question you need to

[00:08:03] ask yourself is how long does someone actually need access to get the result you promise? Is it three months, six months, a year? What does that look like? And here's what I'm actually optimistic about. I think that this shift is good for our clients, for your clients too. Not just for margins. When there's a timeframe, there's urgency and that could help people implement, get them to show up in a

[00:08:28] different way. So they're stopping that collecting of all the things and they're starting to take action. Honestly, this was the reason I got excited about generative AI in the first place. I saw the opportunity for that strategy to turn into implementation. And I think that we are moving from a time of information to a time of implementation and implementation is what people are really paying

[00:08:53] for. So a window of access could actually support that. I think there's just messaging opportunities for you from a sales perspective. So sit with that question. What is the right window for your work? And this might be different for each of the offers that you have. So question number two, what about low ticket offers? How do I build AI into low ticket offers? So this came through our

[00:09:20] Build Lab conversations. Lots of creators have low ticket offers like a $27 workshop, a $47 template pack, a $97 mini course. And naturally thought is, well, if I include an AI bot in that, it's going to be better. Or maybe it was just that people like, it's sexier. People want AI as part of the things that they're buying now. We know that to be true, right? So great idea. Wonderful for your buyer. But the

[00:09:48] question is, does that price cover AI usage? If someone buys a $47 thing and uses it heavily for six months, what could that potentially cost you in inference? And is there anything left over for that to be profitable or even a break-even tool for you? I'm not saying that low ticket and AI can't work

[00:10:12] together. They can, but probably better if you also have other things that are potentially covering the costs here. If you're using a low ticket offer to help sell people into a higher ticket offer, then maybe it makes more sense. But if all you have are low ticket offers and you're just doing volume, it might not make sense depending on how you're building your AI tools. And it's really, it's a

[00:10:38] math problem. So when I build the calculator on our site, this is the exact thing I wanted to make visible. It shows you the percentage of your revenue that goes towards Waze. Once you see that number, it makes your decision a little bit easier. Sometimes you'll feel great about it and sometimes you'll realize that a higher ticket or a shorter access window might make more sense. Either way,

[00:11:03] you have visibility to make that choice and that's our goal. So since I'm asking you to look at your own model, I will tell you what we decided to do as far as pricing for Waze. Our pricing is a low monthly fee starting at $29 per month plus a flat fee per user. So per person who uses your tool. So we landed there because it ties our cost to your growth. So as you bring in clients, that fee will scale with you.

[00:11:32] As a creator, you get a predictable structure that you can build your pricing around and you know that you can just mark up your offers to fit this model if it works for you. So that definitely puts Waze into its own category and makes it so that we are not the right fit for every business model. If you give away unlimited access for a tiny one-time price, you'll want to rework that before you build probably on any platform and we would rather you know that going in. So that's what our

[00:12:02] calculator is for. You can plug in your prices to your different offers. You can see what your access window looks like. You can predict your number of clients and you can put all the pieces together on and see what percentage of your revenue will go to Waze and where are you comfortable there. I actually don't know of any other tool that does it that way and I think that's an advantage that we wanted to create is to really help you be able to have visibility into what those prices look like

[00:12:30] based on the inference costs because that's us thinking about how do we structure our business in a sustainable way so we can continue to be that tech engine powering your business and support your business. So here's a few more questions to leave you with as you're thinking about your business model and where AI fits into it. So what happens when someone's access window ends? Is there a renewal or

[00:12:57] a continuation tier or a community option? So maybe you have a higher ticket program and it goes for three or four months and then after that, could they just access your AI tools that are part of that on an ongoing basis and what would that have to look like to make that math work? And actually ask yourself which of your offers actually need AI inside of them and which might be fine without it? And then there's the

[00:13:23] wild card which is that you will have clients who will be heavy users and some that won't be and kind of trying to predict that do you have a tool that gives you the flexibility based on if you have some heavy users that are going to drive up your inference costs significantly. So that is where my head has been this week. It's been a week of big decisions, a lot of good questions from people in the lab and one

[00:13:52] big takeaway which is if AI is part of what you deliver, it's part of your business model. So decide on purpose. So if you want to come and explore whether Wave is the right fit for your business, then the AI product build lab is the best place to do that. You can join for just a dollar with the code podcast1. The link is in the show notes. We have a live session coming up on November the 2nd. The replay from our

[00:14:18] previous session will be delivered to you automatically if you sign up so you can have the best of both worlds. And if you want to see how Wave might work for you, you can jump over to our website graviastudio.com and check out our pricing calculator. Thanks for listening. I will catch you next week. Thank you for listening. If this episode got you thinking about how AI is impacting your business

[00:14:44] and what it might look like to integrate it in a way that actually fits, I'd love to help you think that through. You can book an AI strategy chat using the link in the show notes. It's a space to talk about where you are, what's shifting and how to move forward with intention. And if this episode was helpful, please share it with another business owner who's navigating this evolution too. And make sure you're subscribed so you don't miss what's coming next.