Why Podcasters WIN At AI Messaging And Everyone Else Is Playing Catch-Up!

This is feed you'd brand with hosts tom hazard and tracy hazard.

Hey everyone. Welcome to Feed you'd Brand. I'm Tom Hazard along with my co host, Tracy Hazzard. Today we're going to get into a topic that is very current at the time we're recording this and airing this. Okay. I mean, as opposed to, you know, a lot of topics are like tried and true for podcasting, best practices. Some things don't change. This is very exciting because an opportunity for podcasters in particular to take advantage of AI. So we're going to talk about podcasting and how you get more human like AI when you use your podcast to feed AI. Isn't that right, Tracy?

Yeah. So I've been doing some long term experimentation for almost. Well, I mean we've been doing it since 2018, experimenting in AI, but really for almost the last six months I've been doing a bunch of experiments on the what you would call agent builder side of things. And an agent is really this model of a. A person who acts like you, your clone. You may have heard that before, but it's also a representative, right? Like an agent, a representative of your content. And so it could be your company's content and you have a customer service agent who's not human. I just experienced that with my. Well, our newsletter is run through a company called Rasa I.O. and I don't know if it's Raza R A S A. I don't know.

If they do actually.

Yeah, I don't either. I never asked. Anyway, we love them. They're fabulous. They have the best customer service and their customer service is all at the beginning side run by an AI agent. But what that agent does with it, and you know, they're better at this than anyone else because they. Everything they do, their whole entire system runs on AI. So they have deep structural knowledge of how to make that work. But everything that they do in that process is prompting you. Helping you actually gets to a human being when the right time is that human being has all the information they need and responds. But that's what so few of us have the capability of doing, is putting that deep knowledge into something. So I'm going to explain the little story about Rasa in a moment and I'm also going to show you some of the things that you're seeing out there that you don't know are AI. That's how good they've gotten. And I want to share with you on the flip side of that. Some things that just really give us an Advantage as podcasters.

This is so true, Tracy. And you've been diving deep into this. We've been testing it a great deal. But I want to make sure that our audience and our podcasters understand some of the trial and error that you've gone through. Because we have lots of content, right? We have podcasts, we have videos of every podcast, we have blogs of every podcast, among actually other kinds of content that can feed AI. And you've really discovered in doing that what the best content is. And that's why I think you're saying there's particular advantage for podcasters now that may or may not apply directly to Rasa. I don't know. I guess I'm curious as your opinion on that, but Rasa's sort of AI support, right? That they have, and that's really separate from what their application does for newsletters. You're not really talking about. Because that's good too.

We really like that it's still all content gathering, right? Like, and that's what you need to think about it. And what I'm saying is, is that they've been an AI company from the beginning, so they're super smart about their AI gathering and logging and capturing and then deciding how to take action on that is something that they're expert at, and that's why theirs works so much better. But this is what we should all strive to achieve. So let me tell the Rasa story. So I was having an integration problem because I was trying to. They just newly got a zoom integration which allows me to show a webinar in my content feed so that I could add it to our newsletter automatically. So every week I go into the Rasa newsletter and I update this coaching call and tell it what the newest one is, put an image of it, make sure the link is updated, like, everything's clearly there. And I redo that manually. But they created this integration between Zoom webinars and their newsletter, and I went to go reintegrate it because of some account changes that needed to change to a different email address as its connection point. And it broke. And so when I went to go re establish that, it was giving me a Twitter feed that said test instead of our actual zoom events. So none of our zoom events were showing. And I was like, what is this? Why isn't this working? And. And so I just went to their AI chatbot in the corner and started saying, hey, I'm having a little trouble with the zoom integration. Like, this is my issue. And, you know, it said, did you try all These and it gave me all the things. And I was like, yep, I did all that and it's giving me Twitter and not Zoom. And they said, I think we should accelerate that to our tech support. Would you like me to submit the help ticket for you? And so I said, naturally, yes. It didn't give me a link to make me go out somewhere and do it. It said, would you like me to do this for you? And then it said, here's. Can you give me any extra pieces of information? Take a screenshot like any of those things. So I quickly recorded a Zoom clip and I dropped in the clip and said, here's my problem and here's what I want. And it submitted the help ticket for me. And in my email I received, hey, we're on it. We're checking this out.

