A (2:43)

So was that in like November, like literally the day, like, you know, like it was like last day in November 2022. Right, like, so it was just leading up to the holidays and you're like, all right, I'm going to go on, on this thing. So, so tell us the rest of the story, what else, what happened?

B (2:56)

Yeah, that's pretty much what happened is that I knew, I played with it, I think it came out and then like the next day I played with it and I thought, this is literally going to change everything. It wasn't amazing, but it was way better than anything I had ever tried before. And I thought, well, if this is the worst it's going to get, it's going to just keep getting better, I better make some adjustments now. And within a week I had decided I'm going to pivot. And I started actually one of our offers, it was like $9 a month. I figured if I get paid to learn this, I'm going to actually stick with it and learn it and not just push it off. So I'm going to do that. I made a super, super like low ticket offer so that people could learn with me basically and I just ran with it.

A (3:33)

So tell us what you're doing now.

B (3:34)

So now we show people how to use AI for automations, for basically creating a team of mini thems that can do the work for them so that they're not spending all their time behind the desk. We show people how to use it for automations, how to create their clones, how to really have this team so they can operate like a big dog without having, like a big dog without having, you know, a million team members that they have to pay and have the pressure of paying and everything.

A (3:59)

So when it comes to AI automation, what do you see as some of the big misconceptions that people seem to have?

B (4:06)

I think a lot of times people think, well, I have to be really techy to do this or it's going to be overly tech heavy. And I, I don't have the time to learn how to do that or this is going to be a full time job for them, they're going to have to learn all these things. And it makes sense that they feel that way because new AI tools are coming out all the time. And if you're talking to people who are developers or who have been AI for decades, it can feel really overwhelming. It can feel like tech talk and jargon and you're not really sure even how to get started. That's, I think probably the biggest misconception I see is that it has to be time consuming and it has to be techie or hard.

A (4:39)

So when we think about AI automation done well, because I can probably speak for most of my audience, they're not actively automating, right? Not in a true sense of the imagination. Like what are the upsides, what are the benefits to automation?

B (4:55)

So I think the biggest benefit is that it's like you're having an entire team without needing to hire a team or have the capital to hire a team. I think that's really. A lot of people want to grow, they want to get their message out, they want to help more people, they want to make more money, but they don't have the capital to hire this big team. So then they get stuck doing everything and then there's burnout and it gets really difficult, especially for people that don't have giant teams or don't want to manage a bunch of people. And so automations can do a lot of those mundane tasks that they may have to hire a VA for and even some of the more complex tasks too with some of their client work that they can have AI or have an automation come in and help with. So they are able to save time, they're able to save money because they're not hiring a bunch of people. And then there's that mental load that comes with having people on your team too. I have a few people on my team that are amazing. I love working with them and I still don't want to have to manage a giant team. That's when I think about like, what does my life look like? What does my ideal dream world look like? It's just, it's not having a giant team. But I still want to help more people, I still want to get my message out there. And so having automations really helps you get your message out there, but do it in a way where you're not stuck in the administrative tasks and you're not managing a bunch of people and feeling drained from that time and the energy that it takes to do that.

A (6:13)

Just out of curiosity for yourself and maybe for some of the people you've worked with. What kind of time savings are we talking about? Like help people just wrap their mind around because a lot of people hear AI helps you be more efficient, AI helps you save time, but maybe help us make it a little bit more practical. What does that meant for you?

B (6:32)

So for me personally, it saves me at least 30 hours a month and then on top of that also saves our team, I would say probably anywhere from 10 to 15 hours a month just from the automations that we have. So in total it really is like we've hired another full time employee without needing to have the capital to hire another full time employee.

A (6:48)

And just so I understand, how long did it take you to get to the point where it saved you 30 hours a month? Is this something that took continuous iteration or was it something that pretty rapidly resolved some of your biggest time challenges? I'm just curious.

B (7:05)

Pretty rapidly. Once we sat down and really got into it, when I first started doing it, I just like tinkered with it here and there and I thought, oh well this thing didn't work, I'll come back to it later. And that was really before a lot of these automation tools had their own like AI helper inside of it. Now they have these AI helpers. So if you get stuck you can just be like, hey, how do I do this thing? And it's going to help you along the way. And so even the space between where you're starting and having your first automation done, it takes so much less time now because there's AI helping you and videos and trainings and things like that. And so the time from starting to actually seeing some real change is really, really fast now.

