Hello everybody. Welcome to this episode of the Marketing Guides for Small Businesses podcast. I'm your host today, Ken Tucker. Today we have a very special guest, Liana Ling, and we're going to be talking about what changes when you're no longer the bottleneck, why customized automations are the future, and how to build them easily. Leanna, welcome.
It's delightful to be here today.
Yeah, we're really excited to have you. Joining us is also Paul Barthel from Changescape Web, and Ian Cantle from Outsource Marketing. So what we're going to talk about really is, first of all, I want to introduce Liana to our audience. She's really a powerful hybrid, a former lawyer who's brought the analytical rigor to the digital marketing world to become the Lead Generation Queen. So that's an awesome title. She's also the former CEO of AdSkills, which she built as an industry-leading training platform for media buyers. And most recently, and I've known of Leanna for quite some time through a lot of things that you've done with Dr. Ben Atkins, for example, but you also have, you know, were part of Perry Belcher's AI Bot Summit this January in Orlando. And I've been in Perry's mastermind for over 2 years now, so. Yeah, so welcome. We're really excited. Anything else you would like to add?
I think that was a lot.
All right, well, we'll get started. We're basically just going to round robin the questions. And so I'll get started. I wanted to ask you about the compound effect. You mentioned that prompts don't compound, systems do. And we— that resonates with us a lot. But I would like for you to be able to explain that to small business owners and other marketing people who are listening to this podcast, a lot of people try to find that perfect prompt, and that's all they do. And they do it prompt by prompt by prompt. So can you kind of talk about this compound effect?
Sure. And it kind of comes from— have you heard of the book, what's it called, The Slight Edge?
No.
Heard of that? So Jeff Olson wrote that. I'm old, right? So this was probably a while ago. Google it, but really Google it for, or ask AI to show you a chart that it has. And I have, I have, as soon as I saw that years ago when I first got this book, this is really what it's all about. So what it shows is it shows a chart of things that are easy to do, very easy and delightful to do, right? And then on the, on the other part is also like things that are really painful to do and that we really don't like to do. And what usually happens is those are the things that we have to get done, but we don't do it because—
Yeah.
It's just painful or it's annoying to do. But his point is that things that are easy to do or things are easy not to do, for example, like not eating properly, et cetera, over time, you'll start to go down. Like, your performance will go down, your health will go down, et cetera. But you don't notice it because you're so comfortable in all those little micro moves, just like either doing the wrong thing or just not doing something. And then the other side of the chart shows if you're doing these little micro things that It's hard at first, like maybe it's frustrating or painful or whatever, but the more you do it, you get used to it and then it starts to compound and then things actually start to get better. And if you look at it maybe like 6 months from now, you'll see a big difference versus in the beginning you may not notice any difference. And I have found like ever since I started playing with prompts and AI and I mean, I've been working on automations for like 10+ years. I find that they do compound. And it usually starts with like one tiny little thing, one tiny move. And, and I would encourage everyone listening that I know it feels overwhelming because everybody's talking about this tool, this tool, do this here, you know, get this GitHub of like 30,000 skills or whatever, right? Just pick one thing and don't even feel like— I used to say, oh, pick one thing that's useful for you, or pick one thing that you think will, you know, help your, your business get more money. I was like, just Pick one thing. It doesn't matter what it is because it gets you going. So I'll, I'll just share like a really quick story of how one of my most massive automations kind of got started from this one little thing. So actually what I wanted to do was just find a way to get my Zoom recordings, um, into, uh, the hands of my VA and also have it posted to Circle and like all these other wonderful things, right? But that was it. I just wanted something to happen.
Mm-hmm.
based on my Zoom recording. And I was like, okay, let me just create what I thought would be a very simple automation. But I quickly found out that it was a lot harder because there just wasn't quite the right trigger that Zoom had available. And then I discovered that, I think it was like the next day or something, I got an email from Zoom saying, hey, we've released an integration with Vimeo. So I just clicked on that. Now all my Zoom recordings go to Vimeo. And from there, I can actually pull information about that video and have it do many, many, many, many, many, many things. And I even created, at one point I created this massive, I was at that time I was using make.com and I created this like massive, massive automation and it actually broke the site and I had to break it down into this like smaller one, subflows. 'Cause I didn't know at the time you're not supposed to put everything into one big thing. But that's, that was probably like a whole weekend of just going through and figuring stuff out. But that to me, that was the moment that That tiny little thing that just annoyed me, right? And that just started this whole journey of just really going way deeper into AI and automation together. So that's just like, again, that's just one example of how I find it compounds.
Yeah, interesting. I'll have to check that book out for sure. That sounds fascinating.
Yeah.
Yeah.
I couldn't help but pull up the chart right now, and it's very cool actually.
So It is, right? I don't know if we can share it, uh, or people can want to see it on the podcast, but it's like, it's something that I, I just— the thing with AI and automation is I always want to go back to first principles. And I just find like business principles, lifestyle principles, just personal development stuff, that's so much more important nowadays because that's what you can take as your foundation. Um, because AI is just—
Yeah.
Like, it's just a yes man, you know? It's just— have you— I don't know if you've seen the Stanford report recently. It was like at the end of March. Have you guys seen that one where it was saying that AI— an actual scientific research paper was published, um, that said that AI, uh, agrees with you about like 49% more than humans do?
Yeah.
And then the damage, right, that that does to us, and we don't even realize it.
Yeah. And I even just what you were saying there, Leanna, about, you know, when you create machines or create automations, don't create massive complex ones. I've had the same experience from the corporate world where we created the beautiful machine and then it breaks and then you have to try to determine, troubleshoot, what do we have to fix? And when you take a more modular approach to it, it's much easier to take the wheels off the machine and put on new ones, right?
Yep.
Yeah, exactly. And then when you start to do multi-agent stuff, it also compounds because again, you start with the one agent and then you just start to build another and another. And then before you know it, you're building like a manager and then you have a chief of staff. And then all those agents are all learning from each other because they can now talk to each other now. And it just, it just becomes this like big auto, um, learning, you know, kind of experience for everybody.
Super cool. I think we're gonna dig into that, right?
Yeah. Yeah. Paul, you have, you have a question?
Yeah. And I learned that a long time ago to build a lot of smaller automations. It's easier to fix a small automation than to figure out what happened in your big complex automation. I, I made that mistake one time.
It's a painful lesson, right? When, when you learn it, you learn it forever.
You can go, you can go from one automation to another. It's not that hard.
Yeah. Yeah.
So you talk— welcome to the podcast, by the way— and you talk about something called the Butler Test, and it's like the ultimate goal of automation has to pass the Butler Test. And knowing what the owner needs before they ask, maybe you should explain that. And then what's like the first proactive system a small business should build to, to see something like that?
Yeah, sure. So what I noticed when I speak with other business owners, like even business owners of pretty large companies, they still are very reactive, right? Like they're— and so they feel like they have to be there in the front of their computer telling their AI or telling Claude Code or whatever, like what to do every step of the way. And I'm like, maybe it's 'cause I'm lazy. I was like, I, why do I want that?
