Latest / The ADHD Skills Lab / The AI Shift Every Founder Needs to Understand (Matt Verlaque)
Transcript
- 0:00SPEAKER_00: I think it's one of the most challenging things in business is to figure out when you just need to go one more, another week, another month, another attempt, another repetition, and when you've gone one too many and you need to actually pivot the strategy and make the change.
- 0:10SPEAKER_00: The way I try to approach it is I always assume that everything is an execution problem until I can prove otherwise.
- 0:14SPEAKER_01: Hello, everybody, and welcome to today's episode of the ADHD Skills Lab.
- 0:20SPEAKER_01: Today I am very excited to have a wonderful returning guest, Matt Verlach, to talk to us today.
- 0:27SPEAKER_01: We're gonna talk about pivoting because there's been a little bit of pivoting that's been going on.
- 0:32SPEAKER_01: But for those of you who don't know, Matt Verlach is the founder of Precision.
- 0:37SPEAKER_01: Precision helps the founders hire AI teammates over chat so they can build a bigger business with a smaller team.
- 0:47SPEAKER_01: He previously co-founded and led to launched to a successful exit in 2020.
- 0:52SPEAKER_01: He served as the CEO of SaaS Academy and co-authored the wonderful book software as a science.
- 0:59SPEAKER_01: Outside of his work, Matt is a husband and father to three mini sidekicks and spends his time weightlifting, running, and sidekick.
- 1:07SPEAKER_01: So great to have you back, Matt.
- 1:09SPEAKER_00: I'm so happy to be here with you, Sky.
- 1:11SPEAKER_00: Thank you.
- 1:11SPEAKER_01: Yeah, yeah, awesome.
- 1:13SPEAKER_01: So two weeks ago, apparently, you changed your um business.
- 1:17SPEAKER_01: I was having a look at it.
- 1:18SPEAKER_01: I'm super excited.
- 1:20SPEAKER_01: Tell me a bit of this journey.
- 1:21SPEAKER_00: For sure.
- 1:21SPEAKER_00: So it was it's a little more than two weeks ago.
- 1:23SPEAKER_01: Yeah, that's why I said apparently.
- 1:25SPEAKER_01: I was like, that's when we find it out.
- 1:27SPEAKER_00: For sure.
- 1:28SPEAKER_00: I I would say, gosh, our our reinvention of precision is probably five or six months in flight at this point.
- 1:35SPEAKER_00: It's interesting.
- 1:35SPEAKER_00: I think that every good business goes through a couple of pivots.
- 1:38SPEAKER_00: Whether or not they have the courage to call it that is you know debatable.
- 1:42SPEAKER_00: But for us, we know we I started this company with this vision of helping entrepreneurs get in touch with their numbers and their data.
- 1:50SPEAKER_00: I've always clearly, as evidenced by my hat, been a big fan of the data and knowing your numbers, because I think it just helps like emotionally overloaded, irrational humans, which is what we all are, uh, make you know better decisions and put some some logic to help you know supplement our gut reactions to things.
- 2:06SPEAKER_00: What I realized as we were building that product and that business is that there's this funny principle in marketing called chocolate broccoli, right?
- 2:15SPEAKER_00: Because everybody wants chocolate, but everybody needs to go eat their broccoli.
- 2:18SPEAKER_00: And what I found in the first year of business, even though we had some success, we delivered some good value for clients and whatnot with our software, it was still firmly in the broccoli camp.
- 2:26SPEAKER_00: Like it was a thing that they knew that they needed, but it wasn't the thing that was going to get them in the door with enough velocity for me to have the type of impact that I wanted to have.
- 2:36SPEAKER_00: So I was wrestling with that.
- 2:38SPEAKER_00: And at the same time, this whole concept of AI agents started to come about.
- 2:42SPEAKER_00: And, you know, if you if you carry the arc of data from the spreadsheet forward, you know, if you step into the pre-AI world, essentially founders would gather their data from the different sources, they'd put it in the spreadsheet, they'd analyze the numbers, they'd figure out the story that the data was telling them, they'd make some decisions around how they were going to process that information, and then they would go get the work done.
- 3:02SPEAKER_00: And so, like the initial thought was well, we can help them with the insights part.
- 3:05SPEAKER_00: We'll help them do the analysis.
- 3:07SPEAKER_00: And quickly, I feel as though the software expectations in the market quickly translated from I no longer want to be shown insights, I actually need the software to help me do the work.
- 3:17SPEAKER_00: And so, not to be left behind, that's where we started to put our software platform.
- 3:21SPEAKER_00: And so we still have all of our data fundamentals.
- 3:24SPEAKER_00: We still plug into people's tools and help them understand their numbers.
- 3:26SPEAKER_00: That's the broccoli that they get once they're here.
- 3:28SPEAKER_00: But the thing that we do with that is then we can use that data and that context that we build to help people hire AI team members that are smarter than just firing up ChatGPT and saying, like, you know, make me$100,000, make no mistakes, kind of stuff.
- 3:40SPEAKER_00: You know, we can actually define the workflows, get really, really specific on what these AI teammates should be doing and not doing and what good looks like, and be able to automate a lot of the like repeatable back office work that tends to stifle the creative instincts of, you know, visionary founders.
- 3:56SPEAKER_00: So that's where we're at.
- 3:57SPEAKER_00: And it's been a lot of fun.
- 3:58SPEAKER_00: I think the business has almost doubled since we made that pivot.
- 4:01SPEAKER_00: And then when we were actually able to acquire SaaS Academy and bring it into precision, so that's been super fun.
- 4:06SPEAKER_00: So there's been a lot of stuff happening really, really fast.
- 4:08SPEAKER_00: But what I can tell you is that people are eager to get some help, and whether digital, human, or otherwise.
- 4:12SPEAKER_01: That is a really interesting, and I think you are right about the pivot.
- 4:15SPEAKER_01: It is a really important thing for business owners to know for two reasons.
