Applying AI to Marketing and Business: Lessons From the Trenches

Before we get started, I want to ask you something. Are you tired of feeling like AI is moving faster than you can keep up? Because I know what that feels like. And I hear it all the time. Marketers keep telling me they're drowning in all the new tools and updates and features and must-try frameworks. And, and they got to do all this while they're also doing their job and trying to run their business. And that's why the AI Business Society was created. It's your personal curation team for AI. We tell you what deserves your attention and what you can safely ignore. And inside you get live trainings from leading AI practitioners and interactive skill building sessions, bite-sized learning libraries when you just have a few minutes, and a community of incredible marketers who are just like you. Hundreds of them are on the inside, and you can join right now and lock in an incredible special sale we've got going on. Visit socialmediaexaminer.com/ai. Give it a shot. You got a 30-day money-back guarantee, so you've got nothing to lose. Go to socialmediaexaminer.com/ai and let's now jump back into today's show. Hello, hello, hello. Thank you so much for joining me. I am your host, Michael Stelzner, and this is the podcast for marketers, creators, and entrepreneurs who want to know how to put AI to work. Today, we're going to explore lessons we've learned here at Social Media Examiner, applying AI to our marketing and business over the last 2 and a half years. I'm joined by Allie Kelly, our Director of Membership. She's worked for Social Media Examiner for more than a decade. Allie and I have worked together to bring our company, Social Media Examiner, to the AI frontier to bring it into the future of our company. Allie, welcome to the show. How you doing today?

Excited to be here. Thanks, Mike.

Awesome to have you. Today we plan to share how we use AI to curate news, and it's going to be something that's going to blow your mind. It's very unusual what we're doing. We're going to talk about how we use it to improve our marketing. And when I say improve our marketing, I mean actually create better converting landing pages and all sorts of fun stuff. And we're also going to talk about how we use it to take one form of content and multifaceted into lots of different kinds of content. So if you're interested in learning more about that, stick with us. Now, first, I want to share a little story. Like many of you, AI has been a journey for me personally. It's been a multi-year journey. I have been on this train ever since ChatGPT dropped in November 30th of 2022, and we're coming up on 4 years. But 4 years for me, and I don't know about you, Allie, it feels like it's been a decade, right? Because this innovation change— I used to say that I thought social media was the fastest changing industry in the world, but Allie, would you agree that AI makes it look like child's play?

Uh, that's a fair statement.

Yeah, for sure. And what I've been doing and what I've been trying to do with our company is to constantly learn constantly apply AI to the work that we do, and then constantly refine it, and then repeat and repeat and repeat. Learn, apply, refine, learn, apply, refine. And it is a crazy journey. First, what I did many years ago is I decided to set an expectation that we at Social Media Examiner use AI to improve our work. And it was very straightforward. Use AI to improve our work, not use AI to replace people, but to make us better people. And what we did was we set that as a company expectation, and then we actually trained multiple times our staff on how to use the tools and the policies and the expectations and what was and what was not acceptable. That was kind of the groundwork. And then we actually made the bold effort of developing content, lots of content and actual products all around this artificial intelligence concept. We launched the AI Explored podcast. 115 weeks ago, which feels like forever. We published our second annual AI marketing industry report back in, October of 2025. And that's coming up on 3 years, which is kind of crazy. And of course, we also launched the AI Business Society more than 2 years ago, which Ali heads up. And we also recently launched AI Business World, which was our conference within a conference at Social Media Marketing World. So we We have eaten our own dog food is what I'd like to say, right? We have actually decided that we're not going to just apply this stuff in every asset, every aspect of how we do what we do here at Social Media Examiner, but we're actually going to create content. We're going to create valuable content and we're going to actually create actual products around this to usher in marketers, creators, and entrepreneurs into this world. And where we sit right now, I'm recording this. Literally on Friday, June 12th, this is the day that Elon Musk's company, Starlink, or what's it called?

I can't remember.

What is it?

SpaceX.

