How to Scale as a Solopreneur Without a Team by Leveraging AI with Mike Todasco

Alright, so let me paint you a picture. You're going through your newsletters and you're seeing these titles: AI replaced my team. I launched a product over the weekend. I'm scaling, no hiring. That sounds wild, right? So you go ahead and you try it. And it's a lot messier than you thought it will be. You buy into a bunch of AI tools. None of them talk with each other. You start with simple prompts. You invest time in, at getting better at prompting. So you go to advanced prompts. You're still not getting what you're looking for. And you can't shake this feeling that everything is just taking way too long. So that's not your fault. Majority of solopreneurs use AI without a system. No workflows, no structure, no AI leverage. Hey, what's going on everybody? I'm Mike Sirius and this is Mastering Tech Growth. Every week we sit down with industry leaders to bring you no-fluff tips on how to grow. If you enjoyed the show, click that subscribe button and send me your biggest growth challenges at masteringtechgrowth.com. I really want to know what you're working on and what keeps you up at night. Joining me today is Mike Todasco. He's a former director of innovation at PayPal, now a visiting fellow in AI at San Diego State University. Mike built labs, launched products, holds over 100 patents. But today we're talking about something much more practical: how to scale a solopreneur without building a team, but leveraging AI instead. Welcome, Mike.

Mike, I'm glad to be here and we're gonna talk about one of my favorite things today. So I'm excited about that. Not only I love like AI and having these discussions, but the things that it can enable for somebody who wants to start a business, it's really amazing.

It's really amazing.

Hopefully we can help some people out.

Two Mikes in the room today.

It will get confused. Yes.

Two Mikes in the room today. I love that. So we were chatting, you know, before recording this, and you said this phrase, which just stuck with me. You went, we are in an age of solopreneur, which makes perfect sense, right? If you're leveraging AI in a real way, in a real enabling type of way, you are able to do so much more with your time. And a lot of functions you can offload to AI assistants, which just effectively enables people to go in and and launch tools, launch services, launch coaching, and many other things. So before we go into that topic, Mike, though, we love asking this: when I say Mastering Tech Growth, what comes to mind?

Oh, other than your podcast, um, Mastering Tech Growth, like, look, I think the first thing that comes out to my mind, seriously, Mike, it, it comes down to experimentation. And that's one of the most important things in the world of AI today, is that you need to just try stuff because we're all on the cutting edge of this. We don't know what is going to work, what is not going to work. This is one of the first times really in the world of software that I've been around that people are making software for consumers or businesses and they're not even really sure of the full capabilities or how these things are going to be used. You saw snippets of this, like take Microsoft Excel or Lotus 1-2-3 spreadsheet software as an example. The original purpose of that was for accounting and the other stuff. And that over years evolved into many things. Like people get really creative with these tools and what they can do and so forth. Like, yes. But with AI, it's out of the box. People are creating stuff that you would never even imagine. I was creating just last night as a joke, I created a fake iMessage text thread using ChatGPT's image generation model for that. Like, was it built for that? No, no. But like, it could actually make a text thread going back and forth like you would take a screenshot of that or any other type of screenshot because it's been trained on so much data. That was never the intent. And there's probably tens of thousands, hundreds of thousands, millions of other uses like that. And so to master tech growth, you gotta just understand, you gotta just play with these things and you just have to experiment.

I love that. No one gave that, by the way, Mike. I mean, you know, we are close to 100 episodes. No one gave an answer of experimentation to this question. We get a lot of choose the right problem, you know, make sure if you're doing something or you're fixing something the market actually needs, ready to pay. We had sales, you know, we had marketing. No one really said like, look, like, If you want to master this art, you have to actually experiment. I love that, especially—

yeah, it's all about playing. It's embracing— one of the things that I do— so later today I'm at my alma mater Berkeley. I'll be giving an AI talk up there as a guest lecturer, which I often do. And I end that talk with saying the one takeaway I want you to have is to embrace your inner child. Like, you need to experiment, you need to try stuff, you need to fail, you need to just kind of like— and, and that's how you work with these AI tools is doing all of that just as you would when you were like 7 years old and just learning how the world works.

Uh, the world was so much simpler.

It was because it's not because the world was better or anything like that, it's because we were kids and we didn't have cares in the world. Exactly.

There you go, there you go. All right, let's go in. So a lot of solo founders use AI to save time. This is, this is how the market right now is positioning the tool, and this is what you're seeing a lot from AI experts coming out: use AI, save time, right? But the real shift really happens when you use it to remove the entire task off your plate. So it's not about making the task shorter or easier, it's just removing it. It's just, it's just using AI to fully remove it, right? So, and according to McKinsey, up to 60, from 60 to 70% of employee time spend right now on repetitive tasks can be automated with AI, right? So the question is, from your perspective, Mike, potentially you're already doing that with your own business, or maybe your day-to-day tasks or practices, or maybe you know someone. What would be the main tasks, or number 1 or number 3 tasks, which AI can just remove from your plate right now?

Um, I want to get to that, but I want to get a little to philosophy first, but I will get into your question, Mike. Um, I think for everybody, with all of these tasks that you have, it's 3 ways to— it's kind of 3-step process. Okay, like I have a task and it is, um, I don't know, sending out invoices, uh, to clients, you know, based on whatever it might be. With all these tasks, you should look at eliminate first. It's like, does this need to be done? Is there value being added? And so forth as part of this process. Like, do we actually need this? Um, I think for solo founders just starting up, it's a good question to ask because you kind of get the first principles type thinking and just say like, huh, do we actually need to do this? All my competitors do this, but do I actually have to do that? For people who've been doing this for a while, might be like, well, that made sense 7 years ago, but does that actually make sense today? So number 1, eliminate. Uh, number 2 then is, is automate. I, or, um, you know, what can you actually do? Are there steps in the process that you could do to automate and so forth as part of that? Um, and then like building off of automation, I think there's a whole nother thing because automation, at least in my mind, like usually implies like there's a human in the loop, something like that. Um, at the end, like to me it's like pure step 3 and the holy grail for this is just AI delegation. Um, and in that you're just going to be pushing this off on a system who's going to be doing it. And yeah, there's going to you probably need some human-in-the-loop type check-in or something like that. But maybe you're not even kicking off the process. Maybe the AI is just running with it or whatever it might be. So like, that's kind of how I think about it. Now, to your question, Mike, of, well, what are the areas where like the biggest opportunities lie?