- 4:23SPEAKER_01: One, it stops people from feeling a sense of shame when they do pivot and and seeing it as a general trajectory, but also so they don't spend too much time starting because they're waiting for the perfect business.
- 4:35SPEAKER_00: Yeah, 100%.
- 4:36SPEAKER_00: It's tough.
- 4:36SPEAKER_00: I think it's one of the most challenging things in business is to figure out when you just need to go one more, another week, another month, another attempt, another repetition, and when you've gone one too many and you need to actually pivot the strategy and and make the change.
- 4:49SPEAKER_00: The way I I try to approach it is I always assume that everything is an execution problem until I can prove otherwise, which makes me sometimes difficult to work with, but that's okay.
- 4:58SPEAKER_00: Because it most often it is, right?
- 5:00SPEAKER_00: Like most often, especially for early stage businesses, say anything south of five million in revenue, even if your strategy is a little bit off the mark, you can still likely make a few million dollars in revenue and then figure out what needs to happen.
- 5:13SPEAKER_00: You know, like we we were north of a million in revenue before we made the pivot, and it, you know, it was a wonderful business and it, you know, grew fast and all that kind of stuff.
- 5:22SPEAKER_00: And I was eternally grateful for all of that.
- 5:24SPEAKER_00: And at the same time, I could still see what the data was telling us is that it was going to be tricky to scale and difficult to scale and would, you know, get capped out a little bit earlier.
- 5:32SPEAKER_00: So I think it's one of those things where often entrepreneurs can pivot too early because they haven't sat with the pain of the problem long enough.
- 5:41SPEAKER_00: But at the same time, some of us will sit with it much too long because we're gluttons for punishment.
- 5:46SPEAKER_00: I am in that category.
- 5:48SPEAKER_00: And um, and I think sometimes you need to know when you need to when you need to tweak it.
- 5:51SPEAKER_01: Yeah, 100%.
- 5:53SPEAKER_01: If you are the business owner who is growing, your operations and delivery are growing, and you're starting to feel pretty overwhelmed, like you're not getting enough time, you're struggling with consistency, and you'd really like somebody else to get it all out of your head and project manage it for you.
- 6:08SPEAKER_01: That's what we do.
- 6:09SPEAKER_01: We are fractional operators for your business designed for ADHD symptoms.
- 6:14SPEAKER_01: We take it out of your head, we set it all up for you, and you just kind of watch.
- 6:18SPEAKER_01: It's pretty fun.
- 6:19SPEAKER_01: Click the links down below.
- 6:21SPEAKER_01: And if you're interested, you can book a call with me.
- 6:24SPEAKER_01: You know, when we're talking about founders with ADHD traits, one of the types that you can find is the pivot chaser, the person who's always chasing the next pivot and kind of going in that direction.
- 6:35SPEAKER_01: But actually, also when you look at the research into ADHD traits, that's one of the strengths is that we can move on a little bit faster.
- 6:43SPEAKER_01: And especially in the entrepreneurship world, that is such a delicate balance to hold.
- 6:49SPEAKER_01: Do you have any sort of tried and true methods of looking at somebody and getting a sense of if they're being a pivot chaser or not?
- 6:56SPEAKER_00: Yes, there is a very common behavior that I've found in a lot of the founders that I've worked with.
- 7:02SPEAKER_00: And it usually shows up when your growth is stalled and you're not exactly sure why your growth is stalled and what you should do about it.
- 7:10SPEAKER_00: And so the interesting thing, I don't mean this if your growth is stalled at zero or you have like, you know, two or three customers, but we if your business has started to achieve even a small level of scale, call it 500K, a million up, and you had this march towards growth and then you capped out, very often it is the founder being, as you called it, a pivot chaser, right?
- 7:29SPEAKER_00: I call it, you know, founder shiny object, you know, squirrel chasing uh tendencies, where where essentially we get so used to this novelty that as soon as the novelty goes away, we feel like we need to go find something novel to make us feel that same feeling.
- 7:45SPEAKER_00: And I where it shows up most often, skies, in marketing channels, where I will have someone and they'll say, Oh, you know, I was doing all these partnerships and then all of a sudden I stopped growing.
- 7:55SPEAKER_00: And they'll say, Cool, what are you doing for growth right now?
- 7:57SPEAKER_00: I'm running Facebook ads.
- 7:58SPEAKER_00: Okay.
- 7:59SPEAKER_00: Uh, what happened to your partnership program?
- 8:01SPEAKER_00: Oh, well, I didn't think it was gonna scale.
- 8:02SPEAKER_00: I needed something to be like more scalable, and you know, ads seem to work really, really well because I watched this webinar, so I started ads.
- 8:09SPEAKER_00: And so then I have to look at them and say, So what you're telling me is that you did thing A and it worked, and you grew up until the ceiling and you stopped growing when you stopped doing thing A and you went to go do thing B.
- 8:20SPEAKER_00: And when you say it like that, everyone's like, damn, yeah, I guess that doesn't make a lot of sense.
- 8:24SPEAKER_00: The the trouble with it is that I think that we feel good about skill acquisition.
- 8:29SPEAKER_00: I know I do personally, where like if I can learn a new skill, it makes me feel useful, it makes me feel like I have a an increased level of utility and usefulness for the world, and I'm like, oh, I did the thing.
- 8:38SPEAKER_00: But if we're constantly pivoting the things that we do, we are essentially enrolling ourselves to permanently be a beginner at everything.
- 8:44SPEAKER_00: And beginners don't scale companies.
- 8:46SPEAKER_00: So the the tough, the tougher thing to figure out is how can you reframe that concept of novelty and try to like get that same feeling from compounding growth where you don't change the inputs, you just kind of feel that feeling from the outputs and the results of what you're doing.
- 9:01SPEAKER_00: And if you can sit with the strategy that was working longer, typically there's there's better results there.
