How Top Companies Are Using AI Agents to 2X Revenue (Without Replacing Anyone)

Speaker A: So I'm going to talk about AI LED gtm and where this is coming from is more along the lines of one of the most underutilized ignore ways to leverage AI specifically for go to market is accelerating and multiplying the team with assistance and agents versus replacing them. And kind of the question I have for everyone is like, who wants to be replaced? Is anybody raising their hand right now? No, you want to replace part of what you're doing to become superhuman, Right? And so we. One of the things I want you to think through is look at how much time, money and energy you've spent if you're a leader developing your team. And my question is like, why do we need to replace all of them now? And don't get me wrong, there's a lot of opportunities for augmentation with job or role or specific function replacement, but the opportunities for augmentation are instant and the results are immediate. And so kind of like what Tom Brady mentioned in one of his talks as he was closing things up is, you know, one of the reasons why he was so successful is because he didn't go for a 60 yard touchdown every single time, right? He took what was underneath to get down the field and score methodically and basically level up consistently. And that's the way I look at leveraging this so that you can get a 20 to 40% lift in only 60 days. So here's what I'm going to do is if we look at this and kind of my question for you, and this is another reframe that I want you to think through is like, all right, if you have the ability to instantly increase the capacity team by 20%, let's say 20% on the line over a two month period, what do you think that would do for revenue? What do you think that would do for profit? What do you think that would do for cash flow? And what do you think that would do for team morale? Right? So those are the four things that I've been blown away with in terms of the feedback I've got from clients I'm working with. I'm going to go through like the macro concepts and then I'm going to get into the micro. So I'm actually going to show you some examples of assistants and agents and how I use them. This is basically the patterns that I looked at with companies that I've been working with. One company is basically last year grew from 35 million to 70 million, on path to double that again to 135 million. Another went from 15 million to 30 million and is looking to do it 60 share their bootstrap. And I've worked with a lot of PE back companies like Thoma, Bravo, Insight Platinum Equity, KKR et cetera in terms of the fund portfolio. So let's get into the details. I don't want to show you real world examples so that you have some ideas going to even walk you through the outputs of some of the agents that we had, uh, as we go through this. Okay, so let me share screen and we will get rocking. All right. I keep. I need to be the closer. Melanie, you didn't tell me that. It's like super cool. All right, here we go. All right. Hello. Hold on. Stop sharing my screen. I think I hit the share system audio on accident. I don't want to do that. Let's try it again. All right. And I should be able to look at my chat now, so I will still need your assistance but I'm going to have this roll in. All right, here we go. So what we're going to do here is I'm going to walk you through some examples because that's one of the best ways that I've learned in terms of executing this. And I always like to look at results first versus shiny tools or features and because that's the way I drive things. Right. Uh, and at the same time I recommend any kind of AI solution should be driven around like what's the KPIs or outcomes you want to and then basically look at the tool or model selection on top of that and then the exit path. So we're going to go over through that kind of step by step. So here's an example. Uh, Fred, I love Fred. He's the Chief commercial officer at Stage 22. And one of the things that we did is we rolled this out to his entire go to market team and he's one of the organizations I mentioned earlier on that doubled in size last year. Talked about it on my podcast as well. The scale up show at the same time path to double again this year. And when of the things that they did was they multiplied effectively their team by 20 to 40% on top of it too, they were having junior reps close the largest deals of the year, which I thought was amazing. And then at the same time team morale went through the roof. Okay. And this was looked at across everything from marketing, sales, partnership and customer success. So that's an opportunity no matter what department, division, space you're in, whether you're a leader, individual contributor. All of these areas are right to take advantage of this. And I'm gonna walk you through some examples in a little bit. Okay, I'm gonna try and hit on all of these. So on the marketing side, here's something that Rami experienced. The cool thing about it is he told me that effectively, once again, he's able to 5 