Oh yeah, SpaceX. SpaceX launched, right? Under— and it's the largest IPO in the history of the world. And it's really all around artificial intelligence. People are piling in and putting money behind this because they believe it's the future. And of course, Pretty soon we're going to see the same thing with Anthropic and the same thing with ChatGPT, OpenAI, the company behind ChatGPT. So we are at an inflection point right now, my friends. We're at an inflection point in the history of the world that has never happened in my lifetime. And I've been through a lot of change. I've been through the internet, I've been through social media, and now we're entering into this age of AI. So as a result, I thought now would be a really good time to talk about what we've been doing so that everyone who's listening can get inspired and hopefully apply some of this stuff that we're about to talk about today to your work. And first of all, I want to share a little story with you. Another story. Literally, it was about 2 days ago that I had this crazy idea to actually do this special podcast episode. And the first thing I did was I went into Claude and I use Claude projects and Claude projects are really powerful thing inside of Claude. And I've trained up Claude projects on all sorts of different things. And what I did is I said, hey, I'm thinking about doing a special podcast episode. And what I would like you to do is take a look at all of the threads that I've had inside of this project, because I want this to be a behind-the-scenes episode. And can you look through all of the history and can you help me maybe come up with some things that you think would be worthy of me talking about on the podcast. Feel free to ask me some clarifying questions. And what it did was it said, okay, what about this? What about this? Boom, boom, boom. And then all of a sudden it looked through dozens and dozens of chats that have happened over the last couple of months, and it was able to identify some really cool stuff that we're about to talk about today. So, um, what's really fascinating about this is that if I was to do this another way, here's how it would go. I'd have to have phone calls, a lot of phone calls, and I'd have to say, hey, what do you think about this? Or, hey, what do you think about this? I'd have to spend hours racking my brain trying to figure out what were all the things that we've done that would make for a good podcast. Instead, I had Claude do this, and then I hopped on a call with you, Allie, right? And we talked it through, and about an hour, we had a pretty good outline. Isn't that kind of how it went down?

Yeah, my favorite part is that I put in a similar prompt to yours, and it gave me the same top 2 suggestions. So Definitely all on the same page.

Very cool. Okay, so let's move on to the cool stuff that you're doing specifically with AI. And this is really this AI news curation thing. I know you've got some really exciting things to share here.

Yeah, yeah. So, you know, as they, as they say with AI, the way to really get the most out of it is to look at the things that you do every single day or your repeat processes the things that take most of your time. And one of those things for me is curating news for our membership. And this is one of our core deliverables. We want to filter out all of the AI news that marketers either don't care about or just have no, you know, no effect on what— how they use AI. Every day in their work. So we don't care about the politics or the mergers and acquisitions and the IPOs and all the financial stuff in the data centers. Like we don't, I mean, it's all important for other things, but it's not important for how marketers use AI. And that's the tricky thing. If any one person goes out there and tries to learn what's going on with AI right now and the things that are most pertinent to them, There's so much stuff out there to filter through. So we want to try to be that filter for them. Um, it's been a challenge because, uh, there's A, there's so much, um, but B, because our criteria for what's important is so very specific. Um, I've had trouble with finding, you know, the, the right way to do this. You know, I want to make sure that our members are getting what's most important. So I've tried just regular prompting to do searches. I've tried prompting to check specific websites. I've tried scheduled tasks through a few different tools. I have tried using like Google Opal, you know, which I won't go into, but, you know, it's like a multi-step process and it's just been a real struggle. I've tried Workspace Studio to try to automate this, to have it actually check my emails where I get the newsletters. Despite all of my best attempts at tweaking the prompts and making it as clear as possible, just that those normal straightforward ways of gathering information were just not doing the trick. Um, so I basically had to find a way to do this, um, in a much more advanced fashion. Um, and so I turned to, um, actually first I turned to Manus. I was inspired by one of the speakers from our training sessions within the AI Society, where she had talked about how she was using Manus and how she had developed these really cool-looking interactive dashboards. So I was like, hey, let's give this a shot because that looks like a way to do this that could work for me. I got into Manus. I actually developed a whole very sophisticated prompt in Claude first and then took that over into Mana so that the first try, you know, didn't waste any tokens and all of that stuff. It gave me something that was moderately okay. I needed to upgrade to a plan to use the more advanced models. I got something that was like a semi-working prototype. And I realized I was burning through tokens like nobody's business.

And real quick, before you go too much further, like, explain to people what you were trying to build, because I got a video from her, like, I think a week ago. Like, she was so giddy. She sent me a Noco— plug for Noco. She's like, you're not gonna believe this. So, like, explain what it is you were trying to achieve, because it's pretty impressive.