I don't wanna be tethered to my computer or even my phone, right? Telling it what to do. I think it should— like, I really think that we need to think about it like a butler. And I, um, my husband actually went to school with somebody who's like a real butler, right? And I was like— and he was telling me stuff he does. I was like, I would love that. Because what, what does a butler do? Like, it— he anticipates your needs. He manages your whole household. He manages all the people in your house and everything that you need. And he anticipates what you and your family need. And the, the idea is that before you even want to turn around to ask for it, what does the butler do? He's got it right there, right? Offering it to you right away. And you're like, oh, of course. Of course he knew what it was. AI should be like that because it can be, right? It doesn't sleep. You can set up things that just monitor things 24/7. And it's just like a mindset to ask it to create things that are anticipatory, right? That just, you know, that are very proactive. So the first thing that I did, because I manage a lot of Facebook ads, so I'm usually That's usually where it comes from, only just because that's the place I need the most help with. The first thing I did was build something that checks the pacing. So something you always want to do is know, are we underspending, overspending, on spend? Because some of these, like Google and Facebook, can just go nuts on us. So the first thing I did was I built something that just looked at the pacing at certain times of the day. And then, you know, if it looked like it was going to be on track or off track, then it would do something like it would either, you know, whatever you wanted to do. It could tell me in Slack, it could send an email, it could have something else do something. But that was the first little thing that I had created and it's now grown into like a, I, I have a very simple one too that just like takes a look at landing pages. It scrapes it, it looks at it, it's for live events. And so it also, it, it also double checks to see if the date is correct on there. In fact, I just got a notification early this morning and saying, hey, the, the old date is still there from yesterday. It hasn't been published yet. Which was kind of cool. So, yeah.
So are you saying that businesses should pick like an annoying problem to get started with, or what would you recommend?
Or it can be something you just want, you know, like, you know, like I have a client who just, we had to build this with a WordPress plugin years ago, but he just wants to know what the pacing is of the sales. That's it. He just wants to know. And then how it compares to other patterns, because that's an indicator to him of how the funnel is working. And it wasn't a pain point. He was like, no, I just really want to know that. For me, I love it to give me ideas. So I also created something that looks at the sales, looks at conversion rates, gathers all the data, and then it sends it to a brain that's sort of like a junior CMO person. And it just sends me these ideas of, hey, you know what? I think we should do this with the ad or do this. And I personally just find it very helpful. You know, but whatever you want, like literally whatever you want. It's easier for people to think of stuff that annoys them, but it's— I don't— I feel like sometimes there's a bit of a— do you guys feel this too? Like some people look down on you like, oh, why are you doing that?
Right?
Like, why are you bothering automating that? I was like, why not?
You know?
So, yeah.
I love the analogy of the butler because it makes me think like, you know, you were saying that it's you know, it's, it's watching out for opportunities to help you, you know, like the butler who's like, sir or ma'am, here's your, here's your umbrella. And you're like, I didn't even know it was raining. So.
Yeah.
Like, imagine what you could do if it was measuring your conversion rates and your leads and your sales. And then, you know, you could tell it to follow us. You could have it like, hey, analyze this like you're Warren Buffett. And then you could have another one that just analyzes like you're like, whatever, your favorite business person. And you're right, like help try and predict when it's going to rain, right? So you can put the things together. It's, it's very exciting.
Yeah. Yeah. Um, you, in your presentation that you have, you had talked about tasks versus outcomes. Um, and you noted that most people automate the wrong things. They're very task-oriented rather than outcome-oriented. Can you give an example of a common trap that, uh, small businesses automate that doesn't really move the needle? It's, it's more focused on the task than the outcome?
Sure.
Like literally pick any of the— any viral video out there that says you should automate this. That's probably it. You know, like one is, okay, so I see people who create things. I mean, I'm trying to think of a good one, but like even if you pick one that say it automates all of your social media, right? What do they do? They say, hey, you know, install this automation or whatever skill, whatever it is you want to call it. And it's going to take— it's going to go pick viral videos and then And it's just going to recreate it, do your digital clone, put it out there. And some of these can take a while to get up and running in order to do that. But now you're just creating more AI slop, right? Like, you're just creating more stuff to put out there instead of really thinking about, well, what are you trying to do here? Okay, so first of all, are you trying to be famous? Totally legitimate. Fine. There's a way to do that. Do you want to get more leads? Are you trying to get more sales? Do you do high-ticket sales and you're trying to preframe people better. Like, just all those things should— will affect that whole beautiful, you know, automation that you created that just completely automates every single piece of thing that you're putting out there and you're posting, you know, 10, 20, 30 times a day. It's just, you're looking at the wrong— you're asking the wrong question, right? Instead of saying, how do I, you know, how do I clone myself and get myself out there 30, 50 times a day? You should be asking like, well, how am I What do I need to build in order to get 10 leads a day or 20 leads a day or whatever, however many, you know, sales, just as an example. Others I've also seen too where it's inefficient, like, and I understand like some people it's fine. Like they're like, no, I will pay the extra rates. I will pay all of that just so I don't have to touch anything and just have Perplexity or Claude do all of it. But have you seen things like where people are doing these personal assistants where you talk into it, then it's translating what you're doing and then it, has somebody else think about it, then it might have somebody else do something else. And like, you've just, you've gone through all these different steps in order for it to send an email to somebody that say confirmed or something like that, right? Like, it's just, it's just, it's a little bit overbuilt. I found some of them too. Does that make sense?
Yeah, yeah, absolutely. The, the, the, you mentioned AI slop, right? I think all of us in, in business and especially in the marketing realm of those businesses, I think we're seeing just the, the explosion of content because it's getting easier and easier to build content. But it's just what you're saying. It's not necessarily strategically motivated. It's activity motivated where they're like, well, I got, I got to be doing this because somebody told me, or I got to be doing this because our competition is doing it. But they have no strategic framework as to why or how to, you know, how it's going to move the needle. Yeah, yeah, yeah, yeah, totally. It's actually making it harder to stand out in the marketplace because there's just so much noise now. Yes.
Yeah, yeah, yeah. It sadly, you know, too many small businesses and probably even larger businesses, they, they don't focus on outcomes. You know, they, they, they lose sight of that, you know, and, and the metrics that they look at, you know, aren't outcome-based, they're vanity-based, you know, too oftentimes. So Exactly.
Yeah. Or
Or like the automations that say, hey, you can make 300 ads a day and put a bunch of them on Facebook. And I'm like, why do you want to do that?
Right?
People are, I don't know, just because I can. And I'm like, well, shut it.
And now you have to manage those ads, right? And the budgets and the pace.
And yeah, and they're like, oh, I have $10 a day to spend. I'm like, no, you're not doing that.
Right now you've diluted your spend over 300 ads.
Yeah.
Definitely.
So, Liana, you talk about kind of using a phased build strategy, 30, 60, 90 days typically. For a team that has limited resources, how should they really define phase 1 and set that as the priority? And how does that help them build confidence and then move forward to, you know, tackling the more complex part of the workflow? Yeah.
Well, I'm, I'm, I'm a big, big fan of just micro moves, micro goals, micro anything. Um, but I do think that you have to spend some time to really understand what you want. Um, just, I can't, I, I'm still amazed at how many people I speak with that they really don't know what they want. They, they have very, very little clarity on what they actually want. Um, and just from my experience, I don't know, I'd love to hear from you guys if you find this too, that the, the, The more clearer you are on the picture of what you want at the end of the day, even like what you're working towards, the better result you'll get from anything that you're building, anything that you're chatting with, you know, AI. And people just don't stop that because just think about that, 'cause we're all, 'cause all the marketing out there, all the viral videos, it's all very tools-focused right now, right? That's—
Yeah.
We've seen this happen in the big martech explosion before. It's, you know, we're going through that right now, right? It's all, it's all, oh, latest tool, latest tool, and the latest AI thing. And the latest, and that's all anybody can think about. And then because they're always chatting with AI, it just keeps patting you on the back and telling you to keep kind of doing that way. But you really have to step back and stop a moment and really think about what is it that you really, really want? Be as specific as possible. I remember I was speaking with these guys who have their PhD in AI, and I was like, how do you think of all these things? Like, what do you do? And he said, just 2 things. They focus on looking at problems, like, and identifying problems and just getting really, really, really good at that so they can get a lot of clarity. And then the second thing is they work on developing their imaginations so that they can be as creative as possible to get to the outcome that they want, or even like imagining these other different types of outcomes that they want. So I know it doesn't sound sexy, but like spend some time just to get a lot of clarity on what you want and you may need to chat with your AI. That's totally fine to get more clarity. But don't do anything if you don't— like, I just think it's a waste of time if you don't know where you're going, where you're headed.