A (7:44)

Okay, so here's what I've heard you say so far. Number one, you don't need to be a techie in order to be able to pull this off. Number two, it doesn't have to be time consuming as you just demonstrated, because these tools seem to be getting incrementally more intelligent and you're not necessarily starting with a blank slate. And the big advantage for a lot of people is going to be regaining time. And I think so many of us, whether you own a business or you work for a business, or you're a hobbyist, you know, like creator who's just trying to do something, hustling on the side. The biggest challenge that so many of us have is not enough time to do all the things that we want to do. And often the things that we want to do are creative tasks, right? Like I want to create more of this or I want to invent this idea that I've always wanted to do or I want to spend a little bit more time doing research. And these are the kinds of things that often end up never happening because we don't have time. And I'm just curious from Your perspective with this 30 hours that you've gained, has it allowed you to kind of play a little bit more and. And to actually talk about that a little bit?

B (8:49)

Yeah. So I write poetry now, and I love it. I mean, that's something that I've wanted to do since I was in third grade. I. I loved writing even when I was a child and really got to do it anymore, really, since high school on, because I was always so busy. And now I actually get to write poetry. And, I mean, like, nobody's actually seen it and hasn't seen the light of day, but it's wonderful because the goal. I feel like it shifted my goal in life from being as productive as possible to living as intentionally and as full of a life as possible. Like, I have realized, I think, for the first time in decades, that my purpose on Earth is not to be more productive. And that has changed so much for me. So poetry. I'm writing that now. I have a book that I'm writing this fiction that's just for fun. I'm getting to spend more time with my family. I bake a lot more. I made Dolly Parton cinnamon rolls or cinnamon bread the other day. I'm getting to kind of be more creative with the food that I'm cooking and stuff like that. I'm getting to work out more. I started wood carving again. It's not good. Like, you can't tell what it is.

A (9:49)

Yeah. So this is fascinating. So what I love about what you just said, and so many people that are afraid of AI because they think it dehumanizes us, is what I'm really hearing you say is it's actually taking things that you don't want to do, and it's doing it for you so that you can do the things you really love. And that's kind of one of those maybe unintended benefits that a lot of people haven't wrapped their brain around of this kinds of things. Because, you know, there used to be days where we would have file clerks. Like, you know, there used to be days when a memo was delivered in a manila envelope and it was delivered by a person to your desk. Okay. If you're of a certain age, you know what I'm talking about? Okay. There were days when the only way you could get something to say someone is through the US Mail. Okay. Like, not, you know, and you have to wait a week for it to show up. Right. So technology does enable certain kind of really mundane tasks to be done with great efficiency so that we can do the Things that we absolutely love doing, which is absolutely fascinating. So, okay, thanks for spending some time with me on that. So let's say everybody who's listening is like, all right, I want me some of that automation. Where do we actually start? Right, because it seems like very advantageous, but it also seems a little overwhelming. So what do you recommend we do first?

B (11:08)

So the tools that I would use, most of them are tools that you've already heard of, you're already using. So we use a spreadsheet. We use make.com or zapier.com I like make.com because they have a robust free version. And then ChatGPT, those are really the main tools that we're going to use. And the first thing I start with is that spreadsheet. So in the spreadsheet we're going to do a time study. A time study is where you write down what you're doing every 15 minutes for four days at least. Ideally you're doing it for a full week, but at least four days so you really get a clear picture of what you're spending your time on. And if you have people on your team, you want them to do it as well once you've done that. And we're talking about everything that you're doing. So in my time sheet, my time study that I do, once a quarter, I'm writing down personal things and then also professional things too. And professional things, I'm writing down the tasks that I did. But also if I found myself procrastinating on something or I found myself feeling frustrated about something, I wrote that down too so that I could kind of identify the things that were slowing me down or the things that I. I found easier. I found myself in flow more. So every 15 minutes I'm writing down, here's what I'm doing during that 15 minute block. And usually when I tell people to do this, they're like, ah, I don't know if I'm gonna have time for that. But really you don't have time not to do it because it's going to allow you to save so much more time in the long run. So it's really important to do that.

A (12:25)

Let me ask some questions about this because I'm curious about this. So like some people do a heck of a lot of things in 15. So like I imagine a spreadsheet, right? Like, help us understand like how this ought to be formulated and what you're talking, like dropdowns for different kinds of work or like, what are we talking about? Like, help me Understand how you would do something like this.

B (12:43)

So you can get really fancy if you want to. Mine is super, super simple. So in column one, I have the dates the first day that I'm doing it. Column two is a second date, column three is the third date, and so on. And then on the left side, every cell is a 15 minutes. So if my day starts at 7:00am, it says 7:00am and then the next one is 7:15, then 7:30, then 7:45, 8, 8:00 clock and so on. And it takes a, like probably five minutes, maybe less to set up. But once you have it set up, you can, you know, use it every quarter or every month or however often you want to do this. But that's what it looks like to get it started.