Yeah, yeah. So, um, it looks like a dashboard, uh, but Uh, it's got settings that, that you can customize where you tell it what sources you want it to check. So I give it individual, uh, websites or RSS feeds. Um, side note, the RSS feeds are a way more effective way to gather information. Um, and then I tell it what my curation criteria is. I have separate settings for what the banned topics are, all that stuff I just mentioned a minute ago. Um, uh, it has like custom tagging. So, uh, once I get all the settings all set up and I tell it exactly what I want, I do like a, uh, a news fetch basically, and then it processes and goes and checks all of those sources. It filters out, uh, the stuff we don't want, and it's supposed to— that Some of this has needed tweaking in the prompting. It eliminates the duplicates, which is a big problem because the same story is being, you know, published in multiple places. So once it gathers all the stories, like from the last 24 hours, it gives me a big list. And then, so I've got a headline, I've got a like 1 or 2 sentence summary. And then I can tell it whether I want to accept that story and work with it. I can defer it and then it'll go into another little section I can go back to, or I can decline it. And if I decline the story, it pops up and lets me enter a reason why. And all of those reasons why are accumulated and collected every 2 weeks and then allows me to refine the curation criteria moving forward. So that over time, um, it becomes a perfect process.

Now, just for the record, did you have this vision or did Claude and you together kind of evolve this sucker before you went into this Manus tool and then ultimately to the next tool?

I had like 80% of the vision. Um, and then as I started using it, I started realizing, you know what, it would be really helpful if I could like archive this from this spot instead of having it live here forever and, you know, little things like that.

Yep. Okay, cool. Keep going with the story. So you had a problem though with Madness, right?

So for the stories that I accept, they go into a new tab. And from there, actually back up for a second, back in the settings, one of the settings I set up are templates. So I can create a template to create a post. In our group about the news, I can create a template to, um, help summarize what the news is about. I mean, the templates are basically prompts. So when I have these accepted stories, then I can go and select which types of content I want to apply to that story, and it will write a post for me using the template prompt. Um, and then I have a post that I can go and use in the community.

So— Real quick, let me ask you this. How many hours, 'cause everybody's gonna be curious about this, did you invest to get to the point where at least you had what you thought was a working prototype in Manus before you figured out what you were gonna do next?

Hours in Manus were maybe about an hour and a half before it really went through so many tokens that I realized it was not sustainable.

Then what about the work before Manus?

Like before Manus?

Yeah, all the work to— you said you and Claude worked on this for how many hours was that before you went from Claude to Manus?

That was like 10 minutes.

So you're talking only about an hour and a half so far. Yeah. Okay. This should blow people's minds. Okay, keep going.

So Manus allows you to put the code from your project into GitHub. So I took the code in GitHub, I duplicated it, and I went out in search of a tool where I could host this that wasn't through Manus that would allow me to use Claude code instead of Manus. I, you know, watched a couple videos and I narrowed it down to a new set of tools. I don't know if you want me to get into all that, but the long story short is that I found a new place to host. A new tool for like the database side of things, and GitHub for the, the code repository. Claude Code can tie into all those things. And I basically told it, I'm moving all this from Manus. I need you to help me clean up the code so it works with these other tools. And then it actually started noticing where there were things that needed fixing. And helped me get it to a place that works right now. And then just the other day when Claude wrote out the new Fable 5 that we get to play with for a short period of time, I had it debug the whole thing. And now, I mean, it's like 95% done. So I'm very excited.

How much time you think this is going to end up saving you when this is all said and done?

And once it works, once it works, the big thing, because previously I was having to go through like 8 different newsletters every day, like skim through each one with my own actual eyeballs. And, um, I can get distracted sometimes. So like that would take some time. It would probably save at least an hour a day off my week.

Okay. So, so when we do the math on this, an hour a day. Let's just say that's 20 hours a month, right? And let's say that's, you know, um, 140 hours a year, right? And that's a lot, right? Can you do 20 times 12? So, and when you start to think about how much time now this is going to enable for you, my guess is, what do you— if once this works, you're going to be able to do this in minutes, right?

Yeah, yeah. And, um, you know, the only cost Now that it's— Manus was going to cost us, you know, at minimum $20 a month. Um, with the current setup, uh, which may or may not change, but with the current setup, um, I'm on free plans of these tools. And the only cost is the cost of the AI usage to fetch the stories and create the content, which right now is working out to be about like 30 cents a day.