to 10x the output of his team because of some core concepts and areas that they've implemented. One of the things that I love that they did is they actually made an improvement in 15 to 20% of SEO traffic in literally one month from basically taking the questions that they were getting from customers on their chatbot or on their site or through support, basically turning those into SEO blog articles. Okay, so there's little things like that that you could do that make a massive difference with a tiny, tiny investment and even a tool like ChatGPT or Gemini. Right, so that's an example. Other things that I'm seeing are some pretty amazing results when companies are working cross functional like you'll see in here. Uh, this one organization basically launched an Olympic themed security awareness training campaign. Used to take weeks, took it in a day. Same thing, they generated 12 MPLs from a single post. And at the same time they've automated, semi automated a lot of their features and capabilities for blogs and award applications. Okay, so those are a couple other examples. And then here's, you know, kind of continuing on the sales aspect of it. There's a sales rep that closed a $30,000 deal by using ChatGPT for non emotional negotiation. You know how many people have been through the example where they get that, that ugly email at the end that basically says like, hey, we've decided to go with another provider and at the, um, they're like 30% cheaper than you. Okay, that was the example he used it in, used ChatGPT to help overcome those opportunities and at the same time win the deal. And a lot of these, what you'll see is there's really simple ways that you could filter some of this into assistance for the repeatable activities or even using agents and assistants combined. So I'll show you some of those in a little bit. Uh, I'm going to keep going through a few of the rest of these because I want to leave some time, uh, at the end of the session as well. But this was one of the things, if you're looking to amp up your team, there's so many cool things you could do as a CEO with this. I nerd out on this repeatedly. So. Okay, Unhealthy. All right. Keep dropping them in there. Okay, here's one of the things from a leadership perspective, right? I saw the CEO on there, so I'm like, I gotta share this. How many people know Brandon Taylor? Does anybody know Brandon Taylor? He's a member of the community as well. Give me a why in the chat, if you know him, maybe you don't, or if you've heard of him. Okay, so I met him at a Pavilion GTM last year and he was really looking to spend more time with his family. And so there's opportunities as a leader to free up five to 10 hours a week by using the PhD reasoning models, especially with O3 that was launched, whether it come for strategic planning, evaluating your team, communicating up and down, and then once again you could put those in assistance as well. All uh, right, I'm going to finish up this part and I'm going to start getting into tacticals. We have some other examples and then if we go down to the individual contributor level, these are other areas that I wanted to highlight. Basically folks are taking the jolt effect reverse engineering loss deal and then using those tactics and strategies to close them. And at the same time, um, I love this example. One of the enterprise top sales reps at this organization basically used O1 Preview, which is now O3 to do the manual work of two to three people in minutes for a complex custom pricing strategy for an enterprise client. So the sky's the limit for this. And the reason why I wanted to show you that is because this is what's going to feed into these two areas. So, so ChatGPT, we look at projects versus custom GPTs. And by the way folks throwing this out there, I just mentioned this to a CRO I was working with earlier today is one of the reasons why I focus on ChatGPT. I want to highlight this because there's a lot of randomness across the uh, LinkedIn sphere about this. Okay, so ChatGPT projects are good, they're not all bad just because they're not a custom GPT. And the three things I want to call out are these areas. Model availability, okay? Model availability means you have a wider range of the more like enhanced models than you do with only a custom GPT. This is one of the things I'm in the OpenAI private forum group and I said, hey, why don't you give the PhD reasoning to GPTs? Okay? I didn't get an answer, but I'm going to keep pushing until I do. So at the same time the other areas are if you think of projects These are only for you. You can't share these with teammates as of right now, but they're also good as like an organizational drawer. I'm going to show you what these visibly look like in the tool in a second. At the same time though, custom GPT, you can integrate these into other applications. Now, I'm not only working with ChatGPT, but there's other capabilities like Copilot or Claude or Google Gemini, and they all have kind of different variations of this. I focus typically on ChatGPT because