Yes.

Look, the first one, you know, every business is different, so everybody's doing stuff, but I'm trying to think like, what are the things that hit across the most areas? Like, the number one thing for me would, would frankly be in the world of email. Yeah, um, I think it's shocking how much time people spend in email clients, you know, and I'm sure a lot of it is essential— talking to customers, finding new customers, talking to vendors and suppliers, so on and so forth— as part of that. I think the process that most of us do it is extremely inefficient. We get nice dopamine hits whenever we're able to send out an email, we accomplish something, we do something, it might be, but especially for solopreneurs, nobody is going to be judging your business acumen on your email ability. It's gonna be based, again, there's elements of customer feedback and contact and all that in there. But like, look, that's the first thing that jumps out to me that is going to impact just about anybody who's a solopreneur. Or a business owner. So what would I do in that world? What are the tools out there that are available for that? First, I would say like before you even get into the AI tools, build a better process. You should be batching these things. Like the work of context switching between email and between other more value-added tasks that you have is probably not worth it. Do all of your customers— are you in a business where your customers and vendors need a response within 20 minutes or something like that? Probably not. I mean, I think just realistically, it is probably not the case. Could you batch this where, okay, you're gonna actually mark off time on your calendar where every 3 or 4 hours you're going to be doing this as email time? Great. That's probably more effective. Take it off of your phone if you need to, whatever, just so you don't get distracted so you can focus more on building product, more on, you know, one-on-one in-person interaction with customer, whatever it might be. But then using, so doing that first, again, do kind of the human procedural psychological type stuff first, but then bring in the AI tools. And a lot of this is going to be built into tools like Gmail and so forth. Like, you know, actually in that same presentation I'll give later today, I use the example of SwiftKey. So SwiftKey was an email app for the iPhone, which launched back in 2010. And one of the things that it did was it would predict the next word in a sequence. Um, and like, that was kind of revolutionary. So you said, "Mary had a little," and it's probably going to say the word "lamb" down there as the suggestion for that next word. And it's using very basic machine learning stuff, which scaled up is kind of what we're using today for— in very simple terms in the world of AI. But that's much more advanced. So it will be in, you know, Outlook, Gmail, all the email clients already. but you can really even supercharge it, dare I say, with like products like Superhuman. I know a lot of people who use Superhuman. It's a company that sits on top of your existing email clients. I think it costs like $30 a month. I don't use it. I'm not a super emailer, so I don't use it. But like for anybody who's in email for potentially hours a day, like that is a huge benefit where it will kind of learn your writing style. Where it will be able to do all of this other kind of stuff. And I'm sure there's other companies other than Superhuman out there doing it too. I'm not suggesting they're the only one, but it's some type of tool like that that can, you know, auto, you know, auto triage things in there in an easy way based on knowledge that it has about you. Email systems, you know, can give you suggestions of like, hey, this is what a good response is. You want to just hit send? You want to edit it? All of those kind of things can be huge, huge time savers, and that's available today. That's not something you have to, you know, some of the things I'm sure we'll talk about are like, yeah, that's 6 months away, that's 12 months away. No, you could do that right now.

Okay, so there's a lot to unpack there, Mike. Um, I love how you, how you started with, uh, with just thinking about the tasks, and it reminded about a quick talk I, uh, I listened, uh, when Elon Musk was talking about how you approach automation So he had this one quote which really stuck with me, which goes, if you're not occasionally adding steps back into your processes, that, that means you're not removing enough.

Oh, sure. Yeah, yeah, yeah, yeah.

So effectively, the mindset is that you have to be very aggressive about removing steps because people tend to overthink what they need to do within their processes. Well, maybe this will be needed later. Well, maybe this is going to be an edge case, and they just overcomplicate their processes and their workflows. What the mindset is actually is be a lot more aggressive of removing steps and occasionally add them back in if the situation appears where the step is missing. That, that sort of mindset is actually better. So we saying first audit your, your workflows, right? That, that's what we're saying. Audit your workflows, audit your processes, make sure they're good. Only then move into automation and only then move into AI. Enabled automation. And you were talking about superhuman, Mike, and I was like, okay, I'm going to share this. So I have AI automation workflow, which I built myself with the automation tools like Make.com or Zapier. I actually use N8N. So I have, when I get, I have an inbox, which is effectively a podcast public inbox. So we get guest pitch through that inbox and other outbound opportunities. So when I get a guest pitch, that actually goes into my workflow, which is fully automated. It goes to AI agent, which is trained on the knowledge of a show with the local Ollama module. And effectively goes like, is this a guest page? Yes, it is. Then we look at the guest, we go into internet, we do a search, come back with some context, it evaluates everything, and it sends back a response going like, that sounds interesting, you know, here's a couple other things. So that step itself, you know, took me some time to build, probably a day or so. But it saves me so much time because I get these emails, like I get 6 or so every week. So it's a good example of a workflow which you're doing every day, and then you're just simply raising a question: can I automate this? You know, does it need me, you know, to be there? So do you have a philosophy of like how you would approach a business? Let's say, uh, you go to a business, something small, you know, maybe 20, 40 people Would you have a process or a framework of how people can think or what could they do to actually get good processes to automate? Is there like a, you know, a green checklist or a checklist, something like that?

Here's what I would do. I think it's a great question. You need to look at it from an outside perspective. So the first thing that I would do was I would probably just— you want to keep a piece of paper, blank sheet of paper out by your computer, in your pocket, whatever. It's probably better to have on a sheet of paper than on your phone. Phone. And I would actually write down, literally, if you work from say 8 to 5, I would write down all the times on that sheet of paper there, right? I write down all the times of day, 15-minute increments. And I would maybe use your phone, or, you know, you could get apps that would do this, like literally have a timer go off every 15 minutes or something.

Yep.

And I'd write down what you're doing during the day. I would start with just like literally doing an audit of what your day looks like.

For a week.

And then at the end of the day, I would go back and I would look at that.