A (13:15)

And what do we put in the cells? Like that's the part that I'm most curious about. And by the way, that was super helpful. And are you doing this for like an entire week? Because you know how sometimes people do different kinds of tasks on certain days of the week?

B (13:25)

If your days are very different, then yeah, I would do an entire week. For me, I do at least four days. I would love to commit to doing it for a whole week, but anytime I've tried to do that, I find myself tapering off. So that' why I found for me, the sweet spot is four days. Most people are like most of my clients, when I'm working with them, if I tell them a week, they're like, I don't know, what if I forget? But if I tell them four days, they feel like that's doable. So that's usually why I start with four days.

A (13:49)

What are we placing in the cells? And do we have to stop every 15 minutes and fill it out? Like give us a little guidance on that.

B (13:55)

Yeah. So in the cells I'm putting down exactly what I did. So if I sent three emails, I would say sent three emails to and then I would specify, was it a client, was it a student, was it customer service? And then if I'm doing a presentation, what am I actually doing? Am I recording the presentation, am I preparing for it, am I getting ready? Am I doing like posts presentation stuff where I'm doing a social media post? I just try to be as specific as possible. The more specific you are, the more this time study is going to help you really see what's happening. And I know that doing it every 15 minutes maybe feels like a lot, and it kind of is a lot to stop every 15 minutes. But I find if I don't do that. There's so many things that I miss. And so if you are listening to this and you've ever felt like, man, I feel like I did so much today, but I look at my to do list and there's still so much on there. A time study is really going to help you because it's going to help you identify those things that are slipping in that you may not even realize. Like, maybe you took 15 minutes because a customer needed help or one of your team members needed help, and you didn't have that on your to do list and you just kind of forget about it at the end of the day. And so by doing it every 15 minutes, you get a really, really clear picture of exactly what's happening. Not just what you remembered happening, but exactly what's happening as it's happening.

A (15:08)

Okay, so let's say we do this for four days and ideally five. Right? Because that way is work. I mean, presumably working days. Right now we've got this sheet all filled out. What do we do with that?

B (15:21)

So now that you have the sheet filled out, we're going to save it as a CSV file or even as a PDF and upload it into ChatGPT. From there, you're going to ask it to analyze the kind of tasks that you're doing and the things that you're doing the most of. I also like to give it my job description and say, based on this, what are the things that I should take off my plate? I also ask it, what are the things that I can delete? That means these are things that are not really moving the needle forward in my business at all. But maybe I've always done it or it feels uncomfortable to hand it off, but I'm still doing it and I shouldn't be. So what should I delete, what should I delegate? And when I'm looking to delegate, I'm looking at what should I delegate to a team member, a human team member, what can I delegate to AI? And when I look at delegating to AI, I'm thinking using agent mode and ChatGPT, or even having a team member use ChatGPT or another AI tool. And then the last thing is, what can I automate? So we're looking at automate, delegate and delete.

A (16:14)

Okay, so you touched on a bunch of stuff there. Automate, delegate or delete. So you mentioned operator very briefly. Can you go ahead and explain what you meant by that? Because not everybody's going to understand that.

B (16:27)

Yeah. So agent mode in ChatGPT, I think I said, operator, I'm in Agent mode. But Agent mode and chatgpt is you tell it what you want it to do, and then it's almost like a team member comes in and does it for you. So the other day, for example, I needed a sales funnel or an opt in page, I think it was created. And I didn't really want to do it. I didn't really want to pass it to a team member because I probably won't use it too many times. And so I went into Agent mode, which if you're in ChatGPT, you go into that normal chat section. There's a plus sign there. You click on that and then you're going to click Agent mode and that's what's going to give you access to this. So you'll click on Agent mode and you'll say whatever it is that you want it to do. In my case, I said, I want you to log into my GoHighLevel account and then create a new opt in page that has this copy. So I copy and pasted the copy that I needed to go in there. And then I want you to create the form for it, the buttons, the design, and the thank you page. And so I just left it on. I have two screens up. I left it on one of my screens while I worked on something else on the other screen. And while it doesn't go super fast, it's way faster than me doing it myself. And so while I was working on something else, the agent mode was working on creating that opt in page for me.

A (17:35)

Yeah, and folks, my understanding is this used to be known as operator. It was only available in the Pro account, which is like 200 bucks a month. And the agent mode is now available to all the plus users. Right. Which are US$20 a month, people. What's your experience with Agent mode? Is it pretty easy to work with?