Yeah, I think it can be a little dangerous chatting with, you know, with your AI because it's going to recommend next steps that it thinks you should take.
Right.
And those look really good and very tempting. And before you know it, you know, what you could have done, you know, in that initial conversation, you know, in maybe 15 minutes to half an hour, as you're drilling down more and more deeply into it where you're driving it, you find that you're being led down a path that creates a lot of really cool things, but really doesn't achieve the objective that you were trying to achieve.
Exactly. Yeah. Which is why, again, you need to just be really clear with your outcome. Keep reminding it what your outcome is, what that you want, right? And just, just like, like any leader of a team, right? You're gonna, you, you have to make sure you keep everybody, you know, on, on the right path. So, sorry. So, to go back to your original question, again, I think that's really needs to be very clear on what we want. And then I think like at any team, whether you're in a big company or you're in a small company, I just say like, pick one, just pick one thing, pick whether it's something that annoys you or there's something you just want. And I would recommend that you use, you know, 'cause this is what I do, like I will ask Again, 'cause I know what I want, like, but I'll ask Claude or, or Gemini or whatever and say like, hey, I, I wanna solve this and I wanna solve this using, you know, make.com or N8N or like whatever tools that you have available. Or, and even, even if you don't know, you're like, I wanna do this and these are the constraints that I want on it. So I, I'll ask it a really good question about what I want. And like nowadays it, like AI will build all the stuff for you. You don't have to kind of get in there always to, you know, drag and drop and things like that. But that just gets you going because I have found that, I think more than any time that I've seen kind of just these new waves of technology coming forward, this one really is, you really do learn from practicing. Like, the more you go into it and the more you try it and the more you overbuild, the more you fail, the faster you actually learn. So, I just want to get people started with something, even something as simple as going to lovable.dev and asking it to generate something for you, just like a calculator or a dashboard or something. It just gets you, and then you see the possibility, and then you start to tweak it, and then you start to understand how things get better, and then you get hooked, right?
Yeah.
That's what I would recommend for any team. That makes sense. Yeah. And then you can just kind of start to go, you know, build up from there.
So you were just talking about N8N. So Zapier versus N8N, and you might have to explain a little bit what those are. I mean, to us it's like, yeah, of course we know what it is. A lot of people don't. But you talk about Zapier speed and N8N is scale. So when should, and even should a small business owner go from Zapier, which has a really user interface, to N8N, which has a lot more features, if you will?
Yeah, it kind of comes back to what you want, what, what's your priority. So they're all good. The main ones for people listening is Zapier, Make.com, and N8N. And N8N is just a little bit more technical to use, and I didn't start in it either. I started from— I remember I used Zapier when it first kind of came out. So But Zapier tends to be the most expensive one. So I couldn't— the reason I thought I would take the pain to learn N8N at the time was I was like, the cost savings was just too much for me to ignore, right? So I think, yeah, like if you can set— if you, in order to set things up quickly, Zapier is great. It's very, very easy to use. It has an AI built into it already. Just say, do this. It has a, I think, not a bad interface still in order to figure stuff out. And it gets you live fairly quickly. I think once you start to see that you're hitting, like you're having to upgrade constantly on your Zapier, that's when I would say start taking those high-volume Zaps and move them over to N8N. Just hire somebody to do it for you. That's usually the easiest way to do it because then they can just go through all the testing and stuff for you. But, or you can have like Claude, um, you know, recreate it, you know, for you and you can kind of learn it yourself. Um, but that's used to me. I'm just looking at it in terms of, um, you know, cost savings because that, that's really why I, you know, kind of use it versus Zapier, if that makes sense.
Yeah. Yeah. Interesting.
Yeah. The beauty of Zapier is that it has so many integrations, right? That's, that's what makes it so appealing and easy to use.
Yeah.
Yeah, so just the same thing with a lot of other things too. Like if you're using— I know somebody, he loves Perplexity Computer, but it's expensive, right? But he loves it, right? It's very— and I— but I know at some point he's going to get to a point where he's like, do I really need to be paying like hundreds and hundreds of dollars a month to do this? I'm going to look at ways to make it more efficient, right? I think it's the same thing with Zapier as well. And I get it, some companies, they have Man, I've seen some Zapier accounts. It's just, wow. It's like so, it's just overwhelming amount of stuff is in there. And I understand like it could be a very, very big job, especially when it's connected to everything. But, you know, the cost savings are just so significant. I would definitely, you know, take a look at it, take a look at it. And like I said, the good news is that the AIs now can just create the code for you in there. Like it's not as, it's really not as bad as when I remember I first learned it.
Yeah.
like it couldn't do that at that point.
You know, one of, one of the things that occurs to me, Leanna, is that, um, and, and I've seen this before with other things, but, um, CEOs, presidents, you know, the highest paid person in the business, um, are, are— I see it all the time. They're getting lured into using AI a lot more. And one of the observations I've made just as a person looking at, at some of these people, because I get caught in it too, is that The more I think about it, the more I think they probably shouldn't even be the one doing the ideation, the testing. They should, they should hire a junior person that is like their AI integration specialist, that all they do is they take the ideas of the team and they try to make, you know, these beautiful machines and see what the results are.
Yes.
Oh my gosh.
Thank you for saying that. Like, thank you for saying that.
Yeah.
I'm like, why are you building landing pages? Why are you doing this?
Right.
Yes.
Yes. Yeah.
One of the other things that came out was something you had coined the super keen CMO, Chief Marketing Officer. At least that's what Ken gave me. So, okay. But, but what the description was, you envision an AI agent that detects patterns and decides what the next test is.
Yes.
And, and how would a small business owner set the guardrails on something like that? Because it almost feels like we're setting the the robots loose in, you know, in our kitchen and letting them go crazy or in our business? What, what are the guardrails? How can they be strategic with this?
Yeah.
So I have this vision of, I haven't quite achieved it yet, but I'm getting close, of creating a self-healing, not just a self-healing funnel, but like a self-healing business, right? That's constantly optimizing over and over and over again. Um, and, and, uh, so let's just take a look at the funnel, for example. And I do have something that runs where it's looking at everything, it's looking at the data, and then it's, like as the CMO, right? And it, and then it also knows everything that's happened in the past. It's aware of what the date is today and sort of what are some, what are some things happening, you know, in, in the world. So, because that impacts marketing, right?
Mm-hmm.
And then it can actually tell you, it can, it can deploy tests or it can actually ask you. And I think that the key thing here is making sure that you build in like just human checkpoints. That, that's all. Like, think about it. Like I, I tell people, think about it as if you hired um, somebody you don't trust, like hired somebody from your competitor who— they're, they're very smart, like they're, they're a nerd, they can geek out, but they have zero interpersonal skills, okay? And you can't trust them with anything at all. Like you can't give them your bank account or whatever. Like how would you set that person up, right?
Right.
Uh, and, and I set that up actually on a computer here in my place here, and I like— it's all kind of walled off because I, the whole time I thought is how would I set somebody up so they can actually do stuff for my business, but You know, they're not going to have— I don't trust them. They're not going to have access to my bank accounts or credit cards or anything like that. And if they're going to do something, they got to ask me first, and then I can— I'll look it over and say, yeah, okay, you can, you can go ahead and do that. So think of it that way, right? Like, and, and that way you can adjust it to your comfort level, you know? Um, like, sometimes I don't mind if, if it goes ahead and it does things because I know that it's all still in draft and that I, I, um, like, I can press the you know, publish button, which the CEO should probably do, right? But, you know, let it do all that other work. And that's— that way it's like, I don't care if it makes a mistake because I know that, you know, it's very low impact, right? Because I can just— I can just press that button. I can decide when to press the publish button. So I would just say think of things like that. And it can be just as simple as you having it go, okay, like, tell me your ideas, just like you would anybody else, right? Tell me your ideas. Okay. You know, you can send it feedback or you can say, yeah, go ahead and you know, go ahead and do it. I think that's a great idea. Let's go ahead and do that. Uh, you know, I think that to me is just a simple guardrail that pretty much works for almost anything. And I feel like it makes people a little bit more comfortable working, you know, just setting it loose, um, if you want to kind of call it that.