Just one day, Mike? Just one day? Like, how long are we doing this? Is this like a full week exercise, full month, or like just one day?

Look, it'd be nice to do for a month. I would just start with a day to do it. Like, you could automate this as well. Now that I'm actually thinking about this, you could use a tool like, uh, B Dot Computer. Uh, it's like— it's a little $50 AI bracelet. I actually have one of them. I usually I don't wear it now, but I've had it. It will actually record everything that is going on around you. So every conversation and so forth. And you could probably then use the transcript of your day to do that. And it's like literally like $50 to get one of those. They're pretty cheap. Yeah, do it for a couple days. It would be ideal. But I also know people are busy. If I say like, ideally do it for a month, nobody's gonna do this for a month. You're busy, you know, it's tedious, whatever. But I think even just doing it for like 1 or 2 days. It's going to be eye-opening. Try and do it for a typical day. And if you're like, well, Mike, there's no such thing as a typical day, great, just do it on a Wednesday, like whatever it might be, it doesn't matter. Um, and then I would actually just start there with like really breaking out what it is. Like, like hopefully people are already kind of saying, well, this is— these are my goals, these are— this is what I want to hit. Like, again, I, I'm a bit of a goal setter. I kind of think about, okay, my priority for the day is this, you know, my priority for the week is this, my priority for the month this and so forth. You don't have to get that granular, but you probably— it is probably worthwhile saying, okay, if, if my day looked like this on the sheet, was that matching to what my priorities were? I would take that list. I'd probably have like a highlighter or something. Wherever something was working towards the goal that you had for that day, highlight that thing, like shine, you know, take a look at that and then step back and say, okay, Well, what did my day look like? In reality, 20% of my time was doing stuff that was towards my priority. 50% of the time was doing busy work that I think is important. And then 30% of the time was like, I actually don't know what I was doing. I was on TikTok or something like that at that time. And so from there, the value is then doing that audit and then looking at I mean, start at that 50%. That's something that's a little bit more solvable with AI. The stuff that is like, okay, yeah, I need to do my taxes, uh, I need to send out invoices, I need to respond to customer complaints, I need to do whatever it is. But is that stuff that can be eliminated, automated, delegated to an AI? And that's how I would start to go through that process. Tackle the things that are— with all of these things, I would say go for the quick wins first. Yeah, go, you know, I— if something is only 5% of your time but you're like, huh, I really could automate that, do that first. You know, even if something else is like 25% of your time, um, I would actually go for the quick wins. Just see what it's like to achieve that, to automate that with AI, or to, to delegate that out to an agent, however we want to call that.

Yep.

Um, I would do those things first and then move my way back up the stack. Back to, um, you know, the bigger things.

That's such great advice, Mike. I love that.

Yeah. Um, one other thing, Mike, as you were talking, I, I was actually looking through— I read a lot of books. I read too many books, I will say. But as you're talking about like eliminating stuff, it got to— and I had to look up the title of the book because I couldn't remember. Um, a few years ago, this book came out called Subtract, and it's all about how our initial inclination is never to remove stuff. It's always to add more. Yeah, it is always to do this. And, and one of the examples, the canonical examples of this, they talk about in the book, and I think I talked about it elsewhere, is the Lego bridge experiments. You are basically— you need— there's a bridge, and if you imagine one side of the bridge is higher than the other, um, like that, and there's a slope down, and your task is to make the bridge the same level. And you're gonna get $5 to do that, and there's a whole bunch of bricks in front of you, and each brick that you use is gonna cost a dollar. What do 90% of people do? They're like, oh, okay, I'll take 3 of these bricks, I'll put this underneath, and I'll have the bridge level, and I get, you know, $5 minus $3 equals $2. I make $2 off of that. Even given those, um, and I'm oversimplifying it here, but what most people don't do is say like, oh, if I just remove 3 bricks from the other side, I will have it and I get to keep $5. The simpler, like, and again, our minds don't go there. Like, we don't think of doing less as being productive. Um, and Elon Musk was a great example of that. I think he's one of the people that is like kind best known for just like eliminating, eliminating until you would like have eliminated too much and then you got to add something back in. Yeah, it is not a natural thing for people to do. So you always think like I'm being efficient and positive and all that, I have to add more. No, no, you got to take a step back from it.

Is that more like a mindset thing then? Then, then like, are we talking about like you have to change a little bit your mindset of how you think about your work? And just become more lazy. That's not—

yeah, I got a lot of that in me. Yeah, yeah.

So it's maybe it's a wrong word to, to say, but don't, don't just be busy for the sake of being busy, right? Just take a knee, take some distance, and really ask yourself things you're doing. Is that actually— is the path of least resistance towards where you actually want to be when it comes to goal setting, right?

Mike, I think you're nailing it. American culture, in many ways, and, you know, and I don't think all countries are like this, American work culture is definitely like this. You know, it is a badge of honor to say that you are busy, to say like, well, I'm, you know, I'm doing so much. And I will say probably younger generations are probably a little bit better than my— I'm a Gen Xer, like, but like, like that, that's almost a badge of honor. So like, yeah, got so much going on and all that other kind of stuff.

Yeah.

Um, I don't know where that gets us because just being busy for the sake of being busy isn't building a better product for your customers.

Yeah.

Uh, it isn't really running your business and all that. It frankly doesn't give you time to just step back and think. I mean, the best leaders, um, out there are ones who take the time to get away from all that, who aren't just tech I mean, it feels really good again to clean out your email or to send out a bunch of emails. You feel productive, you feel like you're doing stuff. It doesn't feel as good or productive to go for a walk and think about things and potentially not have a solution for your problem at the end, or potentially not have something more as part of that. But like, longer term, I would actually argue, well, the walk is probably going to give more value to your business than the email that you're going to be sending out.

Oh, 100%. I love walking. Maybe not ever. I love going for a run during lunch. And I have this app, which is no surprise. It's an AI-enabled app, which effectively you just click on it and it allows you to record voice. And then it records your transcript, of course, and then AI can translate this into actionable items, send it to your to-do, Jira, or email, you know, or whatever. It works so much well, and I get my best ideas when I'm actually walking or running. There's something scientifically around what your brain does when you're in activity and thinking.