- 9:05SPEAKER_00: So I think that making sure that you're kind of plugged into the thing that's working and causing the growth, and that at all costs you don't turn that thing off in search of something else is really, really important.
- 9:16SPEAKER_00: So that's generally my thoughts on that.
- 9:18SPEAKER_01: Yeah.
- 9:19SPEAKER_01: And and you know, that's usually where we will say, okay, it's time for you to plug a person in, somebody who's going to be able to hold the consistency while you go seek the new novelty.
- 9:31SPEAKER_01: But you work with AI.
- 9:33SPEAKER_01: So tell me about the AI teammates that you could be adding into this space.
- 9:38SPEAKER_00: Yeah, the the thing that the thing that is counterintuitive to a lot of people, like I am a I am not the average AI nerd.
- 9:47SPEAKER_00: Like most people, when you go on the internet, and it's funny, but even before we started recording, you were talking about this with the the YouTube videos that I've been putting out.
- 9:53SPEAKER_00: Like most people you go on the internet and they're like, AI is this magical technology and it can do everything for you, and you don't need any people, and you can have a one-person billion-dollar business, and it's this utopia.
- 10:02SPEAKER_00: And I hate to be the bearer of bad news, but at least as of the time we are recording this, that world does not exist.
- 10:10SPEAKER_00: Only it only exists in the world of people who generally want your money, so they can try to like teach you a course on how to do it.
- 10:16SPEAKER_00: So all the AI bros may come for me, but I don't care.
- 10:20SPEAKER_00: The the reality is that it is a really solid piece of technology.
- 10:25SPEAKER_00: It is still a piece of technology.
- 10:27SPEAKER_00: And the best way that I rationalize and teach people how to use AI in the context of actually building an AI teammate is that many of the behaviors that you would use to train and lead a human teammate correlate very, very directly into building an AI teammate.
- 10:40SPEAKER_00: So if I were to break down the building blocks of an AI agent, essentially the way I would define it is that it is an autonomous workflow that has a trigger, meaning that there is some way it knows that it needs to go do its work, that is not Matt or Sky sitting at the keyboard telling it to go do its work or something else that's like the precipitating event for it to go do the work.
- 10:58SPEAKER_00: It does the work, it has a some type of playbook.
- 11:01SPEAKER_00: In AI land we call this a skill, but some type of playbook or process that it's following.
- 11:05SPEAKER_00: And some of those steps require judgment, meaning that it's not just, you know, a code or integration or two plus two equals four.
- 11:12SPEAKER_00: Like it needs to actually make a judgment call, synthesize information, analyze information, do what's called the probabilistic work.
- 11:17SPEAKER_00: And so it needs to do that, it needs to produce an output.
- 11:19SPEAKER_00: And there's typically some type of grading mechanism so it can be self-improving, which is called an eval.
- 11:23SPEAKER_00: Essentially, you like look at what it produces and grade it against a set of outcomes so that way you can make itself better.
- 11:29SPEAKER_00: So it's self-improving.
- 11:30SPEAKER_00: So if it's autonomous where it has a trigger, it's doing a bit of work that has some judgment calls to be made, it produces an output and it makes itself better.
- 11:38SPEAKER_00: At that point, you have an AI agent or an AI teammate.
- 11:40SPEAKER_00: So when you break that down, all that fancy words, nerd stuff that I just said, really, if you can build it so it has a name, a boss to make sure it's doing a good job, a schedule so you know when it should be doing its work, and a job description so it knows like why it exists in the first place.
- 11:57SPEAKER_00: The same things we would give to a human being when we hire them, you're generally going to be taking most of the boxes that you need to actually build and train an AI employee the right way.
- 12:04SPEAKER_01: No, I love that.
- 12:05SPEAKER_01: I love that.
- 12:06SPEAKER_01: So here's where it gets interesting.
- 12:07SPEAKER_01: So I am a co-founder COO, which means that I have a systemizer, former programmer husband who now programs systems and AI, et cetera.
- 12:20SPEAKER_01: That's his job.
- 12:21SPEAKER_01: My job is to talk about ADHD, culture, HR, psychology.
- 12:26SPEAKER_01: That's the piece that I come in.
- 12:28SPEAKER_01: And to talk to the people who hate technology, right?
- 12:32SPEAKER_01: Pieces of paper, post-it notes, that's what we're working from.
- 12:36SPEAKER_01: So can you give me a sentence?
- 12:38SPEAKER_01: So you just said the sentence, and I know that all the software people, those people are nodding along.
- 12:43SPEAKER_01: That makes sense.
- 12:44SPEAKER_01: Can you give me a sentence that says that for the people who are like, I'm just getting used to the computer?
- 12:50SPEAKER_00: Yeah, easily.
- 12:51SPEAKER_00: Imagine for a minute a piece of work that you hate doing.
- 12:55SPEAKER_00: Like if you're listening to this, just take a 10-second pause and picture the thing.
- 12:58SPEAKER_00: It's some data entry, some I need to move this over here, some software that angers you every time you open it.
- 13:04SPEAKER_00: There's everyone's got that work.
- 13:06SPEAKER_00: Right.
- 13:06SPEAKER_00: So the the world that I try to create for people is you can take out your phone and open up Telegram or Slack and talk to an AI tool that we've built and describe the work that you hate doing, and that AI tool can go build you another little bit of AI that will then do that work for you.
- 13:24SPEAKER_00: And then you can see what it's doing and you can grade it and tell if it's doing a good job, and then you can stop doing that work and the technological minions can do the work for you.
- 13:33SPEAKER_00: It's as simple as that.
- 13:34SPEAKER_00: If you can describe the work and the work occurs in the computer, ideally between some pieces of software that we can connect to, you can just use your words to describe what's happening and make the work largely go away.
- 13:45SPEAKER_00: That's the easiest way to think about it.
- 13:46SPEAKER_01: Yeah, and I love that.
- 13:47SPEAKER_01: I love that.
- 13:48SPEAKER_01: And yeah, we we do that for people a lot.