Yeah, it reminds me of when I was in corporate marketing where we would have marketing interns come, you know, either they're in the middle of school or they're finishing school and they're interns in the large Fortune 500 businesses, and they have lots of energy. Uh, they have lots of ideas of how to change the world. Um, but they have no experience, right? And so, you know, you set them loose on an idea, but their, their whole point of their role is to pitch the idea to somebody more experienced who can test it in the environment of real life, real business, right?
Mm-hmm.
So it feels the same as that, where these are the, the agents that we're talking about are kind of junior marketers, and, but we're still, the ones that have to say, you know, approve the expenditures to do that, or approve, you know, you can go to the next task.
Yeah, but I'm so excited about building something like that. And because it's just the amount of data, you know, that it can go through, like, just think how much, how powerful that would be for each one of us, right? Because it knowing like all the historical data, and even if you can get it to get competitor data or whatever, like just all of that, looking at it, looking for patterns that we can't see because we just can't handle all that data.
Yeah.
And I really want it to be able to be like, I'm glad you mentioned about the weather. Like, I really would love it to be like a weather forecaster, you know, for what's gonna happen. Because I mean, it's that, that's all it is, right? I think it's just getting really, really good at the numbers, you know?
Yeah.
Yeah. So I think you talked a little bit about one of the key roles that I think people have to embrace when they're adopting AI, and that's kind of this human in the loop or human-driven AI with the right checkpoints and the guardrails, both on the front end to guide it, but then also on the back end to assure the quality and consistency and brand voice. You've talked about, you know, content repurposing a fair bit as well. And, you know, I think this is a great use of AI. Give— can you give us an example of, you know, a good easy way to get started doing content repurposing? Let's just say from like a staff meeting.
Yeah.
What could you do there? And how can— and then, you know, when you turn AI loose, how do you— how does the business help make sure that it is consistent with their voice?
Sure.
Yeah, I love that question. So I would first start from something that you are already doing. So if you're already— I was working with— I was helping implement a coach where He has these weekly calls, so he's already doing it, right? So you want to be able to leverage what you're already doing. Or if you have a YouTube channel and you're just, you're just already putting out YouTube videos, like just start from that one core thing. So start from something you're already doing. And like I said, it could be as simple as your coaching calls and ask AI to build you a very simple automation that takes What just— and I would even have somebody even just upload the transcripts. Like, don't even get too fancy with this, right? Because like, just say, hey, upload. Because most of, uh, and most of our, you know, note-takers and stuff already have, you know, the transcript created. So just have it put in some type of Google Drive, and that just, um, that triggers from there. And, and then have it, uh, take that information and pull out the main insight, and then just create You can have it write a Facebook post, write a LinkedIn post, write like an Instagram post, write a script for a short video, write a blog post, you know, about it. You could— so all these things you can, you can actually create just from that one thing. I even, for this other person, I actually created, because that's what he wanted. He's like, oh, I can actually create workshops to sell if I talk about something good. So I actually had a step where it looked through the transcript and went, Did I say anything good here? You know, is there, is there a workshop idea here? And I said yes. Then it actually created the whole outline for the workshop based on the template that he liked. It created the words for the order form, it created the email sequence, and then sent a task over to his operations person to, you know, put it all together and try and sell that. Like, there's as many things you can kind of have from just having that one original piece of content. Now, And I know it sounds complicated, but just pick one thing to do. Just be like, okay, how can I create a Facebook post from this piece of content? Or how can I create 5 different Facebook posts from this piece of content? Just pick that one thing to do because adding all these other things is actually pretty easy once you do that. What I found in terms of keeping things to the voice and also trying to make sure it doesn't hallucinate too much is I really try and define the context as specifically as possible. So I make sure that part of the prompt that's writing stuff, right, like let's just use the Facebook post for example, it will have a section that says use this style voice. And so in the automation, I, I have a place where it pulls in this, the, the style, you know, like edgy, whatever, whatever it is the style is. Um, and then I'll have another one that says then make sure like you review these foundational documents. So I also have it pull in, like for example, one of my clients wrote a book, right? And he said, this is the book I wanna pull stuff from. I'm like, great. So I have that as part of the flow. I also had AI create for me a markdown file of just like everything it knew about him and what he teaches and what we should say and what we should not say. I said, please give me like a, just like a cheat sheet, but that's detailed enough so we'll guide it. So every single time it goes to write something, I say like, you need to look at this and stick to this, like this stuff here, instead of going out onto the web and finding out or just going from your own memory. And that's been really, really helpful to make sure that the, like, the author's original intent comes through. Because some, you know, like sometimes AI just makes up stuff. Like it, it, it just pulls out certain words and, you know, it thinks that you're talking about something else because that's what's out there on the internet, not this person's actual thoughts. And yeah, and then it also just sticks to the style better in terms of what we're writing there. And I found that that has improved things a lot for just anything that I'm doing. And even if you're doing it in a project or whatever you're using, they all have similar things, but they allow you to upload files, right? Upload those things, like upload the book, upload samples of the style in there. And so if you're working on it just even in a project, Just have that context in there so that it knows what to do. And just keep giving it more and more examples of what you like as well. And then that's another way to also kind of train it just to get it going on that path. It's really easy to put something up. The hard part and the tedious part is testing, changing things, testing, changing things, testing. I mean, I had one where it was created for somebody in the UK and it kept spelling everything with American spelling. Man, that took like a week to train it out of it. But that, you know, it's like little things like that would just, you know, everybody else would probably just give up. But that's the tedious part, actually. That's the dark side of doing automations and using AI and things like that.
Yeah, I think it's super important too, because, you know, to get back to the AI swap idea, you know, a lot of people think, okay, great, I can just go type in a simple prompt, have ChatGPT write me a 750-word blog post. I put it on my website. You know, it's not trained in their voice. It's not trained in structuring good written content for SEO purposes or for AEO purposes. And, you know, and Google ignores that largely. I mean, it's just not valuable. So, but sadly, that's, you know, that's what I think a lot of— too many people— I wouldn't— I don't want to say a lot, but too many people take the easy way out because let's face it, a lot of small businesses, they don't understand how important marketing should be considered in terms of an investment. They see it as an expense way too many times, including their own time.
Yeah.
And, and so you, you can't cut corners and expect to get good results. You have to go through the training. And, and if you do that, this content repurposing idea alone is a game changer for many small businesses.
Yeah, just do it from stuff that you create yourself. Like, if you're a roofer, create videos just showing, like, just even teaching or showing how you do things, talking about, you know, your clients, instead of, um, hey, go find the most viral videos and just create something for me, right? Because then it becomes just a cycle of AI slop, right?
Yeah.
Um, can I share with you something that's just been really game-changing for me in the past 2 weeks?
Yeah.
It's kind of— so it sounds like the opposite of AI, okay? But believe it or not, I took a pen and paper.
Okay.