they're usually three months to six months ahead of the other providers. Uh, doesn't mean the other providers aren't good and they're catching up fast. But that's why I'm going through this specific example. The other thing, like I said too, is custom GP's are shareable and distributable. That's internally or externally. All right, so those are a couple of areas I just wanted to highlight. Along those areas, let me actually show you an example of what that means specifically in a function. So if we're looking in ChatGPT, you see all these projects over here. This is what I refer to as my own personal assistance. At the same time, Claude has the same capability as what you're seeing here, where you can create projects and effectively with that you can create your own teammates. So for example, as this is a survey analysis tool, I have a LinkedIn teammate, I have a YouTube teammate, I have a copywriting teammate, or team, if you will. I have one for my podcast and then at the same time one for, uh. This is a little bit of an experiment, but I'll show you. It's really cool about basically using an agent to deconstruct the thought leader and then recreate content in an innovative way in their style and format. Now, the goal is not to copy people verbatim, but it's going to give you new opportunities. Okay. And so I'm going to go back to basically the examples, but I wanted to show you. I'm sorry, I probably should hit one other thing real quick. The other area, if we look at this, if we're looking at GPTs, we're going to be on the side over here. You could pin them to the side. And like, one of my most favorite ones that I have that I use is my aic. So I have a CEO version of myself that knows everything about me, my goals, my progress, and I'll even show you some high level details in here. But basically the amazing thing about this is you could put a Lot of different documents and information in here. I don't share this with anyone. Right. So even though this is ah, something I just use for myself, I, I put it into a GPT. All right, so these are a couple examples of like the difference between assistants and GPTs. The one thing that surprised me about this is basically you have the capability. Like I went in there and something one of my prospects or actually one of my customers said really triggered me and I said I'm like this makes no sense. Why am I getting so triggered by this? And this gave me like a really good like therapist quality level response as an executive that I was blown away with. So I've also helped CROs, CMOs, um, CEOs build these out. Because of that, when you're in an executive position it's hard to always have people relate to what you're going through. So anyways, just throwing that out there, if you haven't done that, that's something that like I've been blown away with as well. All right, so let me shift back over. Matt says, does a ChatGPT project leverage all context within that project? What are the best practices for structuring those interactions? Oh, that's good Matt. You know why? Because that's what I'm about to go into. So that's like leading, you know, leading the exact question. Like I didn't give Matt the question to ask me folks, just as a heads up. So I will show that right now and I'm going to show you the exact framework of what to do so you can have it leverage all the context within that project. Okay. And I'll show you how to do that. So I'm going to even go through some specific areas. I wanted to point this out though because Sam's pretty bullish on it that uh, it could literally uh, physically approximate dive do single digit percentage of all economically valuable tasks in the world. Now the other thing that's exciting yet scary and alarming in here is it basically scored the highest on uh, basically humanities last exam like at a ridiculous rate. And so the quality is getting much, much better. All right, so for, for folks that haven't used that, and I'll also include Google Gemini as another example. Those are really amazing tools. I'm going to walk you through a framework for creating an assistant or a GPT right now. So it kind of leads back to Matt's question. All right, so here's what I would look at. If I'm creating, this is the best practice example that I would use. I know There's a lot of marketing folks on here, so I would create something on your audience. Now there's some really unique ways that you could do this with. Okay, a lot of companies have Persona cards or built out icps or other examples that are already basically corporate assets that they have structured so you could use a piece like that. But what I've realized is audience is one of the most critical factors to generating amazing outcomes for whatever you are doing, whether it's sales or marketing, CS account management, you name it, or being the CEO. The other thing that I loved is I took the winners of my sales calls that converted and I took it, not only just the winners, but my best like favorite people that I love working with that have fast close times and high ltv. And I