B (17:51)

Yeah, you know what? At first I didn't like it because there were so many big promises. They're like, oh, it's going to be amazing. It's going to, you know, replace everybody. And it did not do any of those things. And so when it first came out, I was a little bit disappointed. But the more that I've used it now and a few after a few updates they've had, it's actually become really, really handy for me, especially for little things that I can't quite automate. Like I can't really automate it, you know, uploading a seal or an opt in page. And so to have it Create that opt in page for me is really helpful. It doesn't do some things well, like design terrible. When I had it designed, the opt in page, it was really, really, really ugly. But I didn't have to manually go in myself and copy and paste a bunch of stuff and format it. And so now the design, that's something that we can update pretty quickly and we don't have to do like the mundane copy and paste, copy and paste sort of thing. So that alone saves us a ton of time.

A (18:41)

Okay, when you are uploading this data set into ChatGPT and you're asking it to analyze it, any tips on, like, how we. Because I think the way we ask the question is really important, right. Do you have any tips on how to, like, ask it to analyze the information? Because, like, I would imagine there is some science to that A little bit, Right. Or a little bit of prompting technique, right?

B (19:04)

Yeah. A little finessing.

A (19:06)

Yeah.

B (19:06)

So usually what I'll do is I'll say something like this. I want you to review the file that I uploaded. Specifically, I want you to identify the things that I'm doing in my schedule that are not moving the needle forward, that I can either delete, delegate, or automate. And then I'll say, remember, my goal is to. So I'll say, you know, here's my revenue goal that I want to hit this year. Remember, I don't want to hire a bunch of new people. We have these people on our team. Here's their job description. Here's my job description. Identify the things that we're doing. Like I said that we can automate, delete, or delegate. And then I ask it to group it by the kinds of tasks too. For me, that just makes sense, like how I. How my brain works. To see it grouped by tasks is really helpful. Once it groups it by tasks, I say, now I want you to identify the things and it'll give me a list of the things that we can delegate, delete, and automate.

A (19:55)

And just out of curiosity, and I love the idea that you can delegate to AI, right? I mean, are you specifically asking it to include AI amongst the delegation or are you just telling it like, here's my staff members. I mean, I'm just curious about that.

B (20:11)

So I kind of consider AI to be a staff member at this point. Like, it operates enough like a staff member for us that we just kind of include it in the team. But we do say, here's everybody on our team, and that does include AI for us.

A (20:23)

This blew my mind. New data from the US Chamber of Commerce shows that 85% of small businesses using AI saw an increase in their sales and in their profit. But you likely already understand the benefits of AI because you listen to this show now. Consider next leveling your skills by attending AI Business World. It's our first dedicated AI event. You'll learn from 20 of the world's leading experts, the very same folks you've been learning from on this podcast. Isn't it time to take your AI professional development seriously? Get your tickets today at AIbusinessworld live. I can't wait to meet you there. All right, so we've got all this analysis, right, that's been done after this time tracking and chat. GPT in this case has helped us identify some things that we can automate. What's next?

B (21:15)

So now that we figured that out, we want to figure out what's the most important things to automate. Well, first we want to delete anything that we can delete because that's automatically going to save us some time and some mental energy too. But then from there, I want to look at what are the things that I should automate first. And so when I'm thinking about what to automate first, I'm thinking what's going to have the highest impact on my bottom line and my time and the lowest lift. And when I say lowest lift, I mean what's going to feel the easiest for me to automate? If it's something that's like a super long automation, I'm probably not going to start with that one unless it's super, super impact. And so, for example, an automation that is high impact but low lift is an email automation. So that would be where anytime a new email comes into my inbox, it reads it automatically and it creates a draft reply for me and then notifies me that it's ready to review. This saves me a ton of mental energy. We get a lot of emails and I love it. I love hearing from our clients and our students and sometimes it feels a little overwhelming. And so to have those emails automatically read and have this draft reply created that I just have to make it maybe a tweak or two or even just send out, saves me a ton of mental energy. And so that's an example of something that would be high impact for us because it's saving a ton of time. And we're able to get like, if somebody emails in and they ask about one of our offers, we're able to get that information to them way faster. And it's like A three step automation. So it's really easy and fast to do.

A (22:37)

And when we were prepping for this, you also mentioned that you are using this for your social media and also for some of your marketing of some of your programs. I would love to just kind of have you explain a little bit about what you're doing there just to help people. Because a big chunk of my audience is obviously marketers and social is a big part of what they do. So I'd love to hear kind of how you've also automated some of that.