Shocking, right? I know, right? I know, right? So a friend of mine showed me this. He was like, no, get your ideas down. And I don't even have note cards. Like, this is just like scrap paper that I had around. He's like, write down if you're doing— I did this for a whole book even, but he's like, write down the topic and just a few points that you're going to have here, and then you carry these around with you. And then I found, um, I use WhisperFlow. Like, a lot of people were recommending it to me, and I said, okay, I'm finally going to try it. So they have a 14-day trial period, and they gamify it, so you, you have to talk into it every day, right? So I just, I have it on my phone, and so I would just carry these with me, and whenever I had a moment, I would pick out a card with— that's the topic— and then I would get my phone, use WhisperFlow, put it in— I use Evernote, but like, just put it anywhere you want. And, uh, I like— it's been phenomenal. And then I would take that into Claude and say, hey, like, clean this up for me and everything. I've, I've never— I found that, um, like, the output really, really, uh, I feel like it sounds more like me.
Mm-hmm.
Like, this is the only time when I feel like it really sounded so much more like me than anything else. Even when I've hired professional copywriters, you know, to write for me. And it's just some magical thing with just the ideas start on pen and then just dictating it using WhisperFlow and WhisperFlow cleans it up a little bit. And I think it's helping me train also like just the AI too to get more used to my voice, right? But it's been amazing. Like I've written a book, I've written my newsletter, I've written all these different ideas for blog posts and just like pushing it all through. It's, it's a— yeah, it's like really life-changing for me.
Yeah.
I love it too that you're still exercising your leadership and creative skills in the dictation or the ideation phase of that.
Yeah.
Yeah.
Yeah.
It's just, I don't know, I think there's something, there must be something in the science too about, because I like to pace, right? So, or even like I have my phone, I'm outside and I'm just like talking and And again, you just become so locked in on what you're doing, but you're also walking around. And you're right, a lot more creativity comes out. Yeah, it's kind of magical.
Yeah, so you mentioned AI slop and AI hallucinating. And some of that comes from the fact that AI doesn't remember intent the way people do. And AI is not perfect.
Oh, really?
I know. I think a lot of people think, oh, wait, it's AI. It can't make a mistake. And it makes mistakes the time.
So, it's bad.
You have a high-level system called the Digital Integrator, and I think you kind of touched on this a little bit, but it basically tells the CEO what to get rid of. So how does a small business, small business owner feed their personality and company branding performance into AI so its advice is actually relevant to that business?
Yeah, that's a good question. Um, I don't know how many people know that there's a personalization section in most of these chatbots, right? Um, like, ChatGPT has it, Claude, and you can even see what it remembers about you and filter through that. So my first thing is it's probably already taking notes on you. Think whatever you're using— ChatGPT, Claude, whatever you're using— find where it keeps all the memories. It's usually in the setting area. And I bet you'll be surprised at some of the stuff it's saved about you, which is causing it to say things that you're like, why would it do that? Like, at some point, oh yeah, it said I love tacos. I know, like, I know, sorry to offend. I don't want to offend anybody who loves tacos, but like, we were talking about it and it got into its brain that I loved tacos and I just, I don't eat it that often. But I was like, why is all this coming up? And I saw like it kept it in there in the memory and I'm like, okay, like I'm taking that out, right? Just as an example. So find that first. The second thing, and it goes back to when I said, hey, the first step you should do is really figure out what you want. I don't think people are spending enough time creating like a soul basically for their AIs. Do it however it makes you feel comfortable. Like, you know, dictate into something to just to pour out like, here's basically all my insecurities, you know, here's my fears, but also like here's what I wanna work on. And here's the type of person that I work with, like the clients I work with, here's how I like to work.
Mm-hmm.
Here's my goals. Pour all of that into something. You can ask an AI to turn it into like a SOL document or S-O-U-L, or even just like a custom, you know, instructions for your AI so it has all that information, um, at least at the core. Okay, um, and another thing I do, I don't know how much it affects it, but I tell it, I said, it is your mission no matter what to make me successful. So you are to talk to me even if you think I'm going to disagree, etc. Like I said, that is your main mission, is you have to make me successful, and I define what success is. And I have that in all of my personalized, you know, instructions across all the chatbots. The second thing I would do is— and the Digital Integrator is actually a program that Kevin O'Nations and I created. And, and we kept, like, the way that we have it, it's basically like a custom GPT or like project or something like that. Because essentially what it is, is, uh, it, it, we get all the answers to the questions about like what's happening with this person and their business, et cetera. And that gets stored somewhere. And then we have like a specific way of, you know, hey, like here's how you're gonna filter through everything, et cetera. And that gets stored somewhere. So the actual prompt that happens when you feed stuff in is, It's pulling, okay, like, I need the context of who this person is. Okay, I have to follow analyzing this data in a certain way, which is basically, you know, cut out 90% of everything, because usually 90% of what you learn, you don't need to do, right? Find the one thing you need to do, and then just, you know, let them tell them that, but tell them that in the context of what it knows about you and your business. And that's how we kind of keep it. That's how we keep it that way. Like, I think on point. Again, it's— you still gotta, you still gotta push it, right? You still have to be skeptical. It's your enemy, right? So I'll also do things where I will take the output and I'll have it run it through 1 or 2 times, and then I'll also take the output and I'll have— I'll run it through another AI. Like maybe you do Claude versus Gemini versus Grok versus ChatGPT. And sometimes I'll tell it that, like I'll come back and I'll say, well, Grok thought this, and then Claude will be like, oh, well then, you know, and it tries to one-up. like Grok or something, but it's helpful just to see what the other ones, um, think about it, you know?
Yeah, that's a good point because I use ChatGPT a lot and we have clients across all kinds of different industries, and I'll be working on something for a client, putting a prompt in ChatGPT, and it'll come back with some weird stuff, and it's like, okay, I, I know where that came from. So you just have to get more specific with that prompt and say, okay, so like if I'm writing a prompt for, say, a voice AI, and, and I have it give me the basic structure, and it'll start putting emojis in there, it's like, no, don't put any damn emojis in there because it won't accept it. It'll kick it out.
So it loves emojis.
Yeah, it's very— you really have Yeah, something I don't like— ChatGPT. Yeah, I, I still have it. I, I do use it quite a bit because there's so much my history in there. Um, you really have to tame it. Like, yeah, really, you have to.
It was just kind of one of the first ones, so it's, it's kind of like PC versus Mac. It's like, I grew up in a piece— I grew up in a PC world. I made my money in, in IT, and so I have nothing against Macs. I'm just more— I know how a PC works. And I think that's a lot of— if you know how Groq works, if you know how ChatGPT works, you're going to tend to stay with— stay there. You're right.
You're absolutely right. I love that. I love that PC versus Macs. That's so me too, because I'm used to PCs. I have a MacBook, but it still feels like an alien machine when I use it.
I have a friend who got tired of ChatGPT always saying yes, basically, but also always offering kind of a next action because there were— it used usually offers, you know, 2 to 3 next actions, and some of them are helpful, some of them aren't, especially if what you're asking originally is actually the finished product you're after. Um, and so they had built this elaborate prompt and, and trained their version of ChatGPT so that it would not always answer yes. It was being more critical, um, because that's what they wanted from it. And, and it— if, if they were asking for a finished product, it would only give the finished product. So I thought that was interesting. I haven't done that, but I like some of the ideas it gives you.
Yeah, you can tell ChatGPT to put something in a code box that will eliminate a lot of that stuff.
Yeah. AI ad copy, AI-powered ad copy. What can you tell us about that? How can the round— you touched on it a little bit before about kind of having, you know, using AI to emulate famous marketers to help you with stuff like that. But talk a little bit about that and how we can leverage that businesses?