Did you ever think that you would effectively be developing software?

I knew I wanted to. I am just now to the point where I think that it's actually doable and I can do more of it.

Very, very cool. All right, anything else on that before I go to the next thing?

I just want to say for anyone who has dreams of doing this kind of stuff, I didn't know how to do any of this. My, my coding history was a GeoCities page in high school. And, you know, with a little bit of HTML, you don't need to know how to code. You just need to know how to direct the AI to tell you where to go and what to do. And anybody can do this.

It's very, very exciting. And I am personally very interested in this. Anything that I think, like, here's how we have to think about this. Like, Rory Vaden wrote a book called Take the Stairs, I think it was called. And the idea he was talking about, like, when you train other people, he said that you should spend up to 20 times the amount of effort it takes you to do it yourself, and then you're free forever. But when you train AI, it doesn't take 20 times the effort because what Allie is telling us is it's maybe taken 5 times the effort when she's all in, maybe 6 times the effort, right? And it's now going to be done at a level and at a quality standard that is way better than anything we could have imagined before. And that's going to make possible other great things for Allie, for example, improving the user experience inside of our AI Business Society or doing other kinds of trainings and stuff like that. So I'm very excited about that. Okay. I'm really excited to talk about what's next. So we're talking now about how to use AI to improve your conversion rate on your sales page. So what we wanted was to improve, you know, the likelihood that someone would turn into a customer for the particular product, the AI Business Society. Now, here's what happened. We launched this product more than 2 years ago. And back then, 2 years ago was a very different world in the world of AI, right? Back then, everybody was like, okay, how do I get started? That was really the big thing. And the questions were like, is this going to be a big threat for my, my job? Maybe. I mean, it's obviously a bigger question now, but everything back then was like, how do I get started? How do I position myself as the AI expert in my company? These things are still there, but back then these were very early adopters. Here we're now in kind of the mid or peak stage of AI, and now the questions are very different. The questions now are, how do I master AI? Not how do I get started, right? How do I make AI sound like me? How do I make it sound authentic? How do I create workflows? How do I create like, What Ali was just talking about, you know, like a, like a something that literally can do beginning to end and maybe even do better than what I've done. Those are the kinds of questions people are asking now. In addition, we're also in an era where selling training alone doesn't work as well. We've seen a lot of our friends who were in the course industry effectively abandon that space, and there's a reason why because now information has been commoditized. You can get access to whatever you want on YouTube and obviously with all the large language models. So it's changing kind of what people are willing to pay for. And what people now desire is access, right? They don't want information. They want access. They want access to people. They want access to people that are like them. They want access to experts. They want experiences. They want live things. This is kind of what now is valued more than anything else because of AI. So given all that and the, the through line of AI just rapidly changing, I knew that our messaging was completely off on our sales page because it was frankly more than 2 years old. So some of you that are listening, you might be like, oh yeah, you know, I have the same problem. My sales page, I haven't updated it at all in years and I've gotta take another look at it. So what I did was I used AI to help me reposition our product. And just for the record, I have a formal background as a copywriter. Back in the day, I wrote a book on copywriting and it was my career before I started Social Media Examiner. And I've always said this and I'm gonna say it again, the best way to use AI is to use AI in your domain expertise. In the area that you have the most experience in. And the reason why is because you uniquely are able to direct the AI. You're able to tell it what is good and what is bad. You have that discerning eye because you know, because of all your years of experience, that this is an area where I can actually direct someone. For example, if you were to hire someone that works for you, you could really train them up as an intern, right? Or apprentice or whatever. The same thing is possible with AI. When you direct AI, AI actually becomes really valuable because it actually enables you to more quickly do things. You have the eye to know what is good and what is not good. So with that lens, as my background as a copywriter, here's what we did. First of all, we have a lot of data because we publish reports, and I already mentioned we have multiple reports and we do reports all the time. So what we were able to do was segment out our customer base and look at that particular customer base and see what are they actually interested in now, because we've been surveying them twice a year for years and we kind of knew what the hot topics were. In addition, we were able to scrape all the data inside of our community. So if you think about it, if you belong to a community, all those discussions are data points that can also be used. And in addition, Ali has the benefit of actually talking to customers. Right, Ali? So like you're in these meetups and you're hearing from them and we've got the comments after our trainings, all that data. And of course we can also talk to them in person, right? Ali, we were just on the ground at Social Media Marketing World slash AI Business World, right? So we had a chance to gather all that data. So when you take all the data, which some of this is actually written data, right? In the case of surveys and community discussions and in-person data, and then you, you basically do something with that data. It's really powerful. So, one of the first things we did with the data was we redefined who our ideal target audience is because it had shifted. So we know, for example, it's 71% female. We know 66% of them are over the age of 40. That's really important. Okay. We also know that she's no longer looking for getting started. She wants to know how to learn how to put AI automations to work. And workflows to work that can save her a lot of time. So with this new kind of knowledge, and I'm just giving you a sample of it, uh, we wrote up a really detailed description of who that target audience is, what their desired outcomes are, what they want to learn, maybe a little bit about what their fears are. All that information was basically put into our Claude project that we already had, and we re-updated new information inside the Claude project. That was the first part. The second thing is we focused on positioning and we did deep research on our competitors. And I specifically commissioned you, Allie, right, to go out and do that. And did you use Gemini and ChatGPT or what were you using to do the deep research?