- 13:50SPEAKER_01: We spend a lot of time improving these kinds of systems because it is it is so interesting.
- 13:55SPEAKER_01: So, in terms of how you're using it then, are you plugging these people into WhatsApp, into Slack, these AI agents?
- 14:04SPEAKER_01: Like, where are you communicating most of the time when you're working with a client?
- 14:08SPEAKER_00: Yeah, so we uh we have a philosophy the way that we build AI.
- 14:11SPEAKER_00: So a lot of people, when they try to lean into these AI teammates, essentially they end up with like 15 or 20 different AI agents that they then need to like chat with and figure out who you're which one you're supposed to talk to.
- 14:26SPEAKER_00: I I have three young kids and I felt like the human being in the paw patrol.
- 14:30SPEAKER_00: If you've ever watched that, where like I had all these different like scout and rover and whatever the heck their names are.
- 14:35SPEAKER_00: I'm like, oh my God, who do I need to talk to?
- 14:36SPEAKER_00: And that got really confusing for me.
- 14:38SPEAKER_00: So when we built this into precision, the philosophy that we take is one front door and many rooms, if it were a house.
- 14:45SPEAKER_00: And so essentially, what we build, we have a single AI agent, I named her Claudia, and every one of our clients, the answer to everything is ask Claudia.
- 14:54SPEAKER_00: Talk to Claudia, ask Claudia all the time.
- 14:56SPEAKER_00: So you can talk to Claudia in our web app, you can talk to Claudia in Slack, you can talk to Claudia in Telegram, you can text Claudia if you want to.
- 15:04SPEAKER_00: The answer to everything, Claudia is the keeper of the world.
- 15:06SPEAKER_00: She has all the answers.
- 15:07SPEAKER_00: And so that's always the front door.
- 15:10SPEAKER_00: So that's always where you talk to your AI team when we build it.
- 15:14SPEAKER_00: And then Claudia is essentially the router.
- 15:16SPEAKER_00: So she knows the numbers, she knows the overall context of the business, she knows the team, she knows the work that needs to get done, and she can keep track of the dozens of other AI teammates that she has built for the particular client's business, went just simply by chatting with her.
- 15:29SPEAKER_00: So that way we kind of remove the cognitive load from the client of being like, well, which one of these 46 different little pretend teammates do I need to ask?
- 15:38SPEAKER_00: Like, answer is always the same.
- 15:39SPEAKER_00: Talk to Claudia, talk what's going on, got it covered.
- 15:42SPEAKER_00: Like, simple as we can.
- 15:43SPEAKER_01: So, how do you make sure that Chloe, because some of these teammates will have different levels of access to different levels of data?
- 15:50SPEAKER_01: How do you keep the security when you're talking to Claudia?
- 15:53SPEAKER_00: It's a really good question.
- 15:54SPEAKER_00: We spent months and months just on that one problem inside of precision because it's so important.
- 15:58SPEAKER_00: Like, we're built for teams in particular, which is different than a lot of people.
- 16:02SPEAKER_00: What's one of the most common situations people come to us with is they're like, I have a team, everyone's using Cloud, everyone's plugging into everything, I have no idea what's happening, I have no idea where my data is and what's going on.
- 16:13SPEAKER_00: So essentially what we do is every you add your entire team into our software.
- 16:18SPEAKER_00: And then every single team member, they can only integrate things that they would otherwise have access to.
- 16:24SPEAKER_00: And if you integrate something, you can choose who you share it with.
- 16:27SPEAKER_00: So that way, every team member actually gets their own version of Claudia.
- 16:31SPEAKER_00: So, like if you and I, if I worked for you, Sky, like you would have your version of Claudia that probably had access to everything.
- 16:36SPEAKER_00: And you know, if I'm working in marketing, I probably don't have access to your QuickBooks to know what you know your company's net profit is, but you would have access to that.
- 16:43SPEAKER_00: And so everything that we build, if you build other teammates in your account, they can be private just for you.
- 16:49SPEAKER_00: You can tell people that you have them, but not let them see what they do, or you can hide them completely, or you can build ones that are company wide.
- 16:56SPEAKER_00: So you basically get to choose what you do with the access that you have, it's called role-based access controls.
- 17:00SPEAKER_00: And it's one of the toughest things to figure out because, like, gosh, I mean, if you're you're running a 50-person team and you got to do some layoffs and you're chatting with Claudia and then someone else is like, Oh, what's up, Jimmy?
- 17:10SPEAKER_00: Do you know you're on the chopping block in two weeks?
- 17:11SPEAKER_00: Like that would be a horrible situation to have.
- 17:14SPEAKER_00: So we put a ton of work into compartmentalization for that exact reason because it's tough to do without a software platform.
- 17:19SPEAKER_01: Yeah, I would say that's probably like a big thing for you guys because when I'm talking to people who are building these agents, they do end up with like five different agents that you're talking to this person and that person and that person, and you do forget what people have access to.
- 17:35SPEAKER_01: But on the other side, like having that many rooms and giving everybody a version of Claudia that's person specific.
- 17:42SPEAKER_01: I mean, you got to have a hefty database somewhere doing a lot of that stuff because you do not want to make those mistakes.
- 17:48SPEAKER_00: Yeah.
- 17:49SPEAKER_00: That's uh that's the advantage of a pivot going back to what we talked about earlier, is like all of the data infrastructure that we built in the scorecard era of precision, that's what lets us do what we're doing right now.
- 18:00SPEAKER_00: Like we have like varsity level data warehousing, which is you know, nerd stuff, uh basically like making copies and organizing all your data a certain way so we can do that without like bombarding all the software tools that you use.
- 18:12SPEAKER_00: It's usually a level of sophistication that most small businesses don't have access to because it just doesn't make sense for them to invest in it.
- 18:18SPEAKER_00: But that's the that was the first thing our software ever did.
- 18:21SPEAKER_00: So yeah, it's kind of neat.