Yeah, I love it. I've been using it like since, uh, you know, again, since before it became popular. And the AI copy that I wrote, I didn't tell them at the time, but like it beat out some of the copy written by human, like super, you know, very good copywriters out there. So I'm a big fan of it. Um, I'll just, you know what, I'll just walk you through how I would do it on ChatGPT nowadays because I've— I'm finding that simpler is better. On, on Facebook right now. Um, I'm— I, I test a lot on Facebook. I find that just with Facebook, you just have to test, but you have to balance it out with, as we mentioned, like your budget. Like, don't put 300 ads up because it's just not going to work well, especially if you only have $10 a day. But literally how I start every offer is if I'm using ChatGPT, I'll just grab the landing page, the offer, and I'll say act as— and then use your favorite marketer, you know, like, uh, a lot of people like Alan Sultanic, Nothing Held group, uh, Russell Brunson, Perry Belcher, kind of use that. Or you can say, um, act as, uh, the top zero— you are, you are, um, like you're a CMO, or you're like a marketer in the top 0.3% of marketers in the world, right? Just give it some type of role, right? And just say, uh, write the primary text and headline for Meta ads for this offer and make sure that it is compliant with all Facebook rules. So it's going to give you something. And then I use— and I say, can you do— I said, review your work. Can you do better? I really like using that a lot. Seems to help.
Yeah.
And then so it'll start to give me some stuff and I'll probably run this through 2 or 3 times and I may bring it over to Claude as well to rewrite it until it's where I want it to be. But it actually gets me pretty far, you know, and just for that. And then I really spend a lot more time on the creative. So I'll share with you guys like a secret that's been working really well recently.
All right.
Is, uh, look, I, I love videos. Video's great, but it takes so long to, you know, make them, even when you're doing AI videos. So in the beginning when I test stuff, I'll do it, I still do it in 2 ways. One, I'll have it write 20 headlines and subheadlines, and then I'll say, review your work. You know, would, let's just use, uh, you know, Russell Brunson's name, for example. Like, would Russell give this like a 9 out of 10, or would, Would Dean Jackson give this like a 9 out of 10? If not, keep rewriting until like he would give it a 9 out of 10, right? So then you've got this whole list of stuff. What I do is I just use Canva, very simple, do a bulk import, just create very simple plain background with just the headline and the subheadlines as images. That's your ad. So I usually start with that because I'm trying to figure out angles and people just, they just read what's in front of them, right? So that's the first thing. The second thing I'll do is once I have an idea of the angle, I'll go back In fact, I find ChatGPT is really good at this. It's better at this than any of the other LLMs. I'll give it the offer again. I said, here's my offer. I want you to create simple image ads for Meta that— I'm trying to remember the exact words. I can give you guys the prompt to share with your audience after this, by the way, for this.
That'd be amazing.
Because it's longer. But it's like, I want you to talk— I want you to focus on what's on people's minds, right? So I call them like short on-my-mind type of ads. And it's a very simple cartoon. Like, it's not— and then you can actually tell it what style of cartoons, you can match it to your audience. And then I asked ChatGPT just to create the image. And for some reason, it just, it's just really good at this, you know? And it's these like cartoony things. You're like, why would anybody care about this? But it's working really, really well right now. I think part of it is that it looks the opposite of when you just type into Nano banana, create an ad for this. And because I can tell if it's created like that, because now as I see enough of them, um, and I just think there's a lot of that happening. And also Meta's AI creates, um, uh, images for you as well. And they, they, to me, they still look a little bit like the way that AI lays things out on an image. So I just think people aren't— it just doesn't look like AI. It just looks like a simple cartoon, you know. And it's like in some of my campaigns it is crushing it. You know, I partly, I think it's because it's just hitting on what people are thinking about, but they don't wanna admit they're thinking it. And it's just getting really good at get at, at doing that. And yeah, and then just creating some type of very simple cartoon. And like I say, you can use different styles. Like I've tried it, for example, where I thought, oh, I think this audience, they read like the New Yorker a lot. Right. And that's a certain look. Right. Or I did one for moms and I asked it which one and it gave me a couple to try and one was like a very retro colored actually one. And that crushed for a while as well. I was shocked 'cause I was like, oh, like why would, why would it, you know, why, why go there? But it said try it out, right?
Yeah.
So yeah.
Super cool.
Yeah. Yeah.
Speaking of AI doing dumb stuff, I was—
That should be a whole new podcast, by the way.
What's that?
That should be a whole new podcast for you guys.
It could be because—
Speaking of AI doing dumb stuff.
I was working on some content for one of our clients, and it was about exterior painting, and it was an AI-created video. And part of the video was a woman in a bikini measuring her waist. It's like, okay, what does this have to do with exterior painting? I don't quite get that.
But it caught your attention.
It's like, I don't know how this is relevant, but Or where it even pulled that from.
Yeah, yeah, yeah. So the last thing I wanted to say about AI too, like, and just Facebook ads is— and again, nothing against these programs that are coming up that are helping people create more imagery and creatives. I think that's all great. My beef with it is that it's going into the Facebook ads library, it's seeing what other ads are running, and it's pulling from there. And again, that's fine too, but I think we're at a point where it's just regurgitating the same thing. Like, we're all going to the same competitors, the same profiles, looking at the ads that that's running and just reworking that exact same ad. And so, and then it's— and then I'll just do it to the next person, the next person, the next person. So, you know, use AI, but be a human and steer it in a different direction in the way that I did with these cartoon ones, right? Like, it's just because it's not out there. Soon it may not work because everybody's gonna just start regurgitating that and copying that. So you have to think of something else. But that's what these a lot of these software are doing, and everybody's just using the same idea. So ads usually work when you do something that is the opposite of what everybody else is doing.
Yeah, you have to think for yourself sometimes because really, um, it's like when somebody does a study that says 10 o'clock on a Tuesday morning is the best time to send out marketing emails, so all 5 billion people start sending out their— so now you're just caught in black hole of an inbox and people can't delete your stuff enough.
Yeah.
So send it at 3 o'clock in the afternoon.
Yeah. Yeah. Yeah.
And then they think you're a genius, right? You're like, oh, who would have thought of that?
Right?
Leanna, you talk about, or you actually recommend doing a dollar a day strategy for testing creatives. Can you kind of unpack that a little bit and talk about, you know, why that works?
Sure. And, um, for everybody listening, uh, go ahead and Google Dennis Yu, Y-U, Dollar a Day. That's— he's been teaching this for like 20 years. Um, that's where I learned it from. He's a friend of mine, a good friend of mine. Um, and I love the idea about— behind it. He always says that he invented it because he's cheap. He was like, what's the cheapest amount of money I can use to test something with, right? Um, so, so, and it makes sense.
And he—
it also helps you with your brand.
Yeah.
So let's just say, Ken, that you wanted to also get your brand out there and you had a number of videos that you had created. What you want to do is start with Facebook, just run some video ads or engagement ads, and just set it the minimum it'll allow you to spend. Now in Canada it's not a dollar, in the US it is a dollar, okay? But you pick that, you do it for each video. And so again, depending on your budget, you choose however many you want to test— 1, 2, whatever, right? And what you're looking for actually initially is 10% engagement rate, not a 10% of people watching the video. So you just have to make sure you're actually looking at the post engagement rate, because pretty much across the board, anything that has over 10% engagement rate has a pretty good chance of doing better in other— in all other different types of campaigns. And so it's just $1 a day, the minimum. And then once you start to find, you know, just some videos or posts that are doing well, then you can take that and do it on other platforms, right? Like, imagine then you maybe you have like 3 videos, like these are doing really well. So you're like, okay, go to X, run minimum per day per video on X, do it on LinkedIn. I think LinkedIn's like, you have to spend like $20 a day. I can't remember. But, you know, minimum on LinkedIn, minimum on Google, like, and imagine, then you'll notice that people will start to say like, oh, like, I'm seeing you everywhere, right? And this works really well if you're local. Because what I like to do is tell people, Like, I want to make you like the mayor of your town, pretty much, right? And so all you have to do is just like go to all these different, um, platforms, use the minimum per day, and just plug in the zip code or whatever it is for the town that you're trying to blanket everything in and just run it. And people will start to be like, hey, I know you, or whatever, you know, like, um, and then it's this— the goodwill starts to build as well, right? Know, like, and trust. Back to first principles. sales, you know, it starts to build and, uh, it's a pretty— it's pretty powerful. And then also once you find stuff that works, especially in Facebook ads, you can take that same ad and then you can use it in one of your conversion ads. Like if you have a purchase, a lead, whatever, because you know that that creative already engages people quite a bit. And then you can, you can also use it, reuse it inside of those campaigns. Um, one note I would want to let people know, because they complain about this to me, I hear this question a lot. People are like, oh, but you know, I can't measure.