I used Claude for the meta prompting. So I explained what the project was and what I wanted to do and asked it to give me the prompt to give the deep research tool, which was going to be Gemini.

Wait, real quick, we got to pause here. We have, we have some training specifically on deep research that we've either just done or is about to come up. And this is really important because the way you prompt a deep research thing is going to have a huge bearing on how you get the results. So your prompt was probably massive, wasn't it?

Well, it was funny because I told it I wanted one prompt and it broke it up into three and it told me to to do this in 3 phases. So we actually ended up doing a broad deep research to find all the people out there and then we narrowed it down and we did a detailed deep research on specific people we had identified. And then in phase 3, we did the more of the comparison and to help us, you know, revamp the sales page.

Okay. So when you mean people, you mean competitors, we should be very clear, right?

Correct. Yes.

Yeah. So, so this phase of the research was how are our competitors positioning themselves right now? Right. So the first research, uh, just went out and found everybody. Is that right?

Yeah, it went and I gave it some criteria and then it went and basically gave me a summary of all the people that it had found. And then I narrowed it down to about 5 or 6. That I thought were the, you know, the most likely actual competitors.

And then the second research—

the second research went and dug deeper on those individual competitors, specifically looking at things like what are—

what's their offering, how are they positioning, who is that target audience they're going after. And it could discern that obviously by reading the sales pages of all these things. And then the last piece of research Did what exactly?

It took what we learned about each of those competitors and helped to analyze— also analyze our current sales page and told us the ways that we could reposition ourselves and reword things to, to be more effective with our sales.