- 18:22SPEAKER_01: Yeah, that's really interesting.
- 18:24SPEAKER_01: And you know, last time we spoke, we talked about ADHD.
- 18:26SPEAKER_01: We talked about, you know, how entrepreneurs are more likely to get diagnosed with ADHD or have ADHD traits.
- 18:32SPEAKER_01: These days we talk more about traits because so many people are just going, yeah, I relate, but you know, I'm not getting diagnosed.
- 18:38SPEAKER_01: And so have you, you know, since we've spoken about it, has have you noticed the kind of person who comes to you, the kind of struggles that people have that feel like ADHD related?
- 18:51SPEAKER_00: I think so.
- 18:51SPEAKER_00: I'm not an expert in it like you are, so I can tell you what I'm observing, and then maybe you can tell me, because I don't like to play the expert if I'm not one, but I can tell you there's there's really two, I would say, primary archetypes for clients that we work with.
- 19:04SPEAKER_00: It doesn't mean like all founders don't fall into one of these two buckets, but the ones who work with us typically fall into one of two buckets.
- 19:10SPEAKER_00: So they're either somewhat technology averse, and there's a large helping of like FOMO, of fear of missing out, where they're like, I know I need to lean into AI, I don't really know how it works.
- 19:21SPEAKER_00: One of the things that I love, and I I mean you remember this from when you and I work together, is that like we're we're still a very people forward business.
- 19:28SPEAKER_00: Like what we still coach founders, we coach them through the transformation on how to become an AI-first organization.
- 19:34SPEAKER_00: We do a lot of things that like don't scale if you're a software company, and we've built our business model around that, and I'm proud of it, and we're never going to change.
- 19:41SPEAKER_00: So I think for that, that particular kind of persona who's like, I don't quite know what I need, but I know I need this, but I don't know how it works, and I could really use some expertise and some help to make this transformation.
- 19:52SPEAKER_00: We love working with people like that because we can really help them.
- 19:55SPEAKER_00: It's fun because we're all software people, and in today's day and age, every company is a technology company.
- 20:01SPEAKER_00: It doesn't matter what industry you're in.
- 20:02SPEAKER_00: So that's kind of archetype one.
- 20:04SPEAKER_00: And the the second one, which is more interesting in a lot of ways, I call it the like the AI graveyard crew, where it's like they have tried and tried and tried and built and built and built and shipped all of these different things.
- 20:19SPEAKER_00: They got open claws over all over the place and Hermes all over the place and subscriptions to everything, and this talking to that, talking to that, and this spreadsheet, like so many things.
- 20:30SPEAKER_00: And they realized that for that particular persona, the barrier is not can I build this?
- 20:36SPEAKER_00: The barrier is is there a return on time and return on investment for me to build this instead of growing my company?
- 20:42SPEAKER_00: Which is actually where most of these founders should be spending their time.
- 20:46SPEAKER_00: So it's it goes back to the the chasing novelty thing we were talking about earlier, where there like there has never been a better tap of dopamine that you can turn on than building something with AI.
- 20:59SPEAKER_00: And if you follow the incentives, it's obvious why.
- 21:02SPEAKER_00: The AI companies want you to spend money.
- 21:04SPEAKER_00: Spending money means using tokens, using tokens means chatting.
- 21:07SPEAKER_00: So they tell you that you're a genius and they do whatever you say.
- 21:11SPEAKER_00: And boy, oh boy, do you feel smart?
- 21:13SPEAKER_00: And it's tough.
- 21:13SPEAKER_00: And so a lot of the time, our some of our best clients, our fate, my favorite people, have come into this and they're like, I could probably build a lot of this, and I makes me feel good when I do it, but I also realize that I spent the last six months messing around with AI and my business got smaller and it didn't make me money, so maybe I should stop.
- 21:28SPEAKER_00: And I'm like, Yes, I agree.
- 21:31SPEAKER_00: Let us do it for you.
- 21:32SPEAKER_00: So those are kind of the two buckets that we work with.
- 21:34SPEAKER_01: No, I really appreciate that because it's been an interesting shift.
- 21:37SPEAKER_01: You know, obviously, everybody's been moving into AI, and I will say that sometimes the hardest clients we have are the ones who can go and build it themselves because they'll go away and they'll create a clawed skill.
- 21:49SPEAKER_01: And now we have a dashboard that looks really good but doesn't connect to anything.
- 21:55SPEAKER_01: It's very specific, it's not necessarily built on what the team needs, it's more of Like a like a late night fever dream of what I'm saying.
- 22:03SPEAKER_00: Yes, that's a what a beautiful way to put it.
- 22:05SPEAKER_00: That's exactly right.
- 22:06SPEAKER_00: I've had a few of those late night fever dreams myself when it comes to building stuff.
- 22:11SPEAKER_00: It's it's a real thing.
- 22:13SPEAKER_00: It's like, I mean, there are people who even they'll call it AI mania, where it's like it's like, I just gotta like ship this one thing.
- 22:18SPEAKER_00: And people have like actual anxiety when their AI tool is not cooking while they're sleeping.
- 22:23SPEAKER_00: Don't ask me if I know that firsthand.
- 22:25SPEAKER_00: I will refuse to answer.
- 22:26SPEAKER_00: It's a real thing.
- 22:27SPEAKER_00: I mean, I've been through it myself.
- 22:28SPEAKER_00: Like, gosh, I mean, I I flew a little close to the sun on burnout at one point earlier this year, and I was bringing my computer to bed because I was trying to ship some stuff and I would be sitting there and I'd like wake up at one o'clock in the morning and like fire a couple prompts in there and then tilt the screen down and put it down to the side of the bed so I couldn't see the glow, and then I'd sleep for another few hours.
- 22:43SPEAKER_00: Like, I'm not immune to it either.
- 22:45SPEAKER_00: It's it's wild when you can build anything, but then it just shifts the the decision to not can I build this, it's it's should I build this and should I spend the time?