Yeah.
This. I can't, I can't tie this to sales directly. And you need to let that go. Like, just accept it's— you're not going to— like, it's— you might be able to— it's like somebody may say, oh, like, I saw it, like, just anecdotally. You're not— just drop, just, just drop the, the idea that you can. And what you want to look at is your omnichannel, you know, results. Just at the end of the day, how much money do we make? How many sales did we make? How many leads did we make? Week, you know, and how much did we spend? And then you can make little changes and then see if that affects kind of your, your total thing at the top. Uh, because over and over again, just long term, there's so many stories from people just say this works, this works. But in the moment, you're gonna think it's not working, and, and it's going to frustrate you. But it's, it's about consistency. And like I said, I love it when it's local. You know, a friend of mine lives in like a town of like 5,000 people and he did this for his business and he told me like he, he, the first time he walked down the street with one of his kids, uh, going into school and somebody said, hey, I know who you are. Like, you know, you run this thing. And he was like, yeah, he became famous. Right. And he got like, I think he got like more than half the town on his newsletter list too. Like it was crazy.
Right.
Yeah. That's, that's to me is like the power of Dollar Day. And then it does all these cool effects on like helping you with your SEO and stuff. But just Google Dennis's stuff is for free everywhere and he'd be the first one to say just go Google him and like ChatGPT knows all about him. He's legit. So it'll explain it further.
Yeah. How long would you recommend running these campaigns for?
Forever. Always. So when you're testing it out, I would run it for 7 days, maybe 14 days. And then you may just wanna go to the next one, the next one. So in the beginning you're like, you're gonna have a lot of duds, right? You're not going to figure out what works. And some people, video works. Some people, images work. I know somebody else who just has really— he just finds really interesting stories and, and images to share. There's no call to action. He's just trying to get engagement, right? Uh, and that works for him. I've tried that. It doesn't quite work as well for me. Um, for me, I found it's more like videos. But yeah, whatever, whatever works. And then just find something, just let it run. Dennis told me he has one running for like 5 or 6 years on It's just, he just doesn't touch it, you know, because it's working. Um, somebody else told me he does it for his clients and yeah, they, they take a look at stuff and at the end of the year and he's like, most of them are just like, don't turn it off.
Don't touch it. I don't
I don't know how it's working.
Just don't turn it off. Yeah. Wow. So, uh, Ian, Paul, you guys have, uh, any, any last questions that you want to ask Leanna quickly here?
Real, real quick. I do, uh, just Messenger ads versus traditional ads.
Any thoughts Recommendations?
I start with traditional ads, like traditional route. I think that that should be people's bread and butter. Use Messenger as a test on top of that. I do find that they go very hot and cold pretty quickly. I do wonder if it's just because there's less people on Messenger versus the entire ecosystem. So at some point, you just use up that audience. Messenger ads are the ones where after a while, you start to see a lot of trolls come through, and that's your sign to stop. You maybe take a break or, you know, create another batch of them and try it again, right? But I would definitely go traditional first as opposed to Messenger first. There's always an exception to the rule, and the exception would be if you're the type of person where you've been able to do this organically and you kill at it organically, then absolutely start with Messenger ads because you're just amplifying what's working. But for most people, it's really— it's, it's, it's a lot harder to convert those people. And like I said, I just found that, at least just for me anyway, the ones that I run, I've run, um, you get traction and then, and then you start to get just really— you really get a lot of like trolls and weird people, you know, commenting. And you don't want this.
Yeah.
So, but everything works. Everything works. Nothing doesn't work. It just depends how fast and, and how expensive it is.
But literally everything Yeah, I like what you said about the omnichannel because we talk about that all the time. And the world we live in, single-channel attribution is practically impossible. I mean, you just can't do it because your customers or potential customers are everywhere. So you have to be everywhere. And to attribute this sale to this channel, you can't do it.
Yeah. So I think with that, we'll wrap up. Amazing insights, Leanna. We really appreciate your time. And your, your insights. Uh, you know, I've known about you for quite a while, and, uh, I've always been very impressed. So I'm thrilled that you were willing to join us today. Uh, what's the best way for people to, to find out more about you? And, and do you have any offers that you want to talk about, or any upcoming events that you are going to be at that you want to promote?
Uh, yeah, sure. So you can find— I'm pretty much Lead Gen Queen everywhere on social media if you guys want to find me there. I love it when people feel free to shout me out, tag me, or whatever. I'm active. IHuman, me, is usually active on those. So you can find me there. I don't know exactly what my speaking gigs are coming up, but I am actually going to be running some workshops in the Vegas area and probably in Salt Lake City, Utah as well for AI and automation. So I'm really excited about that. And another thing that I just launched Okay, this is going to sound a bit weird that we talked all about AI and automation, but I just launched— I'm launching a printed newsletter and it's called the Leanna Letter because it encompasses not just AI and automation, but also things about business and marketing and stuff in this one 20-page newsletter that I mail out every month. And it's just something I always wanted to do and I finally launched it. And I know it sounds weird, we were just— it's very analog, but that's just something that I just launched, which I'm pretty proud of because it's just— I, I always wanted to do it and so now I'm doing it. So, love it. Awesome.
And getting mail is like a novelty now.
Yeah, right? Yeah, because everything's online. Yeah, I love paper. So I'm excited to dive into that and, uh, and ironically talk about AI and automation in it.
So Yeah, absolutely. Well, we're going to wrap this up. So, uh, Leanna, thanks so much for your time. Um, you, you shared information on a lot of really important topics that I know a lot of our listeners, um, and viewers need to, to be aware about. Maybe many of them have been kind of reluctant to dip their toes in. Hopefully we've given you guys some really good strategies and, and ways to approach things. and think about things differently than maybe the little experiment— experiments that you've tried so far. Uh, Ian, with that, I'll let you take us out.
Yeah, thank you again, Leanna. This has been incredible. So much fun chatting with you today. And thank you to our, our viewers, our listeners. Um, please subscribe, smash that subscribe button. Uh, join us next week when we're going to, uh, do a brand new topic. And thank you for, uh, just being here and be sure to spread it with those that know you that can value this, uh, podcast as well. So, um, until next week, folks, keep calm and market on.
Beyond the Hype: Building Real Business Growth with AI and Automation
The world of AI and automation is buzzing, but amidst the noise, it's easy to get lost in the latest tools and trends. This discussion dives deep into what truly matters: building sustainable business growth by focusing on fundamental principles, strategic systems, and the power of the compound effect. We'll explore how to move beyond "AI slop" and create genuine value, from tackling annoying problems to becoming the "mayor" of your local market.
The Foundation: Principles Over Tools
The conversation kicks off with a crucial reminder: business principles, lifestyle principles, and personal development are far more important than focusing solely on specific AI tools. While the allure of the newest gadget is strong, true success lies in understanding the underlying mechanics of how businesses operate and how individuals grow. Instead of trying to "clone yourself" with AI, the focus should be on building robust systems that consistently deliver desired outcomes, such as achieving a target number of leads per day.