Excellent. Okay, so, so first of all, I hope you all are understanding, like, this research thing alone is super, super, super, super valuable. Okay, so, um, what ended up happening was I got something from Ali, which was a lot— a big old document, right? And not everything in that document I necessarily believed to be true, right? So what I did was I gauged it up against, um, up against my experience, right? Because I've been a copywriter for a very long time, and we kind of came to terms on what we felt like was really the core position of this entire thing. And, and one of the things that we discovered is that we— one of the unique value propositions that we were underutilizing on our sales page was the brand reputation of Social Media Examiner, and specifically the reputation of me as the founder of Social Media Examiner, and the fact that, you know, 17 years of doing this thing. And that is something that matters to this audience because we are known to this audience. And there are a lot of people in the industry that are actually, um, really, um, nobody's ever heard of them before. Right? And they're not necessarily our competitors or people that are just coming out of nowhere claiming they're experts and having courses and stuff like that. So we wanted to position the, the reputation, uh, which was really important. And, um, in addition, it recommended that we, um, focus on the trusted relationships that, that, that I personally had built over many, many years because of my podcast, the AI Explored podcast. And it also suggested that access was a really important thing. And the fact that I show up live on all of these expert-led trainings is something that we should position properly. So basically all— and that's— I'm just giving you a sample, but all this new insight and all the biography changes and stuff all led into me, uh, retraining my Claude project. And then what I basically said is I said, here's the sales page, go rewrite it. Okay, and it rewrote the entire darn thing. Okay, and then I did my craft. I brought it into a Google Doc and I, and I started, you know, slowly tweaking it and challenging it. And I spent a lot of time on it. It was a huge undertaking. And we completely crafted a brand new sales page. And then this is all inside of a cloud project, right? So now I'm in a position where I have to do all the other things, right? Not only do I have I have the sales page now refined. I'm actually asking Claude to be a very critical, um, uh, conversion rate optimization expert, right? And asking it like, okay, which kind of headlines would work best? And very slowly working my craft because this is my craft. And then what ended up happening was now that I've got this new model, now that it's trained up on this new messaging, now I can actually start having it do the things that I normally do. For example, help me come up with a script for a podcast ad when we launch the new product. I mean, when we, you know, relaunch the product. And here's the angle I want to take. And it would write the scripts and would understand that. And then of course, I would ask it to create a series of emails. So now I've got like this thing that's kind of like really knowledgeable, trained up on how we do what we do. Now, a few tips on how to do this with excellence. I always ask when I am creating emails, because emails are kind of our secret thing that we do here, I always ask it to create multiple versions. So what I'll do is I'll say something like, create multiple versions of an email based on this theme, and I'll give it the theme, and it will just do its thing inside of Claude. And I, right now, I use Claude Opus 4.6, even though I know there are newer models, it's the one that I prefer. It uses, I think the, it's got the right balance of token use along with really creating exceptionally good output. And what I'll end up doing is I will ask it, hey, which of these is your favorite and why? And it will justify it. It will say, well, based on what I know about this target audience and this and the positioning and stuff, I believe this one is the best. Now sometimes I'll look at it and I'll say, okay, well, I like number 6, but I also like a little bit of number 2, so let's combine them together. And we'll go back and forth and then eventually I'll move it into a Google Doc and I'll do my final edits on it. And, and another really important tip is when AI, when Claude in particular, if you use Claude, because this is what we use mostly for this kind of work, when it seems to be making the same mistake over and over again, like my signature line, I'll just say update your memory so that my signature line always looks like this and it just learns., and it— you don't have to worry about updating your project instructions or anything along those lines. So that is, that is what we do in order to really streamline, uh, all of our messaging and make sure that there's congruency between what people see on the sales page, what they see in all of our ads and everything. All right, Allie, you've got some exciting stuff to share. So over to you.

Yeah. Um, so when we talked about the, the things that, that we do inside the management of the AI Society that like really uses AI in the places where we've made a lot of progress. One of the things that we do with every training session is we provide what is essentially a resource kit with every replay. So we don't just post the video, you know, like with the slide deck, although we do do that. We want to make that replay page as usable as possible, and it actually serves two purposes. One, we want to extract maximum value that we possibly can out of that training session so that when either the people who watched it live can come back and get even more added value, or the people that just go to watch the replay can actually get a lot out of it, even if they don't have time to watch the whole thing. And then the second part of that is the saving members time, because time is like the hardest thing to compete against. And we want to be able to provide a space where people can get, you know, their ROI from the membership., in the most efficient way possible. So after every session, we take the transcript and we're basically doing like a form of content repurposing. Um, so in our case, we take this transcript and we let it do lots of different things for us.

How do you get the transcript?

I'm sorry?

How do you get the transcript?

Yeah, so the transcript is delivered to us from Zoom. After we do the webinar. So I take that file and I convert it to different format that, that works with the AI. And then I will take it into Claude where I have certain skills set up. And if you're unfamiliar with skills, skills are kind of like Claude's version of like a custom GPT and ChatGPT or a Gem. In, uh, Gemini, um, but it's a specific set of instructions, um, that is kind of predetermined ahead of time.

So real quick on skills, um, actually what makes skills cool is that they don't have to just live in a folder like a project. They're usable across all of Claude, and they're exportable into ChatGPT and Gemini and all that kind of stuff. So you've created a skill to process these transcripts specifically from the recording. Okay. And by the way, you could do this with the podcast folks too. So keep going.

Yeah. Yeah. So bringing in the skill is just a little slash command and keywords that I've assigned to that skill. And then I don't even have to write in a prompt. I just call the skill, I upload the document and I hit go. And then it will give me a new session description that we can put on that replay page that is accurate to the things that were actually discussed in that training. Then I call in a second skill, which is for takeaway tips, and then it knows the exact format I want, how I want the sentences structured in a way that's easy to read, and exactly what kinds of tips I'm looking for. So then it gives me like 20 tips, and I sometimes narrow it down to maybe like 15 or so. But I'll put those on the replay page. At the same time, my teammate Cassie is working over in Gemini where she has Gemini Gems set up that help her with identifying the specific topics that were discussed in the training and what the timestamps were for when each of those topics was started.