Mike, I love— if we could go on a slight tangent with this, I love what you're saying because, and actually I am exactly the same as you, and it probably took me— I was probably 30 years old before I realized that was the case with me. For me personally, I hate running. I actually hate it. And I think my mind going in this spot of thinking of ideas, thinking of all this, like nothing happens for me for the first mile or so, like nothing. I'm like, frankly, the first mile, the whole time I'm like, well, I should just go home. Why the heck am I doing this? I hate doing this. But once I get past that first mile, like my mind, 'cause so I don't bring any music. I don't bring anything. I want my mind to be bored. I actually have an audio recorder that I bought on Amazon. Just like a little digital recorder to do the exact same thing that you're doing. Because I don't want a phone, I don't want any distractions. I have like a separate audio recorder that I keep by my front door just for when I run. I go out there, I will talk to it, um, and after 45 minutes or so, I come back home and I listen to that and I just start typing feverishly. Like, my kids and family know to kind of just leave me alone after my run for 5 minutes. Yeah, because I want to like capture all the thoughts and ideas.

Yeah.

And the one thing I will say, for everybody it's different though. And this is one thing, when I was really teaching innovation and creativity and things like that when I was at PayPal, I would give a talk and I was like, "Everybody needs to find what their equivalent for running is." Yeah. There was the Masters of Scale podcast and Reid Hoffman. I think at one point they did a montage 'cause he would ask people this question. And for Reid himself, it was, he would go to conferences He would sit in the middle of a row and he would go to talks where he had zero interest in those talks, um, and he would just have a pad of paper. He would get so bored, maybe something would inspire him, but maybe not, and he would then just start writing, you know. Again, no devices, no nothing. The, the thing that kills creativity, the thing that kills ideation is this, is having a phone by you because you're like, oh God, I'm so bored, I need to go on TikTok. Oh God, that is the worst thing that you could possibly do for Great. I mean, again, I'm not saying you should never do that, but I'm just saying like you need time away. You need to like feed into the boredom that comes in your life. You need to find the ways to do that. The founder of Spanx, Sara Blakely, she would talk about— she would— her car was her place to do that. So when she would drive to work, sometimes on her commute, she would go for like an hour. She would just circle. And all that, because she would find like her car was her meditative thing. And if she had some ideas, she had something going, she would just like keep going around. She's like, I'm gonna just keep doing this until she'll show up late to work. But like, that was her ideation. Yeah, running, getting bored at a conference, driving, whatever it is. Like, everybody needs to find that for themselves.

And everybody needs to get— yeah, get in the state of being bored so your brain actually thinks on real problems. Yeah, people are not intentional. A lot of young solopreneurs, maybe not even solopreneurs, just in general, people are not intentional with the time. So I, I get— I annoy my wife with this question all the time, right? So she's on the phone doing some Instagram thing, checking some recipes or something like that, or just watching, just scrolling. I come in, I'm like, what's your intention with this? What exactly you're looking for? And she just flips off. It's like, just leave me alone, because Like, if you're going on your phone, right, and you're getting sucked into all of that, like, endless scrolling or just browsing or whatever, like, you have to have intention. You must be looking for an answer for some question. That means the time is getting spent intentionally. You clearly have some challenges to overcome. But if you're just doing it for the sake of doing it, you just got hooked by the sheer hookiness of these apps. They're very well done to just make sure you're spending time with them. So I always, I always ask myself if I find, you know, in these moments I'm like, what, like what I'm doing here, what I'm trying to achieve with this time. If the answer is no, just, just put it down, go for a run.

And I want to give you a great example of that. So I was at lunch with somebody, a friend of mine just recently. We were talking at lunch, mentioned something. He's like, oh yeah, that's interesting. He's like, yeah, I got to look that up. So he takes out his phone and then all of a sudden he just starts typing away. He just starts typing away. And I'm like, what are you doing? He's like, oh, I'm unlocking my phone. And I'm like, what? What? Like, what's going on? You've been out of it like 15 seconds there. What do you do? He's like, yeah, that's my passcode now. And I'm like, security thing? He's like, no, I want to be intentional whenever I need to access my phone. If I could just put a fingerprint, if I could just do a face scan, if I could just do whatever it is with that. Then I'm going to just mindlessly, whenever I'm bored, I got to pick this thing up. Now I got to actually think. I have to focus for like 15 seconds. And like, honestly, if it's not worth 15 seconds of your time to actually do that, is it really worth even picking up your phone? I love that. And I thought that was just like a great little hack. So like, you know, if anybody is struggling with this out there, that is a simple thing that you can do today. Make a really ridiculously long passcode, turn off Face ID, turn off all of that other stuff. Make it hard to actually open your phone. And so it's not turning it off, but it is actually saying then, oh, if I actually need this, I can get in there. If I'm actually just sitting in line at a grocery store because I have 26 seconds to kill before I get in there, no, I don't need to actually look at my phone.

I love that. That's a very, very, very neat, simple hack to implement to get, get away from that easy access to all of that dopamine, dopamine loads. You get from all the social media, right?

And I will say the most ironic thing about that story, Mike, so I just put, um, he is the person I know who has by far the most followers of anybody I know on TikTok. So he knows better than anybody out there what this is, and he knows like, hey, for myself, this is the worst thing in the world to do. Like, he's actually extremely successful on that platform. He knows that. He's like, yeah, but like, I can't get sucked into it. Like, he knows the, the poison that it can be.

Okay, I love that. Right, so let's jump back into the— in, in where we drifted, which is talking about the test. No, no, it's cool. I really enjoyed that bit. We just went on a, a little bit away from the episode, but I'm sure people enjoying as well. Um, come back to the test, Mike. So you mentioned all of this setup, uh, doing the, the paper. I, I really love this approach, just, just auditing your time. And then highlighting all these tasks. So let's say now you have your low-level tasks or low-value tasks, but you, you know, they still need to get done. You are solopreneur, right? So they become your first or a good candidates or first list of candidates for automation or AI-powered automation. Now a question is, you're not technical, so you can't really judge well what can be automated and when cannot. How would you approach that process of thinking about what can be automated and how?