- 22:53SPEAKER_00: And that's that's the harder one to figure out.
- 22:55SPEAKER_01: Yeah, no, I a hundred percent agree.
- 22:57SPEAKER_01: You know, one of the things that we've found is that AI and social media are starting to become a bit similar in that way, you know, attention taking, you mentioned it as well.
- 23:05SPEAKER_01: Like you use credits, so they want you to spend more.
- 23:08SPEAKER_01: So there's a lot of conversations, a lot of sense of like we're overdoing it.
- 23:12SPEAKER_01: Like one of the things, you know, we talk a lot in our business in the internal about AI slop because a lot of what happens is there is this creation, especially if you're in um working in companies like we are that are using AI, and not all, you know, I'm not speaking to any specific clients or anybody specific here, but like there is this creation of AI slop that then has to be sort of thought back in order to like end up with the stuff that is useful, or you want to, you know, what we actually want to do is just rebuild it so that that doesn't happen.
- 23:49SPEAKER_01: But a lot of times the process of creation is uh is not a useful process.
- 23:54SPEAKER_01: It's the best way I can explain it.
- 23:56SPEAKER_00: I have a friend who coined this term.
- 23:59SPEAKER_00: He said for for people that just get a bunch of AI slap and like just pollute everyone else and just hand it to like he give you these long 67-page reports that are just like blocks of prose that have all these weird words in it.
- 24:09SPEAKER_00: And he says, You're being a meat proxy.
- 24:11SPEAKER_00: And it's essentially like we are we're the we're the proxy made of meat for these AI tools, just like passing along their words.
- 24:17SPEAKER_00: I know, like I know you're big on culture, and so am I.
- 24:19SPEAKER_00: It's it's one of the most important, I think, cultural things to run defense against is that exact behavior.
- 24:25SPEAKER_00: I know in our company, I gosh, I posted this last week that just like a reminder that you, the human, are the owner and accountable party for anything that any AI that you supervise puts forth on your behalf.
- 24:36SPEAKER_00: So, like, we don't blame the AI, we don't go to the AI, we go to the human who's in charge of the AI.
- 24:42SPEAKER_00: It's even in precision, in the software platform, we have an org chart where the AI shows up, the AI agents show up under the human who's accountable for them.
- 24:48SPEAKER_00: So you always know what meat proxy you need to go to to go solve the problem because we can't, yeah.
- 24:54SPEAKER_00: It just, it's it's almost disrespectful to be right because it's like saying, I, as the author of this thing, I'm not gonna take the time to edit it.
- 25:01SPEAKER_00: So I'm gonna make all of my individual readers parse through all this trash, and they each individually have to edit it to extract the meaning.
- 25:08SPEAKER_00: It's wildly inefficient.
- 25:09SPEAKER_00: So it's a it's a really important thing.
- 25:10SPEAKER_01: Yeah, it's and it's interesting because I think it's it's something that we're learning in real time.
- 25:15SPEAKER_01: This is such a fast-moving industry, and there was a moment where everybody just went, Great, the AI will make it, and then everyone will read it.
- 25:21SPEAKER_01: And then for a while, a little while, everyone was like, Oh, M-dashes are a bit of a thing, and now we can just feel this AI slop, we can sense it in every piece of writing, especially.
- 25:34SPEAKER_01: And so with my with the operational side of things, what happens there is that you know, a lot of the time we're coming in to reduce the time that the founder is spending on the business itself.
- 25:47SPEAKER_01: And then what's happening is that people who are lower down in the business are sending all of this, you know, to being the meat proxy, if you will, up the chain.
- 25:56SPEAKER_01: And so people whose hours are even more expensive are having to pass through the AI to get the information.
- 26:04SPEAKER_01: So it's only really making it faster for the person who's doing it, but that person isn't always the one who's like you know, the one whose time we're trying to reduce.
- 26:13SPEAKER_00: For sure.
- 26:14SPEAKER_00: Yeah, it's the exact opposite of what we're trying to reduce really.
- 26:17SPEAKER_01: Yeah, which is funny because you know, we also work like I I have a community now of operators and and project managers, and we meet once a week and I train them.
- 26:26SPEAKER_01: And one of the biggest concerns they have is people are gonna take, you know, AI is gonna take their jobs.
- 26:31SPEAKER_01: I'm like, well, not if you speed run it by handing something over that's not you didn't look at, that's probably true, but you know, these are important levels.
- 26:41SPEAKER_01: So I really love your idea of we blame you, not the AI.
- 26:44SPEAKER_01: I think that's really great.
- 26:46SPEAKER_00: Yeah, yeah, it rolls up to so there's an article that I read.
- 26:49SPEAKER_00: There's this company, every.com, and the the founder's name is Dan Shipper, which by the way, what a great name for a CEO of Shippers.
- 26:55SPEAKER_00: Just ships everything.
- 26:56SPEAKER_00: I love it.
- 26:57SPEAKER_00: But um they they are very, very AI forward.
- 27:00SPEAKER_00: They run a portfolio of of SaaS products.
- 27:03SPEAKER_00: And what they found, you know, they were one of the earliest adopters of like like hardcore AI-driven software development.
- 27:09SPEAKER_00: There's like like maximalists.
- 27:11SPEAKER_00: What they found is that their throughput increased by 100x.
- 27:14SPEAKER_00: Sure, they can go a lot faster, but their human needs actually also increased with the additional AI that they were using, because like adopting AI at that level is still going to create human jobs because humans are still required to provide oversight over the AI to make sure that the quality is of a certain threshold.
- 27:30SPEAKER_00: So I really like even on our own website, we say supercharge your humans, not replace your humans on purpose, because I don't think that we should be part and parcel just replacing the human beings and striving for this, you know, anti-human like workplace, because it also sounds like a depressing environment to participate in.