B (22:58)

Yeah, so I was noticing that I wasn't getting out as much social as I wanted to. And it was frustrating because I was like, I have so many good things to say. And you know, people always refer back to me and say, oh, Brittany, help me with this. Brittany, help me with that. But nobody knew about it because I wasn't getting out enough social. I wasn't sending out as many emails as I wanted to because we just had a lot going on. We have a daughter and soccer practice and all the things. And so I thought, I wonder if there's a way for me to automate this where it still sounds like my voice. Because that was a big concern for me, especially being a former copywriter. I really wanted to make sure it actually sounded like me. And so where we started is this trigger. So a trigger basically means that this is the thing that starts everything else. For me, the trigger is uploading a voice note. So what I do, because I try to fit everything into like the pockets of time that I have, what I do is I'll drop my daughter off at school and then I'll sit in the parking lot for a second so that I can press record on this voice note that I use. It's just like a generic voice note recorder on my Android. But Otter AI is also a really good tool for this. So you press record and then for the next four to five to six minutes, I'm just sharing my thoughts on a topic that I see coming up in AI a lot or question that a client or student asked me. And I'm just sharing my thoughts on it, basically brain dumping into this voice note. And then once I get to the office, which is literally six minutes away from her school, I stop the recording before I get out of the car. I upload that into the Google folder. And because I uploaded it into the Google folder, that triggers everything else. So that creates for us an evergreen email, two Facebook posts, Instagram posts, Instagram Reels, captions, YouTube scripts, LinkedIn article, blog post. And I feel like I'm even missing one or two things. But there's just a lot of content that it creates for us. And it all stems from that voice note, which is great because then I know it's going to sound like me because it literally came from what I was saying and my ideas and my thoughts.

A (24:52)

Ooh, I love this. Hold on a second, we got to talk about this. So, and by the way, if you're on an iPhone, there is a voice app, a memo app kind of thing built into it. So, so basically when you hit record, are you just on the fly discussing an idea with your quote unquote AI or are you just free flowing as a, as a writer, kind of an outline of what it is you think, like, what's the substance of what you're recording? Do you understand what I'm asking?

B (25:18)

Yeah. So I would love to say that there's like an outline that I'm using and stuff like that, but I get distracted really easily and I know that's not usually going to work for me. Like I'm not going to remember as I'm driving, driving. And so for me, what I'll do is I have a list of topics that, you know, questions that people have been asking all in a spreadsheet. And so before I drive off, I just look at one of the topics and I think, oh, I could talk about that today. Really, it's just whatever kind of comes up for me that day. I'm like, yeah, this is something that I can talk about today. And I just brain dump, basically. Word vomit. Anything that's coming up for me as I'm thinking it. For those of you who are external processors, if you have something to say and you talk it out to people, this is a really, really great.

A (25:57)

Oh, yeah, I am a verbal processor. So for me, I'm very excited about this.

B (26:01)

Yeah, it's super helpful. And so you just talk out. It's almost like if you were standing next to somebody or talking to somebody and they asked you that question, what would you say to them?

A (26:09)

Or it's like having a phone call with one of my staffers where I'm driving and I'm just ideating on something, except I'm recording it. Fascinating. Now, what about on the marketing side of things? Because I know you also have automated some of your marketing.

B (26:20)

Yeah. One of the other problems I realized was that I would often forget to talk about the things that I'm adding into our programs. So we do a really good job of taking care of our students and taking care of our clients and really going above and beyond for them. But no one knew it because I kept forgetting to tell people I was great at the, like, delivery side of things, but not always great at the marketing side of things because that to me feels a little bit more uncomfortable. I'm like, oh, I feel braggy or I feel, you know, whatever.

A (26:45)

Yeah, you don't want to be self promotional.

B (26:47)

Yeah, there's like this friction that was coming up for me and so I thought, how can we remove some of this friction? And for me that was creating an automation so that at the end of the day I say, here's the new training I added and here's how it helped people, or here's what somebody said about how our program is helping them. And so from there it creates, you know, all the content for that. It creates an email, it creates a blog post, it creates everything that we would need. And so now all we have to do is upload that. And if there's any friction there for me on that, then I have a team member, a human team member that uploads that for me so that there's not this friction anymore.

A (27:20)

So what's the trigger on that one? Is it another voice recording or is it written or what is it exactly?

B (27:25)

This one's a form. So we fill out a form at the end of the day and usually at the end of the day I'm all talked out because I usually do multiple presentations every single day. And so at the end of the day I'm all talked out, so I just fill out a form for that one.

A (27:37)

Fascinating. Okay, so thank you for sharing those examples because it's super helpful for people to like because for no, people have never done this before. They can't even imagine how it would work. But all three of these examples are super eye opening in. Okay, so now let's get to the technical side of it. So here's what we have so far. We have a time audit, right? And then we have a analysis of our time and we've interacted with AI and we've identified some things that we should stop and some things that we should delegate, some to AI, some to other people, and we've also identified things to automate and we're really zooming in on the automation. We talked about three different examples. Email automation, social media stuff through a voice memo, and then filling out a form to generate some marketing emails and such that you could use to send out to your list or whatever.