Yeah, I would say it's baby steps with all this. One tool I'm going to suggest to people, which is, I think, mostly free— I don't know if there's a paid version, there probably is a paid version— but you could use a tool like Make.com. And that is a way to actually connect many different services. And I don't know if your guests in the past have talked about services like Make.com, but like, here's how it could work. Um, you could have— say you have, you know, multiple social channels that you need to, um, send something out on, um, that's related to your business. And yeah, you gotta log into Pinterest, you gotta log into Instagram and do all this other kind of stuff, and that takes some time, and you want to like, you know, at least start to automate What you can do is you can have Make— you could kind of, uh, you know, have it tied to, let's just say, ChatGPT. You can put into ChatGPT and say, hey, I want you to give me, um, all of these different, uh, like, this is how I write for all these socials, so give it examples of posts. Uh, and when you get a message, I want you to break it down for all these socials. Then Make can actually go in there and then you give it your login credentials in some way and it can actually take those and post those or save it as a draft and then post it or whatever it might be. It's using different language for each one of those. That's just a way to take something that's, okay, you're going to have to do this 7 times, maybe you just have to do it one time, write it generically. It's going to be able to tweak it for all the different channels and the right tone that you want it to be tweaked in. And it's going to be able to post it with one thing. It might not do it perfect the first time, and, you know, you might say like, oh, it's a little bit off with how it's doing it and whatever on Pinterest or something like that. That's not exactly what I want it to look like. Um, so you just— and again, this is experimentation, you just kind of guess and check and so forth as part of that. But like, that, that would be one easy example, uh, and make What Make is, it's a way to just connect apps to each other. If you want to connect a language model to your email and all that, it's an if this then that type of experience where if you see this type of email come into my Gmail inbox, then I want you to take the information from that, put that into ChatGPT. I want you to tell ChatGPT to give me this data, and then I want you to take that data and put it into QuickBooks or something like that as an entry. Like, if this, then that. And it's just kind of defining that whole workflow between software applications.

100%. That's a brilliant advice, Mike. That's a brilliant advice. I can add a little bit to it. What I would do as well, and that's what I do, and I give advice to all people looking for AI-powered automation is that majority, if not all of these services such as Make.com, uh, If Not This Then That, Zapier, N8N, which are like probably the biggest, uh, automation platforms right now, they have what is called GPTs or ChatGPT. So they have GPTs which are trained on their documentation. So it's a very, very good way of actually understanding if you should be automating this, how easy it will be automated, and what potentially process of automation you can do. You just go to their GPTs and you go like, hey, here's my workflow. You know, I want to achieve this, right? Can you tell me how can I automate this? And it's just going to spit you out literally a workflow of automation. And you can look at this, see how complex it is. You know, does it make sense? Do you want to do it yourself? Maybe you want to delegate to someone. But here you go. Like that would be your starting point. You audit your time. You get your potential tasks to automate. You chit-chat with GPTs around automation. And then you can either automate yourself. Or pass to someone else, which leads us very nicely into my, uh, another question and another quote, Mike, which is about, uh, solopreneur systems. So according to Harvard Business Review, which basically says that consistency in the tool, uh, is the single biggest factor in successful AI adoption. So effectively we're saying that, or what they're saying is that don't buy into your hype of all the AI tools out there because there's a lot of them. Yeah. But really concentrate on couple of tools and, and make good use of them on, on be— or become a, a super user of these tools. You're gonna get a lot more from this than actually just trying the newest, uh, kid on the block, as they say. Right.

Which leads me to— oh my God. Like, uh, 100% agree with that, Mike. I, Look, I think here the worst thing that you can be doing is not using any of these tools today. The fact, like, if you're frozen and you're like, well, I heard OpenAI is going to be coming out with like GPT-5 soon, so I'm going to hold off on that. And yeah, I know my company has this or whatever, like, oh my God, just don't. Like, the tools that exist today are amazing, like across the board. And I don't care if you're using like Claude or, you know, let's just talk like big, you know, foundation models. I don't care if you're using like Anthropic's Claude, like, you know, Grok by xAI, ChatGPT, Google's Gemini. They're all good. Um, pick one and go with it. That, that's the honest solution. I will say like it is so competitive. It is so cheap to use these tools.

Yeah.

Uh, Mike, if you would have given me everything that ChatGPT, just as an example, could do today, and you took that back in time 3 years ago, 4 years ago, let's say, uh, back when I was at PayPal and said, hey, how much would you pay per month per head, um, for this kind of software that could do all that? I don't know, $10,000 per person per month or something is what we would pay. Like, like, I mean, I couldn't even— I couldn't even fathom Like, like, of course it's gonna give us— this is $20 a month for these tools. Like, um, again, like, and I have no financial stake in any of this. Like, I, I don't. I intentionally do not. Like, but what I tell people is like, just pick one and go with it. Um, if you see a thing that's like, hey, well, the models say that this thing is, you know, they just launched something, it's, it's better. Like, don't just run and and like go to the next new shiny thing. Just, just stick with where you're at. If you find out like months and months later if it's not improving, whatever, then it's okay to reevaluate that. But like, all of these things are so damn good. Just, just pick whichever one works for you.

Um, I would, I would probably give some kudos to ChatGPT. I, I use— I want to say use a majority of the most popular ones, and I, I really like as well, you know, routing tools, which effectively gives an API on top of all those other tools. So effectively, you know, get your main API, but then you can just simply switch between all of these tools with your prompts and just, you know, get very quick experimentation happen and see how they respond to your prompts and your content. And I want to give an edge to ChatGPT, you know, not associated, you know, not affiliated in any way. Still doing best for at least the workloads I'm doing, which a lot, you know, a lot about the podcast, a lot about social media, a lot about content production, newsletters, and blog and stuff. A good edge. Grokex has a very good approach in indexing human forums, I want to say. So it's like, let's say this, like I'm preparing for an episode and I'm like, okay, I want to know what people are asking around this topic. And I want to really, I really want to go into like Reddit and I really want to go into Quora, which is, you know, Q&A type of forums. I just want to see what, you know, what people are sort of talking about. This does very well, does very well in a sense like here's just Reddit answers actually have good indexing. ChatGPT a little bit falls short on this and it tends to default to like blog posts and, you know, and blogs. So some differences, but yeah, ChatGPT choose one. Become better at prompting, you're going to get huge value. Just, yeah, yeah, yeah.