All right, so for clarity, for any of you that don't use Zoom as much as we do, there is this function within Zoom to what they say record a clip. And what that really means is a screen share video. Or it can be a video video of just you, you and a screen share video, or you're not on it at all. And it's only a screen share video. If you're a Zoom user, it's really easy to use. You create a clip and you get a link. You can easily then drop in to an email, a text, or in this case, the chatbot for rasa support, Correct?

Exactly, exactly. And we use it all the time internally. We use these, not for any kind of social media, nothing fancy. We just use them for like internal tutorials and help and things like that. So it's just super simple and easy and integrated to what I'm already doing. So I send this forward. I get this the next morning, 12 hours later, like, they're just so good at it. I get an email back from the team. And this is what was beautiful about it. I not only got an email back resolving the problem, saying, you were right, there's a glitch and we fixed it for you. Go in and log out and log back in. And it should be fixed. Right. And that's what they say. But below it, it has a complete copy of the chat that I had with the chatbot.

You mean so the whole history of the entire chat you had and all the context.

Yes, all the relevant part of it. Right. Like the part that was about. That I've had. Cause it kind of. It must save your ongoing. So like it asks you if you want to go back to where you were before. And I started A new chat, because where I was before was something totally different problem, Right? And it comes back and it gave me that history. That's a beautiful use of this log and relevance, because it means that the chatbot's moving on and keeping this information and forwarding to the human who needs the information so that I don't have to go back in, remember what it was wrong, try to recreate it like, you don't have any of those problems. That's what we really need is this feed mentality to all the content that we're creating and using with our AI system. This ties to what we're talking about today. And so when we create that continuity, we create ability for the AI to train and learn faster, for it to make more relevant support results for us or give us more relevant results if we're looking for an answer to a question, because it knows where we're coming from, it knows the midpoint of the conversation. It doesn't start fresh every single time. So when we are creating our agents, when we're creating stuff, the feed, like we do in podcasting, is ideal because that feed is growing with us. Now. Some of us cringe at our early episodes, right, Tom? I mean, we were just looking at volume one of wtff.

Oh, when I think of how contrived some of what we did was because we just weren't experienced that, it is a little embarrassing. The only good part of it is I looked a whole lot younger than.

Yeah, absolutely. What was it? We had one of our longtime listeners on wtff, the first time they ever saw us on video, they said, tracy, I thought you were taller, and Tom, I thought you had more hair. Isn't that always the way? Right. And so anyway, the idea, though, that those early episodes, like, it gives it all conversational growth. The AI is actually learning with us. It's growing with us. And now we really hope our listeners are going to binge on it and grow with us, but they don't learn. Our listeners don't learn at the speed that our AI will. Right. I mean, it's just not possible. So in order for it to give us what we're really asking for or to understand why we really, in this particular context today, why we created an audible and did this topic today is much more logical when it's been following all the things we've been talking about. And. And it's not looking at the episode in isolation. And so this ties to what I tested and tried out.

And is this where the benefit of podcasting will be revealed. Tracy, in terms of informing AI.

Yes, exactly. So now I want you to be thinking in the back of your mind, Podcast is a feed, content is a feed, right?

The constant stream of more data. Is that right?

But it's also in a timeline, Right. If you think about that now, we recommend to people that they're jumping around, that they don't do it in a linear way. Like, it's not a training, but it still has. It builds off of each other. Right. Early on this year, one of our clients and who comes live to the coaching call regularly said that she felt like we were layering. Like we were creating layers, one on top of the others, layering topics. And that once you grasp this, then it was a lot more easy to implement. Implement the next thing. Well, that's the way you teach the best, right? That's the best process for teaching. Because I can't take it all in at once and do it all. I need to master something and then take it. The next level AI actually has to do the same thing. Otherwise it has to do all this research and try to understand the groundwork and foundation that it's built on or it misunderstands it. And so this is what I want you to be thinking as I'm talking here today. Feed. Podcast is a feed. It's a natural feed. It's already naturally happening organically. You're creating it that way. So it has the best shot at being the best feed into an AI. So I tested that theory.

And what did you do?

I went to an AI agent creator and I took it and I said, well, the most comprehensive piece of information we have is a blog, because our blog has our social media clips in it, it has our YouTube video in it, it has our podcast audio in it, and it's also edited. It's not just.

It's not the raw. Right?