B (28:25)

Right.

A (28:26)

So now we're at the point where like, okay, we've got a lot of great, like front end of this, but now what's the back end that makes all this stuff connect? You've already mentioned some tools, but I would love to just kind of like help people understand like how to connect all this stuff together.

B (28:40)

Yeah, of course. So we're going to start in make.com and like I said earlier, I like using that tool, that automation tool, because they have a really robust free option so that even you're just tinkering around or you want to just kind of try it out, you can try it out for free or very low cost. And so we go into make.com and we're going to set up a few things. The first thing we need to add is that trigger. And in this case the trigger is going to be a Google Drive module. And a module is just the first step. Basically modules are steps. And so the Google Drive module is going to be to watch when a new file comes into the drive. The next thing is going to be to download that file and that's also going to be from Google Drive. Once we do that, we want ChatGPT to pull the transcript. So the next step or the next module is going to be pull transcript from ChatGPT. Now in make, it's called OpenAI, which is like the parent of ChatGPT. And this is a place where sometimes people get stuck. With OpenAI, you're going to need an API key. An API key basically allows you to connect one tool with another tool. And so you'll go to platform.OpenAI.com and you'll add a little bit of money in there. You won't need much. I think $20 usually lasts me like six months. And I use a lot of automations with ChatGPT, so it won't cost you very much at all. And so you'll have the API key that you'll get from there. Once you have that, you'll upload it into make and then that's really what allows everything to connect. Once you've done it once, you don't have to keep doing it. So that's exciting. And then from there we're going to have ChatGPT review the transcript and then it's going to start to create the content. I usually start with the biggest piece of content first, so the blog post or the LinkedIn article, because then everything else can pull from those two instead of always reviewing the entire transcript to create each new thing.

A (30:26)

Okay, cool. I have a million questions which, which is which is kind of normal. Okay, first of all, we're using make.comhelp everybody understand. You said that there's a free version of make.com but, but I would imagine at a certain point you gotta pay for it. Right? So how does that work? Just explain like to everybody like where the money comes in, just so they understand that.

B (30:46)

Yeah, so you pay monthly per a certain amount of credits. So for example, for $10 a month you get 10,000 credits. So when you're starting on the free version, I think it's like a thousand credits that you get at the beginning, which is enough to really test it out and see if you like it. But 10,000 credits should last you a really good long time. You can do probably about a whole month on 10,000 credits with this one automation. Unless you're sending voice notes to yourself maybe every then, then maybe you need a higher Tier. But for $10 a month it's 10,000 credits. But you can even start it on that free trial and get, you know, like I said, a thousand or a few thousand credits to get started.

A (31:20)

And is every little action counted as a credit and some multiple credits, Is that generally how it works? Depending on what kind of action you're doing?

B (31:26)

Yep, exactly. So this one, if I have 15 different steps in it, it's going to be 15 credits.

A (31:32)

Okay, cool. Now as far as makes, my understanding is for those that want to go beyond ChatGPT, you can choose Claude or Google Gemini. Right. If you're on Google platform and you love Google, you could just stick with Google. Is that true? Like, and have you messed around with the other models? Because I know I love Claude as a writer and you're a copywriter as well. So are you using Chat GPT for everything or have you messed around with Claude at all?

B (31:57)

I do use Claude occasionally. I prefer chatgpt for most things only because most of my students and clients prefer that and it's generally a little bit ahead of the game and it is can do a little bit more. So for me I usually do that with make. The other reason I really like it is because it's easy to share the automations once you've created them. And so I share it with all my students. So that's why I usually stick with Chat GPT only because most of them are using it and it's easier for them to import it in for those.

A (32:23)

Folks that work for a company where they have a bunch of staff like my company. And by the way, we actually have a make automation similar to what you're talking about this podcast. When I'm done recording it, I export all the files, which are really big wav files, into a folder along with my notes, and it triggers an automation where it emails three different people and then the humans take over. It's just a very, very simple automation, and that just allows everybody to get notified the second stuff happens. But, you know, obviously there's so much more we can do with it. Like, I would imagine, like, is it complicated? Because my team has told me it's not super intuitive sometimes to set these things up in the beginning. Beginning. Maybe it's more now than it was back when we set them up. But, like, talk to me a little bit about, like, I know you said you don't have to be technical, but some people do find it a little confusing. What's your thoughts on that?