The Compound Effect: Small Actions, Big Results
Drawing inspiration from Jeff Olson's "The Slight Edge," the concept of the compound effect is highlighted as a powerful driver of success. This principle illustrates how small, consistent actions – or inactions – can lead to dramatically different outcomes over time. Easy, passive choices can lead to stagnation, while small, deliberate, even difficult, actions can compound into significant positive results.
This idea is beautifully illustrated by the journey into AI and automation. It often begins not with a grand plan, but with a seemingly minor annoyance. One speaker shared how a simple desire to get Zoom recordings to a Virtual Assistant (VA) and posted to a community platform sparked a massive automation journey. This initial hurdle, though frustrating, led to exploring integrations, complex workflows on platforms like Make.com, and ultimately, a deeper understanding of AI's potential. The key takeaway? Start with one thing, even if its immediate utility isn't obvious, to begin your AI and automation journey.
Building Modular Systems: The "Butler Test"
As businesses delve into automation, a common pitfall is creating overly complex, monolithic systems. Drawing from corporate experience, the wisdom of building smaller, modular automations is emphasized. Just as complex machinery is difficult to troubleshoot when it breaks, so too are intricate automation workflows. It's far more efficient to fix a small, self-contained automation than a sprawling, interconnected one.
This leads to the concept of the "Butler Test": the ultimate goal of automation is to anticipate needs before they are even articulated. Imagine a butler who manages your household proactively, offering assistance without being asked. This proactive approach is the ideal for AI and automation, moving beyond reactive tasks to a state where systems anticipate and address needs. This can manifest in practical ways, such as systems that proactively check Facebook ad pacing for underspending or overspending, or scrape landing pages to ensure live event dates are current.
From Annoyance to Insight: Defining Your "Why"
So, where should businesses begin their automation journey? The consensus is clear: address an "annoying problem" or something you simply want to know. While annoying problems are often the easiest to identify, the motivation can also stem from a desire for insight. For instance, a client might want to understand sales pacing and compare it to other patterns as an indicator of funnel performance, not necessarily because it's a pain point, but for strategic understanding.
This insight can then be used to create systems that analyze data, identify patterns, and even suggest strategic actions. The idea is to have AI act as a "junior CMO," analyzing data and offering suggestions for ad campaigns or overall strategy. The beauty of automation is its versatility – it can be built for "whatever you want."
Avoiding "AI Slop": Focus on Outcomes, Not Just Tasks
A significant trap in the AI and automation space is the creation of "AI slop" – content or processes that are easily generated but lack strategic purpose or genuine value. This often stems from a focus on automating tasks rather than achieving desired outcomes. For example, simply cloning oneself to create hundreds of social media posts daily without a clear goal like fame, leads, or sales is inefficient. The question shouldn't be "how to clone myself 30-50 times a day," but rather, "how to get 10-20 leads a day."
Similarly, overly complex personal assistants that involve multiple steps for a simple confirmation are a prime example of task-oriented automation that doesn't move the needle. The explosion of content due to the ease of AI creation, without strategic motivation, contributes to increased noise, making it harder for businesses to stand out. Many businesses automate simply because they're told to or because competitors are, without a clear framework for achieving tangible results.
The Power of Micro-Moves and Clear Objectives
For teams with limited resources, the key to effective AI and automation implementation lies in "micro-moves" and "micro-goals." Before embarking on any project or adopting AI, it's crucial to have clarity on desired outcomes. The current marketing landscape often prioritizes tools over clear objectives, and AI can inadvertently reinforce this. Stepping back to define goals is paramount.
AI PhDs emphasize two key areas: identifying problems for clarity and developing imagination for creativity. Spending time to gain clarity on desired outcomes, even with AI's assistance, is essential. Starting without a clear direction is a waste of time. AI recommendations can be tempting, but they may lead away from the original objective. Therefore, clear outcomes and consistent reminders of those outcomes are vital.
Choosing Your Tools: Zapier, Make.com, and N8N
When it comes to automation platforms, several key players emerge: Zapier, Make.com, and N8N. Zapier is known for its user-friendliness and speed, making it ideal for getting started quickly. Make.com offers a visual workflow builder, and N8N, while more technical, provides significant cost savings, especially for high-volume automations.
The decision to switch from Zapier to N8N often hinges on cost. As Zapier upgrades become more frequent and expensive, migrating high-volume Zaps to N8N can be a strategic move. While N8N has a steeper learning curve, AI's ability to generate code is making it more accessible. The key is to find the right balance between ease of use, functionality, and cost-effectiveness for your specific needs.
The "Super Keen CMO" and Guardrails for AI
The vision of AI extends to creating an intelligent agent that acts like a "super keen CMO" – an AI that detects patterns, decides on the next test, and drives business optimization. This concept involves an AI agent that analyzes data, past events, and current happenings to inform marketing strategies.
However, the question of control arises: how do we set guardrails for such AI agents to prevent them from running "crazy" in a business? The answer lies in a "human in the loop" approach. AI can generate ideas and perform analysis, but human checkpoints and approvals are crucial for quality, consistency, and brand voice. This allows for adjusting AI's autonomy to a comfort level, with low-impact mistakes if AI is still in a draft phase. AI agents are like junior marketers; humans must approve expenditures and next steps.
Content Repurposing: From Staff Meetings to Social Feeds
AI offers a game-changing solution for content repurposing. Starting with existing activities like weekly calls or YouTube channels, AI can extract main insights and generate a wealth of content for various platforms. Transcripts from coaching calls, for instance, can be uploaded to trigger automations that create posts for Facebook, LinkedIn, Instagram, video scripts, and blog posts. AI can even identify workshop ideas, create outlines, order forms, and email sequences.
To maintain a consistent brand voice and avoid AI hallucination, defining context specifically in prompts is key. This includes style guidelines and foundational documents. By providing AI with a "cheat sheet" of knowledge about a person and their teachings, you can guide its output and prevent it from relying on external information or its own memory. Uploading relevant files like books and style samples into AI tools provides crucial context.
The "Dollar a Day" Strategy: Cost-Effective Creative Testing
A highly effective and cost-efficient method for testing creatives is the "dollar a day" strategy, popularized by Dennis Yu. This involves running video or engagement ads on Facebook at the minimum daily spend for each video. The goal is to achieve a 10% post engagement rate, which indicates a high likelihood of success in other campaigns.
Successful videos can then be promoted across other platforms, making a brand visible everywhere, especially locally. This strategy aims to build familiarity and goodwill, becoming the "mayor" of a town by blanketing local platforms with targeted ads. While directly tying engagement campaigns to sales can be challenging, the overall omnichannel results – sales, leads, and spend – should be considered. Consistency is key, and successful creatives can be reused in conversion ads.
Navigating the AI Landscape: Skepticism and Personalization
As AI becomes more integrated into our workflows, a healthy dose of skepticism is warranted. Running AI output through multiple iterations or using different AI models for comparison is recommended. Personalization is also crucial for feeding personality and branding into AI for relevant advice.
Chatbots offer personalization or memory sections where you can input information about your fears, goals, and client types. This can be turned into a "soul document" or custom instructions for the AI, ensuring its mission aligns with your success. The "Digital Integrator," a custom GPT-like program, exemplifies this by storing and filtering data to pull the most crucial information within the context of a person's business.
The Future is Analog and AI-Powered
Interestingly, a game-changing method for creativity involves pen and paper, seemingly the opposite of AI. Initial idea generation using pen and paper, followed by dictation and refinement with AI, can lead to content that sounds more authentic and captures your unique voice. This process, when practiced consistently, can train AI to better capture your voice and has been used to write books, newsletters, and blog post ideas.
Ultimately, the journey into AI and automation is about embracing experimentation, learning through practice, and building from a solid foundation of business principles. By focusing on outcomes, avoiding "AI slop," and strategically leveraging these powerful tools, businesses can unlock new levels of growth and efficiency.