It's not the raw speech to text that comes with, like, the crappy captions you read. You know, it was really funny, Tom. We were meeting with someone and we were explaining them, the last 10% of the most speech to text is 90% accurate, but it's the last 10% that actually we care about the most as a business, as a human, because it's the one that last 10% is where they misspell our name and where they don't understand what the word potatoes is. And so if you actually look at the transcript in our own system from Zoom, it says pot, statize, P O T S a T I Z e. It's not Even a word.

And it also naturally spells our name incorrectly with onesie instead of two, among other terms. Right?

Yeah. So if you're in business, if your whole point of podcasting, if your whole point of creating videos, if your whole point of everything is getting all screwed up because of a bad speech to text model that doesn't know you and doesn't understand what's critically important to you, it's not going to work. So here I think the one that's completely edited is going to be the best option for me. And it is going to be. And also because it's most complete, it has everything I need in there, everything that I've created, and it's all in one spot.

It'll be the best data to feed the AI. That's what you thought.

And I. So I thought, aha. Potatoes creates blogs. This is part of our services. This is going to make that service so valuable. Like, this is true. Guess what?

Didn't work.

Didn't work. In fact, it crushed the system. It got so confused, it couldn't handle it. And it got completely messed up by the multimedia that was in those blogs. And it came out just jumbled and all kinds of wrong. And I, the other reason is because a lot of these agents have references. So I thought, I want it to reference that blog URL, because that's where someone could get the most complete answer to their question. And so it gives them an answer and it says, and now go to this blog. Right. So I thought that that would be the logical way. And it just couldn't handle it at all, in fact. And so every time it was trying to answer a question, it was too much data for it to go through. And because, and this is the key here, unless I were to put those blogs in one at a time, when I put a blog feed, it actually doesn't go in any linear order, which was shocking to me. I would have thought it would have gone in date order. But our blog feed isn't set to have the dates in the feed titles. It's not set in the URL. So many people will take out the date because you don't want the date in your URL, because then the search engine doesn't recognize it as current data. Like, so it's an SEO reason you remove the date, but because we removed the date, it didn't work. It didn't put it in any order. So it was creating its own logic to the order and that was messing it up. So it didn't work.

All right, so you're saying to inform an AI chatbot. It really didn't work.

It really didn't work. Now, I'm very sure that I could have, like, really played and messed with the blog feed to get it to work, to get it to ignore things, but it was too much work for me to do that. Right. It was just not quick and easy. So then I said, well, the next most prevalent thing is YouTube. Everyone has a YouTube channel. And the great part about the YouTube is it's not just going to have my podcast in there. It's also going to have all our tutorials. Everything else we might possibly want is like, our YouTube is just full of stuff. But that's the problem. The YouTube is full of stuff. And for some reason, its speech to text is somehow worse coming out of the YouTube.

Oh, I've seen that with the auto captioning. Yeah, it's not great.

I don't know why, but maybe it's because it's pulling an SRT file that this particular agent I was using, it wasn't doing its own speech to text. It was choosing the SRT files. And they were already so bad. But it was like missing stuff that it wasn't missing before in my blog version. Right. So it got worse as well. But it was also. It was just too much content and it was too unstructured because it would ignore all the playlisting. So it just went everything in date order.

Wasn't there also an issue of the fact that the video content was so large in terms of file size that it took the AI a long, long time to try to ingest it?

Yeah, so it did take quite a bit of time to process it. But once it processes it, it doesn't do it again like it, you know, but it would update it, you know, if a new video got added. But because it was looking at everything in the order at which you loaded them into your YouTube, not the order which you want them to be in or, you know, it just. Because sometimes we put tutorial videos in between podcast videos. Right. Like, they're just. And they're coming at random times, or we update something or we create a webinar. All of that stuff was meshed together and it messed up the understanding of the content. Does that make sense, too?

Yeah, that makes sense. All that seems to make sense to me. But. So you did get to the Holy Grail, it sounds like, though.

So I did. I got to the next step and I said, okay, I'm gonna. I wiped it all out, actually, you know, because I. I Deleted everything and started fresh each time because I didn't want it to be confused.

Well, also, you wanted to really test each type of data, Right?