B (33:13)

When I first started using it, I felt a little bit lost, but that was, I think, about two years ago now. It has improved so much in those last two years, and it's become so much easier to use in the last two years that even if you're just starting out today and you have no automation background, you could still get started. And one of the reasons is because in the bottom right corner it says beta A and if you click on that, you can basically say, here's the automation or here's what I want it to do and here's the tools that I want you to use, and it will actually set it up for you. Now, there may be a few things you need to edit or change, but by telling it the outcome you want and telling it the tools you want to use, it's going to create most of it for you. So instead of starting from a blank screen, starting from scratch, you have most of it already created and you just have to tweak a few things.

A (34:01)

So you mentioned email automation. I would love you to kind of explain how that works. Like how the heck do you get a draft message in your inbox? Because I think people are. You don't have to get into like the super technical weeds of it, but I'm curious about how you get there. Because it's one thing to drop a file into it, it's another thing to give it access to your Gmail account, right?

B (34:20)

Yeah. So I start with a trigger from Gmail and it basically says anytime a new email comes in that meets these criteria. So if, like a. If I get an email from PayPal, it doesn't need to reply to PayPal. If I get a marketing email, it doesn't need to reply to those. And so I'll say if it meets these criteria, then I want you to. Now the next step is ChatGPT. So ChatGPT is going to review that email. It's going to create a response and in the directions or the prompt to ChatGPT I say, you know, here's my brand voice, here's how, how I normally respond to stuff. And then after that it's going to create a draft reply and that's it. It's three steps. You could also add in a notification to yourself in Slack or Telegram or WhatsApp where it says, hey, there's a new email. Or even at the end of the day, hey, don't forget to check, you know, your draft folder. But it's three or four steps depending on if you want that last piece in there or not.

A (35:13)

So then when you open up your email, you just have a bunch of draft messages in there waiting for you to review basically.

B (35:17)

Exactly. So it literally used to take me an hour to review emails or my team to review emails, but with this autom we can have it done in like 10, 15 minutes.

A (35:27)

And did you say ChatGPT is actually reviewing its own drafts to make sure they're like legit or does it just craft them? I'm curious, like, can you create one kind of task to like write the thing and another one to cross check it? I mean, is that, do you do that kind of stuff?

B (35:41)

Yeah. So if you were doing that after that initial ChatGPT module where it's reviewing or it's like reviewing the email and creating the draft, you would add one more step with ChatGPT and that one would say something like review and then you'd refer back to the chat GPT draft, review that draft and identify where you usually. I'll say identify where you missed the mark. That usually seems to get it on the direction I want it to go right away. But see where you missed the mark and you can even give it a rubric of does it start with the person's name or does it have this in it? Does it have my sign off? What I would do if you wanted to have it review is I would put in some of your emails, like your, the emails you replied to and how you replied to them into ChatGPT. Not an automation, just regular ChatGPT. And then I would say review how I wrote these emails or how I replied to these emails emails and come up with an SOP and a rubric and so then it would give you the SOP or standard operating procedure. And so you'd have like a step by step, here's what it should look for and then a rubric that you can just upload into that automation so that it does what you want it to do and reviews how you want to review every single time.

A (36:45)

A lot of people that are listening have played around with Chat GPT projects or chatgpt custom GPTs or cloud projects and they understand the concept of like giving it a system level instruction, but they also know that they typically have the option to add examples and data sources. How does that work from the automation side of things? Because in this case we're using the API, right? And we don't have the benefit of like the learning knowledge and all the kind of stuff you have when you just use ChatGPT. So talk to us a little bit about like that because that seems like it's a unique challenge that comes with automation.

B (37:22)

I find that giving examples of emails doesn't do as well as like a rubric would. So giving it an example, what I find usually happens with automation is that it doesn't understand you don't want it to use the example as like content or context. Even as many times as I told it. I don't want you to use this as content, I don't want you to use this as context. Look at the format or look at the style. It doesn't get it. It gets it sometimes, but it doesn't get it 100% of the time. Whereas if I'm using a rubric or an sop, it gets it way more consistently. And so that's why I lean more towards rubric or SOP instead of giving it an example example.

A (37:58)

So when we talk about rubric and SOP standard operating procedure, does this mean we're just giving it a monster system instruction for lack of better words? Or does it mean that you can attach data somehow to like the email writer part of the process? With ChatGPT you have the instruction set that you've given it and then you have all these, this rubric or the standard operating procedure, like help people understand how those things connect together. When it goes across into OpenAI, I.