For 95% of the use cases out there, any of them are going to work fine and any of them are going to be just great for you out there. Look, and again, I just feel like so often people always are looking for— I remember my dad would always used to say that, like, he, like, when he got his first computer, I don't even remember when it was. Like, he was like, well, isn't it just going to be obsolete in a couple years? Um, and I was kind of— and it was always like, well, if it's always going to be obsolete, that like— like, I mean, like, it was always like his mindset was like, well— and, and I feel so many people have that. Well, like, if it's just going to be better in a month, why am I even going to do this today? Well, it's going to help you in a month. And when you get to the point where that tool is better, you are going to be so much more educated. You are going to be a better user of these things. So you'll be able to use these things more effectively. If I give 100 people, you know, Gemini or ChatGPT access and say, hey, do a task, I guarantee the people who are going to have the AI do that task most effectively are the people who use AI the most often. That's just how it is.

Yeah, yeah.

Like, these things are not superhuman yet. They cannot read minds. They can only do what you tell it. And the better you are talking and conversing and knowing what it's good at, knowing what it's not good at, the better the output's going to be. And all of that just takes you spending time experimenting and interacting with these.

Yeah, 100%. Another quote which I want to share is from Zapier's reports, which basically goes, 88% of small businesses want fewer tools, and they want the tools they use actually to just be better and what they need. There's so many tools. I mean, honestly, like, if you solopreneur or a business owner, you're using probably 10, if not 20 different apps for different things. That's number probably low. It's probably going to be like 60s, 50s, or something like that. It's a big number of different apps, right? So, and making sure these apps talk with each other so we have our good workflows is where the automation really comes in play. But in general, if someone would come to you and say like, here's one tool you need, you can forget everything else, you would make that choice without a hesitation. You would be like, excellent, you know, I just open one and everything is there. So don't debate the tools, just choose one, become much better at it. And I want to ask this question, Mike. So we touched on emails, which we all spend a lot of time with them and approaching emails and automation around emails will be a very huge win. People looking at this and solopreneurs looking at this, what else comes to mind? Like, you know, content production, like social media, I don't know, like document evaluations and things like that. Like from your perspective, like just giving some ideas for solopreneurs where they could just without much difficulty get some real wins.

Yeah. And I actually want to comment on your one tool thing as well, Mike, that you bring up, because I think that's a great point on that of like, you know, And I actually think that's the future we're moving towards. I think we're just going to have one tool at an operating system level. You're going to pop open a screen, whether it's a phone, whether it's whatever, and you're just going to see stuff. It's not going to be, oh, well, to send an email I need to go into Outlook to do that. Um, I need to analyze my whatever, I'm going to go into Excel to do that. Like, we take those things for granted, like, well, of course it's going to be like that. Like, I don't think there's going to be Microsoft Office in 10 years or whatever it might be. I think these things are going to be running behind the scenes. I think you're going to have an orchestrator at an operating system level. And I think we actually will have a future where you don't even know what tool you're using for all this stuff. And I think people are actually going to be happier with that experience. And we're going to look back to like today, 2025, and be like, yeah, I had like a dozen different like tools that I was using and all that, and that's just going to seem crazy. So I think that's the future we're going towards.

I love that future. I will choose that future like today. Just give it to me.

Sorry, sorry. Say again. Say again. I really forgot. I totally forgot it.

I would choose that future today, Mike. I love it.

Oh, yes, yes, yes.

That vision is great.

What was your actual question? Because I totally went on a sidetrack.

With that. So the question leading to this was, yeah, what ideas we can give to solopreneurs right now where we know that it would not be difficult to leverage AI and get some injection of, uh, effectiveness or totally removing the task?

Yeah, yeah, no, sorry, thank you, thank you. Um, here's what I would say, and again, I want to be as universal in how I apply these things as possible. So I'm like, what was— what is everybody dealing with. Um, for me, like, look, I would again look at the types of things that maybe, you know, we talked about before, the things that are going to be most, uh, time-consuming for you. So there's kind of doing that audit, but then there's also the other side of things of like, well, what are the ones that are going to be like the most expensive that you maybe have to outsource, or how can you even outsource more effectively? So I'm going to go through two different things with this here, different tools for each. Um, one, while AI is not a lawyer, I would use the AI to do legal things all the time. I do now. Anytime I get a contract for something, the first thing I do— first thing— it doesn't matter if it's a term sheet from a, uh, something I'm investing in, a contract for me doing something, whatever it is, First thing I do is I throw it into Claude and I say, hey Claude, what's wrong here? Anything unusual? Anything non-typical? Anything non-standard in this contract? And it'll come back and it'll say, hey, this looks a little weird, blah, blah, blah. That's a lot of what an outside counsel does. Now, if this was like a $10 million contract or something, what would I do? I actually probably would start with that, to be very clear. But then I'd probably also send it to a lawyer that I'm gonna be paying $300 an hour for it and say like, hey, can you double-check this thing to make sure? But, you know, if it's something that's lower risk, I would be doing that all the time.

Yeah.

And that's just a really easy way, uh, to use these tools for something they're quite good at. And then you can make the judgment from there. Okay, do I need to escalate this thing to my attorney who's going to charge me for every second, um, you know, and only give that attorney the really important stuff? Or is this something that I think it's fine and I'm just going to move on from that? Because look, in past days from that, what were you going to do? You're going to read through the whole thing, and you're— and if you're not a lawyer, like, what is that? What is that going to get you? So like, that's the— and I think about all those other examples. The other thing I would say is, even if you're a solopreneur, you probably have people that are doing some semblance of work for you today. Maybe you're hiring people on Upwork. Maybe you're using Mechanical Turk to do stuff or whatever it is. The big thing I would say is you need to empower those people with AI tools as well because they might not know how. If they can work more effectively, it is only going to help you in the end with that. For example, I have a PR person who connects me with awesome people like Mike and other podcasts and so forth. Because I hate reaching out to people directly. I hate doing that. For me, it's easier to have that. But one of the things that I did was I created something in NotebookLM where I uploaded every single, or whatever, up to 50 different transcripts of podcasts that I have done in the past, a bunch of my writings, all this stuff. I have this in one notebook and I've basically trained and teased it of like, hey, these are the Goals, these things Mike likes to talk about, blah, blah, blah, and all this other kind of stuff. So now my PR person who I'm paying to do this can actually be a lot more effective, a lot more targeted, a lot more poignant to put me on like really good interesting shows or connect me with interesting media outlets or whatever it is that I'm looking for and targeted with by using these tools like NotebookLM. And for folks who don't know, NotebookLM is something made by Google. It's using RAG, which is Retrieval Augmented Generation. It's a little bit of a different technology than the traditional language models where it hallucinates less, references the things in there, and it's largely for free. I'm not paying anything for NotebookLM and get these really awesome things that generate and you could share those with somebody on your team. Empower those people that are not necessarily working for you full-time, Um, you know, even if they're only working for you 10 hours a week, you're like, well, then they're going to use this tool. Well, yeah, let them. Let them be more effective with their other clients as well. They're just going to appreciate you that much more and be able to be helpful that much more with you, uh, in doing that.