- 27:45SPEAKER_00: But I think that if you're worried about AI replacing you, I would look at it at a skill level and not necessarily like a uh existential labor type level where you may need to change the skills that you have and be more in an oversight and an editor in an orchestration role.
- 28:00SPEAKER_00: If you can do that, you will likely vastly increase your earning power as an employee because all of a sudden you can then be the supervisor of dozens of AI agents that are likely doing the work you used to do.
- 28:12SPEAKER_00: If you can cross that chasm, like your earning power will be through the roof relative to just being an individual contributor today.
- 28:18SPEAKER_00: It's just uh it's a different way of looking at it.
- 28:19SPEAKER_01: Yeah.
- 28:20SPEAKER_01: If you've been telling yourself for years now that you're gonna get organized, you're gonna get a sense of relief, a sense of clarity in your business, but you just can't figure out where you need to start to get that done.
- 28:32SPEAKER_01: I want to give you that information for free.
- 28:34SPEAKER_01: Go ahead and click the link down below.
- 28:36SPEAKER_01: So two minutes of questions about your business, and within the next 24 hours, I'm gonna send you an audio note explaining what your business struggles are and how they make sense with your ADHD and a step-by-step process for you and your team to get organized and get unstuck.
- 28:51SPEAKER_01: It's free.
- 28:52SPEAKER_01: I just want to help you out.
- 28:53SPEAKER_01: So click the link down below and uh let's get chatting.
- 28:57SPEAKER_01: What do you think is like one of the things that people misunderstand the most about working with you and working with these AI agents?
- 29:06SPEAKER_00: Like, forget working with me.
- 29:08SPEAKER_00: The the biggest misconception that I think people have about AI, I'll give you two because they're very tightly linked to one another.
- 29:15SPEAKER_00: The first one is that it is a magic potion that can do anything you want.
- 29:21SPEAKER_00: So the the part of what we do even in our sales process is I have to talk people out of the things they're asking us for.
- 29:27SPEAKER_00: Because like we can you can do a lot of stuff with the the type of you know software platform that we've built.
- 29:32SPEAKER_00: But I mean, there are people who are like, I just need some AI to like CEO the business for me.
- 29:36SPEAKER_00: Can you make that?
- 29:37SPEAKER_00: Like, no, not well, not unless you hate your business.
- 29:40SPEAKER_00: Then sure, like whatever.
- 29:41SPEAKER_00: I'll just, you know, sell it to me.
- 29:43SPEAKER_00: So I think one is just like scoping down the things that they think they want and like sprinkling some reality on it.
- 29:48SPEAKER_00: The second one, though, is a little bit more nuanced, specifically for people who are already trying to solve this problem.
- 29:53SPEAKER_00: There is a very short distance between not having it and having it work once.
- 29:59SPEAKER_00: And there is a very long distance between having it work once and having it work every time.
- 30:04SPEAKER_00: And that gap between it worked once, here's my dashboard, here's my prompt, here's my article, here's my YouTube script, here's my thing, that working once, work twice, work three times, that is so far removed from something that can actually like carry the responsibility of revenue generation in a professionally run business.
- 30:24SPEAKER_00: There's so much that goes from here to there.
- 30:26SPEAKER_00: And I think that people sometimes don't understand that until they've been down that journey of trying to make something that is bulletproof inside of a technology business.
- 30:35SPEAKER_00: Like having it work once is not that impressive.
- 30:39SPEAKER_00: It's very easy to do because you've only tested it under a very finite set of circumstances.
- 30:44SPEAKER_00: You haven't yet introduced the variables that will actually make it fail.
- 30:47SPEAKER_01: Yeah, no, I agree with that.
- 30:48SPEAKER_01: I really like that because so often I think, you know, we talked about that like late-night fever dream kind of shipping process.
- 30:55SPEAKER_01: It's about getting to a place where something works once, but there is a lot of systems that are in place to work multiple times and then also work without you, like as the founder.
- 31:07SPEAKER_00: Yeah, and then and the next level to that to actually keep it working without you is if you if you think about, if you think about the work that we do as humans, right?
- 31:14SPEAKER_00: We're interacting and interfacing with external systems all day long, right?
- 31:18SPEAKER_00: If you log into QuickBooks to go check your revenue and QuickBooks changes the menu, like that's not going to trip up sky.
- 31:25SPEAKER_00: You'll be able to find the button unless they did a really bad job with the menu.
- 31:28SPEAKER_00: Like, maybe not.
- 31:29SPEAKER_00: You get the idea though, right?
- 31:30SPEAKER_00: Like humans are relatively adaptable.
- 31:32SPEAKER_00: And so the the thing that like part of what's in that massive distance between worked once and works forever is that self-learning loop and the ability for the AI that you've built to actually recognize when circumstances and inputs have changed, figure out what changed, self-correct, and then lock in the lesson so you don't have to keep learning the lesson every time.
- 31:50SPEAKER_00: Like if it's not self-correcting, it's not durable because humans are self-correcting.
- 31:53SPEAKER_01: So who do you think is at risk with an AI world that we're living in now?
- 31:59SPEAKER_01: Like who are the people who should be worried about their jobs?
- 32:03SPEAKER_00: People who are doing there's like there's an inverse relationship between the number of variables that you have to deal with in your work and how at risk you are.
- 32:10SPEAKER_00: So if you are doing data entry that is the same every time, right?
- 32:14SPEAKER_00: And like there's like government jobs like this, there's some jobs like this in larger corporations where you've got to move the letters and numbers from pile A to pile B.
- 32:22SPEAKER_00: I got to enter stuff in the spreadsheet, I gotta like process these invoices.
- 32:26SPEAKER_00: Like, if that is your job, it is definitely a risk.
- 32:30SPEAKER_00: The more uncertainty that you deal with on a daily basis, I think the less your job is at risk.
- 32:34SPEAKER_00: And if you're looking to de-risk a job, if you are in one of those like low variability roles, figuring out how to automate yourself out of the job before somebody else does is a really wonderful way to approach it because then you will have developed the skill that you will need to exist in a post-AI world.