basically analyzed those sales transcripts and created an audience card which I'll show you in a second what that looks like on those winners. So I could duplicate those. Okay, so I just want your wheels turning so you understand what's possible there. The next part that I look at when building an assistant or a GPT and you could do this with agents and I'm going to link where agents come in with this is in this case we have marketing, right? So you could do specific tasks. What I mean by that is like I need to create a landing page so you could use OpenAI and I'm going to show you all examples of this soon. So this is going to be all theoretical but I'm bridging the gap. You could use examples of like, okay, identify a 50 page report on the best practices of building on a landing page and then you could infuse that into basically an assistant or a GPT. So you're using the agent to create the training data on the jobs and expert side and then filtering that in with your audience. And what you're going to do is start stacking other, other training data in there. Okay, so for example, there's the expert training data, there's Seth Godin here on the left, there's Gary Vee on the right. You could basically reverse engineer their social content, their content strategy, their key frameworks. And, and the way that you do this at scale really fast is you could use it with a deep research agent, whether it be in Google, which is unlimited, or you could use it in ChatGPT, which you have a limited of 10 uses per month or I think it's 12 uses per month now, um, for a plus user or a Teams user or 120 uses for a uh, basically a pro user which is the $200 a month version. Next step that I would recommend is or do you have the capabilities to do is not only do you have the ability to reverse engineer frameworks from thought leaders and it doesn't need to be people just on social media, it could be authors, it could be a lot of different examples, but you could reverse engineer brands. So, and I'm going to show you this one. For Apple, I reverse engineered their entire brand and it includes everything from their, their visual branding board to their copy to the typography, all those different examples in like a 50 page doc that I could start to emulate if I wanted to do that in my content, my creation, my messaging, my pov, anything. So we got our audience, we got our jobs or tasks and then we got experts. All right, so what that leads to is this next one is innovation. Now the beautiful thing that happens when working with these GPTs or assistants is if you stack these three examples, audience jobs, you can start to generate new things at an insane rate. Okay. And this is what Steve Jobs actually nobody knows about. And sometimes I was, it's funny, I was at the CMO Summit, I was sitting at a roundtable with about five different CMOs. And one of the things that they said is I'm like, hey, you could reverse engineer this, you could reverse engineer Apple's brand. And they're like, I don't feel good about that. Like that's kind of, it's kind of, that's like they felt dirty, right? And I'm like, here's the thing. Like innovation comes from combining two different things in different areas. An example of that would be abt, which is a company out of Chicago. And they basically infused. It's, it's like blowing up one of the most amazing electronic stores that blows Best Buy away, which sounds like Best Buy, but uh, they have Vegas type hospitality in an electronic store. So think of those combinations, right? That's the same thing that we're going to do here. I'm going to walk you through some of the examples of these outputs. So I'm going to show you my audience guide. Right? So this is something that I created and I basically had a reverse engineered. Now you could use. Actually let me ask this, how many people have used O3 before? We have some pretty savvy users that use agents. Here's what I'd say. It takes a little bit of time, but you gotta be patient with it. But you could basically throw sales transcripts in here and create this training data. So I have this Everything from core voice and tone attributes, language styles, writing, do's and don'ts, content structure, preferred formatting, this communication style, and all these different areas, right? Content themes, framing, content examples. So these are all taken from actual calls. So you can reverse engineer the language of your customer and then you can infuse that to create content enablement, messaging, any kind of detail that you want right now. Here's what I did with Apple. Do you all want me to show you the Apple one? People typically love the Apple one. And I could go, I could go crazy on this one, but I could show you the Apple. Let's do this. I'm going to show you the exact example. I'm dumping around to different accounts while I'm looking at this so you can see it. All right, I'm going to show you the exact example. I'm pulling it up here and we'll show you different areas. Do you want to, do you want to know how to prompt to create this training data