I really love that. Thanks for— thanks for bringing this up, Mike. And, and just, um, to dig a little bit on Notebook LLM, it's— it has a context of what you just give to it, right? So it's not that, let's say you take Ollama as a pre-trained module and you add your information on top and you're sort of going like, here's me, and it's still going to have these huge layers of trained data which will be no relation to you. So it might just hallucinate and give wrong facts. Where Notebook LLM doesn't have that. It's literally what you give to it. That's where it's going to come, come back with the information, right?

That's exactly right.

Yeah.

I mean, at a very high level, that's right. It is relying on the things that you have given it as its source data. Yeah. So it's effectively— look, there is a language model running on the side, but effectively it is so, so much fine-tuning it on the data that you're giving it that those are the sources that it's going to be referencing. Compensing and the output. Yeah, that is going to come from it. So, so that's exactly right. It still can hallucinate, it still can get stuff wrong. You still need a human in the loop for this thing, uh, but it's a lot better, a lot more accurate. And you know, that was just one example. One, one way I always tell people, you know, usually not for solopreneurs, people in bigger businesses, like, you should record your one-on-ones, throw those into NotebookLM, and say, hey, how could I be a better boss? Or, you know, or likewise, share that thing with your employee. Your employee could then ask the NotebookLM, hey, what are things given all these transcripts? Am I doing a good job in this job? Like, what do you think? How could I be a better employee? All that leads to all these really interesting conversations.

Yeah, it's a good shout. In general, when you use AI and this is like a general notion right now, when people utilize AI, they really expect AI to figure things out, right? And they just go chit-chat and they go like, oh, why are you doing such a shitty job? Like, why are you not telling me what I want to hear? So they sort of expected to figure out, but according to MIT Sloan, which basically writes AI systems perform much, much better when they are treated as collaboration partners, not just tasks executors, right? Which effectively, where we're going is that if you hire someone, let's say you're hiring a junior team member, you wouldn't just leave him alone or her and go like, figure out, right? You, you would give some type of onboarding information. You know, here's our processes, here's our document, go read it, go learn it, and, and then, you know, let's have some sessions, let me give you a feedback. So you would coach that person to become better.

Yeah.

So if you have the same approach with your AI agents, you will have much, much better results when it comes as well to AI-powered automation or in general how you use that. Is that, is that like a good summary of what we were just saying?

100%. Like I would say you have to do this to use AI effectively. Let me give one very personal real-life example. I wrote about this a couple of years ago, been living with the benefits ever since. I decided one week that I was going to have AI give me all the meal recipes that we were going to have in our household that week. So I fed in, you know, 4 people in my household, my wife and 2 kids, myself. Everybody's got different preferences. I was the one cooking. I'm a horrible cook. I'm lazy. I don't like to use a lot of pans, like all this other stuff. Put in all of these things, long, long description into the AI.

That's it.

Okay, now I want you to give me 5 meals that I could have for the week. Um, and for the first one, actually give me, you know, shopping list and, uh, tell me what the recipe is. Great thing about cooking with AI, if nobody's ever done this before as well, is, um, when you screw things up, as I inevitably do, you could actually say, hey, um, it's now— I'm almost done and I forgot to add the butter. What should I do? You get to ask questions like that all the time. Super helpful. But here's the thing that we did that was magic. At the end of every one of these sessions, we would have the meal and I would say to everybody in my family, "Okay, 1 through 10, how'd you like it?" Then we would feed that back into the AI. This is a type of reinforcement learning with human feedback. Basically, you're reinforcing and educating the AI model with what it is. So it knows, hey, um, my daughter only thought this was a 3, or the whole family thought it was a 3. You gotta pivot, try something different here with that. Next time it's going to try something different, like, or if we were all 10s, it would do that. And so through the week, we're doing that each time, and it would give me new and different recipes and so forth. It's tweaking the algorithm. It's kind of figuring out, oh, okay, you like more of this, you like a little bit less of this, don't make it too spicy, whatever. Especially if you can even just give like a sentence reason why you gave it a 3. I forgot about that. I actually did have my family do that. That is super helpful. Again, that was just an example using in the household. Use it anywhere. Use it everywhere you do. If you say, "AI, I want you to write a marketing copy. Here's an example of what I do. Now write a press release for this new product." I do this all the time where then I would then go back into the AI and say, okay, cool, you gave me that. Here's what I actually did. Look at the outputs I had and learn from that. In Claude or any of the other ones, you can then continue that session if you want to just have one save for press releases. It's going to actually know and learn more of your style. It's going to learn what it did wrong. ChatGPT has that infinite memory now, supposedly. So like it'll just kind of learn that organically and you can even do that in a new session. But ultimately you want to be giving feedback to the models and just saying, hey, you got this right, you got this wrong. The more you give it that, the more it's going to be tailored to exactly what you want to see.

You mentioned Infinitive Memory. Like, is that like a new thing?

Yeah, yeah. They, they announced this just a couple of weeks ago. So ChatGPT or OpenAI came out and said, hey, ChatGPT has infinite memory. It's always had some level of memory.

Yeah.

But supposedly, like, it will remember everything from every conversation it's ever had about you.