B (38:27)

Don'T want to make it super long. So the SOP and the rubric that I give it are really short. And so what I'll have it do is take the original SOPA that it gives me and I'll ask it to summarize it and still give me those step by step instructions, but make it as clear and to the point and direct as possible, so that we have maybe four to seven sentences for the standard operating procedures and then four to seven sentences for the rubric, and that's it. I don't want it to be something that's super, super long, mostly because if something changes or I want to make an update, I find that sometimes people won't read through the whole thing if it's really long. And then when it's not giving them what they want, they're like, oh, I'm so frustrated. Usually it's just because they forgot to remove something in the old instructions or remove something from the old rubric and put in the new one or the new pieces that they need to know or that it needs to know. And so by making it short and sweet and to the point and really direct, it's easier to make updates when you need to make updates. Needs.

A (39:21)

Do you feel like the automation side of it is evolving? Like, you know how like custom GPT seem to be like getting really sophisticated in what they can do, right. Do you feel like the automation side of it is equally as sophisticated? You know, like a custom GPT can do research and all these kinds of things and has all these cool things built into it. Do you feel like the stuff going over the API, which is effectively what we're talking about here with make, do you feel like it's getting strong or if not even stronger than what you can do with custom GPTs?

B (39:49)

Oh, yeah, for sure. And I think the reason is because now we're adding in more layers to this. So now it's not just automation, now it's you can also add agents which will take action on your behalf for you. So instead of just, let's go back to the email example. Instead of reading the email, creating the response, checking its response, and then putting it in the draft folder, it can also send you a slack message and say, hey, here's the response. Do you want us to send? Send it. All you have to do is click yes or no and then it will send it for you and even follow up with them if you wanted to. And so what I'm seeing with automations is that they're getting a lot more, not necessarily complex, but a lot more helpful, I guess is the word there. It really is becoming like an AIT member.

A (40:32)

Let's talk a little bit about the human side of this for our company, our staff, because there might be some resistance a little bit to automating some of the tasks that Judy or Sam or whatever ever. Do you know what I mean? So do you have any tips on, like, how to, how to get them to embrace this kind of stuff?

B (40:48)

Yeah, I think it can be really scary for some of our team members because they see, well, they probably see some people online talking about how, oh, you can replace your team members. And they might even be worrying about that, or they may be thinking, you know, how do I bring enough value to the table that they still want to have me on the team? And so for me, what we do is, first of all, we want to sell them on the value of, of not only what they have to offer, but also it's almost like this power couple, them and AI together. For me, I don't always just want AI, and I also want the efficiency of AI. So for me, I'm looking at how can I combine the power and the personality and all of that with a human team member and with AI? And so looking at, I look at their enneagram, I look at different personality tests they've taken to see how to best approach it with them. And I use ChatGPT to help me kind of identify what are the things that I need to touch on so that they feel valued, so that they know that their position here is safe, Safe. And then I talk about how it's going to help all of us, really, with our team members when we have, like, the mundane things off their plate, they're able to also go into more of the creative things. And what I found is that when my people are in their zone of genius and I'm able to be in my zone of genius, like we're unstoppable. But a lot of times we just get so caught up in the everyday, boring, mundane stuff that has to get done. But now AI can do that, and so they're really able to thrive a lot more and think of ideas and really own the ideas they're coming up with and feel creative versus feeling like they're stuck in the random clicks that they have to do.

A (42:16)

You know, Brittany, this has been really fascinating and I know some people are going to want to connect with you on the socials and, or go directly and maybe join whatever communities you've got going on. Where do you want to send them? If they're interested in interacting with you on the socials or if they, they want to potentially work with you. Where would you want to send them?

B (42:36)

Yeah, absolutely. On Instagram, you can find us at Women Thrive with AI. And then on YouTube, it's Brittany Long. And I think the little at sign it's Women Thrive with AI for that as well. And so then we have our membership automations on demand. If you want those automations done for.

A (42:50)

You, if they want to work with you, do they just go to the socials or do you want to send any particular place?

B (42:55)

Yeah, you can go to aicoachondemand.com and that's going to take you to the automations information.

A (43:00)

Brittany, thank you so much for sharing your insights with us today.

B (43:03)

Thanks so much for having me.

A (43:04)

If you missed anything, we took all the notes for you over@socialmediaexaminer.com a79 be sure to follow this show on your favorite podcasting app. And if you've been a listener for a little while, we would love a review. And we'd also love it if you would share it with your friends. You can tag me on Facebook, LinkedIn and or X. Also check out our other shows, the Social Media Market Marketing Podcast and the Social Media Marketing Talk Show. This brings us to the end of the AI Explored Podcast. I'm your host, Michael Stelzner. I'll be back with you next week. I hope you make the best out of your day and may AI help you become more successful. The AI Explored Podcast is a production.

B (43:44)

Of Social Media Examiner.

A (43:48)

Get your tickets to AI business world right now by visiting AIbusinessworld live.