from, from Apple? How about maybe we start there? That might even be better. I think we should do that. Okay, well, I'm going to do that. I'm going to show this to you right now. One of the simplest things you could do to prompt chat GPT is this example. And you know what? I'll even. I could drop this in the chat if you want. I know everybody's like prompt hoarders, if that's even a word or a thing. I should get like a license plate that says that, uh, I have dreamed in prompts before. I don't know if anybody else has, but. Okay, kind of sad, but true. All right, so here's what I did. The good thing about, like the bad thing about OpenAI's deep research is effectively you only get 10 shots. You only get 10 at bats, right? If you're using. Not a power user like me, right? But what you could do, and I found the best way is to identify what your goal is. And then you could use a verbal tool like whisper, where you just speak what you're thinking or you could type it out. So I typed this out and I asked X3 to create a prompt for me for deep research. Okay? And this is, this is very, very simple, right? It wasn't super complex. And what you can see here is it basically wrote this out. So I'm going to put this in the chat. You are now, are you a prompt? What's that? Who said are you a prompt? Darn it, it's too long. Okay, well, we'll do something Maybe I'll drop it. Hit me up if you want it. And um, I also have a gift for you at the end, if you stay till the end as well. What's this poll? Melanie, are you a prompt? Is that a joke? I don't know. Are you a prompter or designer? Prompt hoarder. Okay, it cut off on my end so it just says are you a prompt? And I'm like, okay, this is fantastic. All right. So anyways, this is the prompt it created for Apple's Visual Style Guide. Now as you know, I said visual Style Guide, but it's got everything from like local variations, positional guidelines, do's and don'ts, color palettes, typography. Right now I'm like, hey, I want to rewrite this as a prompt template so I can blow away a, uh, top 1% CMO, right? So I had him rewrite it. Him or her or it, right? And it's much deeper and richer. So sometimes the best ways to do this are to effectively really go deep on this and iterate back and forth, as you can see what I'm doing here. So then when you launch this into Deep Research, you get amazing output. And as you can see, I went through multiple areas. I added like key phrases and copywriting principles. Okay, this is a deep level of nerdery, as you can see. I'm going to show you the output. But this is like how this is just a prompt. This isn't the output, this is just the prompt. All right, and now here's my ultimate tip. One of the simplest things to do, and I tell my clients this all the time, is tell after it's done, whatever it output it said is make it 10x better, more impactful, more visually stimulating and you'll be impressed with the level of results that it comes up with. That's all you have to say is once it creates a prompt for you or an output, say make it 10x better. Right? Make it 10x better. Really good tip. Okay, so now this is what it created. This is the final winner and it's got the brand overview, logo usage, color palettes, custom typography, imagery implementation. It actually gave prompts and key phrases, copywriting, data driven insights, brand resilience, and then final deliverable. Okay, so that is the example of what I gave Deep Research. Now all I did is I dropped that in here. Then I went to Deep Research, executed that and it created a massive doc and file for the outcome here. Now I'm not going to show you this whole thing. I'll actually show you in a Google Doc. So it's a little bit more digestible or from a visual standpoint. But as you can see here, this is like the table of contents. It goes through all these different areas specifically on Apple. So if I wanted to basically have Apple and Tesla make a brand baby, I could mix those two together and create a new brand innovation just from that, put that in as an assistant or agent and then start to generate content ideas, output from that. Okay, I know I'm really hammering home the outputs of like the training content of how important that is. But that'll just give you an example of what, what it looks like. Now the interesting thing too is like I love the copywriting principles that it executed in here. Really, uh, talks about clarity and simplicity and emotional resonance, but at the same time, you know, it really even talks about crisis communication, key phrases and messaging. And it even gave me LLM prompts in here, which I was pretty impressed with because we talked about it. But it's everything from like a HERO page to a digital ad social media. So it goes through a lot of the basic stacks that you would need to leverage. I would use the Google Docs as is training, uh, data. Or you could use data that you have and synthesize that in unstructured format. So that's an example on the brand side. Now I'm going to show you another example of, um, kind of the assistance and what an output