Interesting.

And I think that's going to create— make the product really sticky because it's— so you can actually go in there and say, hey, what do you know about me? I did a fun thing and I actually said to it, Hey, what do you think I look like? So it's just like, give it— like, it's never seen a picture of me. So I was actually really curious and I posted this on my LinkedIn because it was like, okay, it's actually in the ballpark of like what I look like. It knew I was a white guy, knew I was in my 40s, knew I had a beard. I think just all of us Silicon Valley folks look alike. Like, I think something, but like, you know, I was able to get some reasonable facsimile of me. But like, it is actually really intriguing to just see what like having memory, remembering all of these conversations, it's just going to make for a better experience later. It's going to make the product more sticky, which is what they want. They want you to stay with ChatGPT and not go to other ones. But I think it's going to make a better experience for everybody who uses it.

Yeah, I love that. I didn't know that they did that. Infinitive memory. Need to go read on stuff. But that's, that's, that's definitely 10 things in like one week.

And this was just among like O3 Mini and, or, or O4 Mini and all these other things that they announced. This was just one of the many that they had. They tend to just throw a bunch of stuff out there like every 3 months.

Well, I want to ask this, Mike, from honest perspective. So let's say I'm not a technical solopreneur, right? So maybe I know some stuff because I'm still likely building services or products or something in that regard for online. So that means I am to some extent understanding the technologies, but I'm not technical. I'm not a coder, right? So I have no background in it. How likely is that I'm going to be automating stuff for myself with like Make.com or like Zapier? Should I be actually trying to do this? And, you know, I'm looking for your honest opinion. Or will that be actually more waste of time and I should be looking to hiring someone to do that for me?

Here's what I would say. I would say it's all baby steps with this. Like, yes, I think if you're not technical, not a coder, like, I wouldn't say like, okay, you should just integrate agents into your workflows and all that. Like, let's let's take steps here. Um, ChatGPT, Claude, any of those tools are amazing at what they can do, and they respond in natural language. You don't have to know a single— be able to write a single line of code to be able to talk with these tools. Um, I'm not an engineer. I, I, I worked in innovation and all that. My background is in finance and accounting. That's how I actually started my career. I can't code. Uh, You know, I've over the years learned to understand how technology works and things about it, but I, I can't do that either. Um, so what I would say to people who are, you know, as long as you're comfortable just talking with them, use these tools as much as you, as much as you're comfortable with. If you think like, okay, it would be good to automate it, what I would then do is I would actually talk with the tool that you're most comfortable with. Like ChatGPT and say, hey, I think I want to automate email. I heard this podcast with this Mike, these two Mike guys, and they were talking about, you know, automating stuff using Make. I don't know how to do this. You know my level of technical expertise, ChatGPT. I'm thinking about this. How, how do you think I should go about it? And you know what, that is going to give a better answer than I could ever give because it's going to be much more customized to you. Because then you could say like, ChatGPT, I tried to go into Make and even just creating an account was difficult. Here's some screenshots or whatever. And then, then maybe you just make the decision like, okay, it's not the right time, or have it help you with that. Uh, I— we are not far away. I think actually on my Pixel phone it actually has the feature now, at least within the Chrome browser, where it can see everything that you're doing within Chrome. Like, literally, it's almost like a video camera. Is if you turn this on, you effectively have like a camera running on your device at all times. And so you could converse with it. You could say like, okay, I'm on make.com, what should I be doing now? Okay, do I click over here? And like literally having a back and forth conversation. Yeah, that's going to be standard pretty much everywhere within 6 months. So even if you aren't technically savvy, as long as you have an agent that you kind of trust, for lack of a better term, uh, you're going to be able to kind of do things in a browser to have it help you automate those things as part of it. Uh, so that's where I would start with it. Could you hire somebody to do it? Potentially. But honestly, I think, uh, in many ways, the, uh, the AI tool that you're already using and comfortable with, it might even do a better job. At least it'll do a quicker job. Than that and have enough to hire somebody to do some of the more basic things at least.

That's excellent. So I think to summarize the whole episode, pay $20, get ChatGPT. We're not advertising. Maybe you want Claude, maybe you want, you know, AccessGrok, which is, I think it's free effectively, but choose one tool, use it, and you effectively start by just talking with it and do your time audit. Find the tasks which are low-hanging fruits, quick wins, which should not be done by hand. Make sure you want to automate this. Go to ChatGPT, or again another tool, have a chat. How can you automate this? Just ask it, help me to automate this. See what it says. Decide if it's too complex, if it's worth your time or not. If it is, go learn it, go automate it, go to the next level, or delegate to someone. But the biggest takeaway from this episode is that you should not be doing many things which you're doing today in and out. That should be automated. I think that's the biggest takeaway we want to give here. So Mike, if people want to get in touch with you and just, you know, thank for the episode or have any other questions, where they can find you?

Yeah, I'm on LinkedIn primarily, so you can see me on there. And if you want to send me an email directly, send me. It's just Mike Tedesco at Gmail.com. Just, just my name. Send me that. I'm always happy to help people out in any way I can share some thoughts. I live in this world, so I just want to help folks.

Excellent. We'll have a link in the show notes 100% for the LinkedIn. We will probably not add your email, Mike. If you would get—

if you made it this far in the podcast and want to send me an email, like, just let me know you made it this far.

Yeah, I'm happy I'm just thinking right now, I mean, if I'm some type of spammer, I'm probably transcribing podcast episodes just for emails.

Oh, I mean, here's one of the great things about AI. Like, spam is so much better now than it was even 10 years ago, largely because of AI running behind the scenes.

I do agree.

It's still not perfect and so forth as part of that, but, you know, and AI, and, you know, AI is going to be coming on the spammer side now, but yeah. Gmail and all that is really good for filtering stuff.

We coming to good times. The spam will be much better. One good use out of AI, the spam will come so much better. Right. Well, thanks everyone for tuning in today. And as always, if you enjoyed the show, make sure you subscribe and not miss any other future episodes. If you have any growth type of challenges, make sure you visit masteringtechgrowth.com. Share that with us. We'd love to know what you're working on, what, what keeps you up at night. And as always, until the next time, stay safe, see you online.