Explain why we're using Gemini for this particular application.

Gemini is really— I mean, honestly, it could be either tool.

Well, Gemini is good with reading videos. That's the key thing, right?

Yeah. Well, we're actually just using the transcript with Gemini as well. But you could use just the audio or just the video. It could work with any of those things. Got it. Um, yeah, like I had a video in Gemini earlier today and it did things that astonished me, but, um, okay, cool.

So, so she's creating all the timestamps and all that kind of stuff.

So keep going. Timestamps, uh, also on that replay page and it helps her to format the transcript in a way that makes it really easy for people to read. So she'll take the reformatted transcript, put that in a document. And, and then the transcript also helps me develop a lesson that we put in our weekly Friday newsletter for members. So all of this stuff, some of it we didn't even used to do because it would have been way too time-consuming in our former membership. But since we've had AI Society and we have access to these amazing tools, we can actually add value to these replay pages for our members. And we can— once we get the video from Zoom, that takes the longest. That's the longest part, is just waiting for the video to get prepared. Once we have the video and the transcript, it takes us like 30 minutes to get it all posted where it used to be like we'd have to wait till the next day to have everything ready.

Well, how was the old way of doing it?

Does somebody actually have to watch the video and then kind of discern what the heck was I mean, because, yeah, well, back in the day we actually sent, uh, the video off to a separate person who did a transcript for us, um, and we'd have to wait a week for that transcript to come back, which just sounds ridiculous now. Um, now it's, it's just easy peasy.

Well, what's really cool about this is it sounds like you can have this up pretty quickly after the training, right? Like how long Does it? How long after the actual training is done is this resource thing put together?

I would say within an hour and a half to 2 hours it's posted.

Yeah.

And yeah, like, you know, we used to have to have somebody like go back through the actual video and write down the timestamps of when those topics were started, but no more.

And folks, if you are creating any kind of content that is video, or audio, you could take the exact same process that Ali just talked about and you could create multiple resources from it. Um, for example, we've got a different team, our editorial team, which will do all sorts of things from the actual interviews that happen on this podcast. And right now it's a pretty manual process that involves a lot of experts. We've got a video expert, we've got an audio expert, we've got writers. And obviously we're working to try to figure out how we can automate a lot of that process as well. So if you think about all the things that you do, especially if you're a creator, uh, there's a lot of stuff that you could rapidly speed up the value. And if you can speed the value, that means you can publish the content more quickly, which means it's more valuable to the people that watch or consume or are part of whatever you're doing. So lots of exciting stuff, folks. I hope that you try some of the things that we've been talking about. Now, if you know that you want to go deeper, if you know, hey, I feel like I'm falling behind, I'm slipping behind, I can't keep up, and there'd be value in you actually being part of a community where we have really great people like Ally who are curating the best of the best, and we bring in incredible trainers They teach you how to do some really, really cool stuff so that you can go from, I have no idea what to do, I have no idea which tools to use, to, wow, I'm doing this stuff. I'm actually able to do things that I could never do before. And as a result of it, I am more valued to my employer. I am more valuable to my clients. I am able to do things for my own business that I could never do before. Well, if that resonates with you, then consider joining the AI Business Society by visiting socialmediaexaminer.com/aisociety. Our most recent training actually is on Claude Skills and Claude Cowork, which is incredible. And I know personally that I've got a lot that I could do here. And, uh, it's called How to Turn AI Assistants into a Digital Workforce with Gemma Bonham Carter. So folks, give it a shot, check it out. Socialmediaexaminer.com/aisociety. Allie, thanks so much for joining us, joining us today on the episode.

It's been fun.

Thanks. Before you go, if you're tired of figuring out AI all on your own and testing tools that don't seem to work the way you want 'em to work and constantly rewriting scripts and prompts because they don't sound like you, the AI Business Society was built for you. Every week we cut through all the noise. Every training is led by expert practitioners, and you're surrounded by marketers who are just like you. And right now you can join for $497, which is our special sale price, and your rate is locked in for life. And you got 30 days to ask for a refund if it's not for you. So what have you got to lose? Check it out at socialmediaexaminer.com/ai, and I'll catch you on the next episode.