looks like. I use this one. How many people like Alex Hormozi? He's a pretty damn good copywriter. And what he does is he has really, really strong language integration. That's almost like addictive. He's like a dopamine dealer when it comes to that. So I'll give you an example. I basically reverse engineered his content from X so that I could put this as training data in assistant. So I use deep research to do that. That's what this file is over here. And then what you could do is I'm like, all right, I'm gonna, I'm gonna like ask it. Just throw it a phrase and say I want 10 posts created on it. Right? And I wanna show you the results of this. Cause this was, this is pretty impressive. So I started with AI literacy. Let me show you this. Uh, and so look at this language though. These are like 10 tweaks that are pretty powerful. Like, here's an example. People complain about not having enough time while ignoring free no code agents and giving back 20 hours a week. Your competition isn't smarter than you. They're just leveraging tools you haven't bothered to learn yet. You know, there's a lot of like fire statements that are created from that. You know, I asked a hundred entrepreneurs that's holding their business back. 87% not enough time. Well, 92% couldn't name a single no code agent they've tried. The disconnect between problems and solutions has never been greater. Right. This is like strong. So anyways that's the like an example. So if we go back to it, really what we're doing is with these. And just like I showed you, I have a team of probably 20 different assistants that I've trained by using deep research as an agent to create high quality outcomes. And if I were to basically hire that team to do that on my own, it would probably be $20,000 a month of all the work those individual providers are doing. So I need to map it out. I imagine it might even be higher by now. But those are the things that I'm looking at bringing to individual go to market leaders, individual contributors or CEOs so that you understand like basically what's possible now. You don't need to wait, you don't need to throw the 50 yard bomb and wait till all agents are autonomous and you can just replace people left and right. Because I don't know about you, but I don't want to be replaced. I want to keep working. I want to help. And so to kind of wrap things up is what I would say is let me share this last kind of kind of hit home area and then I have some free resources for you to leverage because I love it when people give me free resources to leverage. Okay, so this is what you effectively want to build out. This was an example I did for an SDR team is like we built out all these different agents and we started seeing conversions double. We started to see a significant increase in pipeline. And these are the areas that you can start to add digital workers now with the tools that are available versus waiting that could give you that 20 to 40% left. All right, so quick teaser. So don't just think of it as time savings, think of it as capacity multiplication without hiring. Right. While also basically using the frameworks of top thought leaders that have invested sometimes 20, 30 years to build that expertise that you could use on demand. Right. You're going to walk away with a 90 day GDM sprint. You'll be able to apply this to real world problems. So please check that out. That's one. And at the same time it kind of Wrap things up is, like I said, if you want to talk about this for your team, feel free to hit me up. You can hit me up in Slack or LinkedIn. I can let you know I've worked with about 12 different teams to 15 different teams. Can't remember because they keep happening for basically rolling this out and creating a superhuman team within only 60 days. So they could add 20 to 40%. All right. the same time, you know, if you don't have an AI strategy for a go to market plan, then this is something that I could talk to you about as well. So connect with me on Slack on that or LinkedIn. And then at the same time, I have a resource for you. So in conjunction with Captivate, I am working with Chris, uh, Gannon over there and his team, and we're doing an AI agent and AI benchmarking survey prompts and use cases for go to market that I built. I think there's like 170 of them in there that you can literally use. And there's about 20 different cheat sheets that are built out in this resource. Just by answering that, uh, benchmark survey at the same time, you'll have the ability to really unlock and know how you compare versus everybody else, which, when I work with CROs and CMOs is one of the biggest questions is like, am I behind? And some people, the answer is yes. I disagree with Owen on the last one. There are people that are behind, there are people that are ahead, and there are going to be winners and there are going to be losers. So, um, but the point of this is so that you could see where you stand, you could do something about it versus that ambiguous thumb in the air. Okay, that's pretty much all I have for today.