- 32:50SPEAKER_00: So I think that that's that's today's truth.
- 32:52SPEAKER_00: I think from a kind of more like a more global labor standpoint, if you fast forward another few years, as soon as AI starts to cross paths a little bit more frequently with robotics, I think we're gonna also have like a massive change in like the skilled labor, physical labor markets, right?
- 33:09SPEAKER_00: Where at some point it's going to make less and less sense to have humans building houses and putting on roofs and fixing sinks and doing all these different things.
- 33:17SPEAKER_00: I just don't think the technology is there yet.
- 33:19SPEAKER_00: But I will tell you, if you don't think that that's coming, you are likely wrong.
- 33:23SPEAKER_00: So if you're watching this and you're like, I'm good because I like fix roofs for a living, like you're good for now, but you may not be good forever.
- 33:31SPEAKER_00: And it like the the global impact of when that happens, I don't know what it's going to do, but I know it's going to be different than what it is today.
- 33:38SPEAKER_01: And I will say as well, like when I'm working with founders, they're crying out for people who take on responsibility and take on leadership positions.
- 33:46SPEAKER_01: So if you are in a space where you're going, yeah, well, I crunch numbers.
- 33:51SPEAKER_01: What do I do after that?
- 33:53SPEAKER_01: Have a look around, see what you could become responsible for, see what you could lead.
- 33:57SPEAKER_01: Like that is a skill set that's whether you're talking with AI or whether you're talking with people or both is very, very valued, especially now as we're sort of really shifting the idea of like what it what it looks like to lead in a business.
- 34:11SPEAKER_00: Yeah, a hundred percent.
- 34:13SPEAKER_00: I I think that I think that like the the days of the individual contributor are relatively numbered.
- 34:18SPEAKER_00: And I think that like the I don't know, I I don't love the like higher, lower hierarchy type thing, but I think like the the lowest role, I don't know a better way to say it, like the lowest role will will change from being the individual contributor and will start being like the orchestrator of AI-driven individual contribution.
- 34:34SPEAKER_00: Like I think that's where that's like gonna be the new entry level in five years or or less, is that it's gonna be orchestrating this thing because it just won't make economic sense for most businesses to have one human doing one job, except in some exceptional, you know, edge cases where there's some type of skill set that's tough to replicate.
- 34:50SPEAKER_00: But I think that you figuring that out is so important as an employee for your own earning power.
- 34:55SPEAKER_00: And I love, Sky, what you said about you know, businesses are craving people to take on responsibility and take on leadership.
- 35:01SPEAKER_00: I also see the angry people who comment on my Instagram when I talk about like, you know, having tough like bad bosses or bad management.
- 35:09SPEAKER_00: So if you if you're in that thing and you're like, Sky, I would love to work for a company like that, but like my company doesn't reward any of that stuff.
- 35:16SPEAKER_00: It's you know, just people who are, you know, suck-ups or you know, the owner's cousin or whatever, it doesn't actually matter.
- 35:23SPEAKER_00: Still develop the skill.
- 35:24SPEAKER_00: They should still do exactly what you said, because that skill will be marketable to a smarter company that will hire you to go do that new job, and you'll probably make more money at a different company than you would at the one you're in.
- 35:33SPEAKER_00: So, either way, if you want to, yeah, if you want to increase the earning power, you have to make that jump.
- 35:38SPEAKER_00: Otherwise, like progress is coming for you.
- 35:39SPEAKER_00: I agree with you 100% there.
- 35:41SPEAKER_01: Every week we are hiring for people who are project managers, and we have a whole bunch of little hoops and bits and pieces that we're doing because we're looking for people who can take responsibility and initiative.
- 35:55SPEAKER_01: Like, it's hard to find.
- 35:56SPEAKER_01: If you're doing it, trust me, it's it's well designed.
- 36:00SPEAKER_00: Yeah, and I can tell you there's two founders on this podcast right now who take that as a breath of fresh air.
- 36:05SPEAKER_00: There's nothing better than someone smart coming to you and saying, Hey, Sky, I noticed this problem.
- 36:09SPEAKER_00: And also, here's my recommendation and here's what I already did.
- 36:11SPEAKER_00: And it's like, say it louder for the people in the back.
- 36:15SPEAKER_00: Let's go.
- 36:16SPEAKER_00: It's so good.
- 36:17SPEAKER_01: Well, it's been so great chatting to you again.
- 36:19SPEAKER_01: Please tell everyone where they can find you.
- 36:21SPEAKER_00: Yeah.
- 36:22SPEAKER_00: So if this AI stuff is interesting, especially if you're running a business, I put together a like a download.
- 36:27SPEAKER_00: You can just go to precision.co slash sky, and it's basically a cheat sheet that will let you figure out like what types of roles and workflows you can automate in your business, what you shouldn't do.
- 36:36SPEAKER_00: It's this exact same process that I run when we work with someone.
- 36:38SPEAKER_00: So I just I put together like a one-page on that.
- 36:40SPEAKER_00: You guys can snag it.
- 36:41SPEAKER_00: Precision.co is the business, Matt Verlach.
- 36:44SPEAKER_00: I'm on all the places that you would expect, the YouTube, Instagram, all that kind of stuff.
- 36:49SPEAKER_00: Drop me a note.
- 36:50SPEAKER_00: We always reply.
- 36:50SPEAKER_00: It's usually me.
- 36:51SPEAKER_00: And uh yeah, I'd love to jam.
- 36:53SPEAKER_01: Thank you so much for listening to this episode of the ADHD Skills Lab.
- 36:56SPEAKER_01: If you liked it, leave us a five star review.
- 36:58SPEAKER_01: It helps other people learn more about us.
- 37:00SPEAKER_01: And thank you so much to our wonderful team for making us sound good, look good.
- 37:04SPEAKER_01: We couldn't do it without you.