Right. And so I did have to, because some of the content did not exist in our podcast, like our tutorial videos for our portal, for instance, you know, for using the Productize portal, we only send those to clients. They're actually in Vimeo for some of them because of the way we load them in, and some of them are in YouTube. So I wanted to. So I had to create special playlists for those, load those in first because I wanted our products in first. So I loaded those in first, and I loaded our product pages first so that our, Our. That the AI would understand what I sell. So if you had, if you had any ebooks or other things that you were downloads, providing our social media guide, I dumped those things in. So I put in all of those, like basic documents and in basic information so it understood the company.

That makes sense.

So I did those first because I knew that those were all individual stuff. Then I went in and put the binge factor and feed your brand. Only those two at first.

Well, those are both our podcasts about podcasting, which is our primary business. So that makes sense.

Right. So I put those two in first and I. I loaded them in as.

Feeds, RSS feed for same thing you would syndicate to Apple and Spotify and wherever.

Yeah, And I was shocked at how quickly it processed. It was done in a tenth of the time, the individual ones. And I probably only put like 50 items in. So, like, it wasn't, you know, a tremendous amount of stuff between the tutorial videos and stuff like that. I put all of that in and it processed really, really quickly. And as soon as I started asking the same questions I had been asking before, the answers were coming up exactly accurate with the most current information first. So, like, we might have advice about some Apple. Something to do with the way Apple syndicates or something to do with the way Apple works. And Apple has changed in the amount of time we've been doing this podcast. So we've done like 300 episodes. And, you know, the advice we might have given about how Apple treats something in episode 10 is not the same as we gave it in episode 150. And so it was going to the most relevant, most current information and getting the answers right the first time.

In fact, Tracy, I remember we have an employee who we have assigned now to really, it's her primary job to assist us in this endeavor with AI and our own AI chatbots and, you know, creating different apps that we're going to use in our client portal and other things. And she took your AI chatbot and was testing it, and she said she was really trying to break it. But then in the end she said, but I think it really broke me, which was a funny way of her saying, I couldn't get it to give me a bad answer. I couldn't get it to give me something that was not accurate, among many, many other things. And it's a really high compliment for the quality of the data going into it and how the AI is interpreting it and to provide responses. So it's the way I've been saying is it's wildly accurate using this podcast data. So if you have a business or any other good reason why you might want an AI chatbot, and there's about a dozen different use cases that we've been thinking about, but where you have a podcast that's in alignment with your business and in any way teaches something and educates people, informs them, or you just. You have. You need customer support, and you have a lot of things that, you know, people are involved in answering and doing these things. Your podcast is the best data to use to inform AI of anything. I mean, and we're a business, and we have all different kinds of data and all different kinds of media. But here, Tracy, you proved the podcast RSS feed content, the audio episodes are the cleanest, most efficient data and created the most accurate results. Am I missing something there, or is that right?

No, that's exactly it. But there's a couple of other nuances I want you to think about here. So one of the big problems is once something's in the training model, it's extremely hard to get it out. So once it ingests it, it's in there. Like, you can't.

Can you guide it not to give an answer anymore if it's not quite right?

Yes. I mean, but you have to manually go in and say, stop answering this question this way. Answer it this way. You have to redirect it, and you don't always know somebody's going to ask. And how often do you remember that the advice you gave two years ago was wrong, is different now because things changed. Right. You know what I mean? Like how. I mean, it happens all the time. Like, every so often we go, oh, we got to record another portal video, because you know what? This. This little tip change, this little tool changed, and we forgot about it. And so you scrap an entire video and you got to start all over again, right? You cannot do that once you've trained your model without a feed system that has a currency and relevancy factor that is predominant. So in other words, if it's more recent and if it's duplicative, right? Like meaning like I've covered this topic before, the most recent information matters most, right? So you're giving it that relevancy, that current is relevant, current matters more. And so for some things, that destroys historical information, right? Like, and this is why in the general AI model it's screwing up because the most current information is AI regurgitated information. But in your model, if you are doing that new content and you're not using AI generated content as your newest content, right. It's still you're going to be covering it. So like we probably covered, you know, I mean I think I just this year I covered guesting four times and I probably covered it 20 over the years. But the difference is in nuanced that how it works today, there's way more messaging than there used to be. It used to be super easy to pitch somebody to be a guest on your show. Now it's a lot harder. Right? Like all of those things have changed and the nuances are training and growing with your show with you're building on it, right? And so you're not going to delete and reinsert. You can't do that with the training model. It's kind of like a year ago. You guys know I spent an inordinate amount of time in jury duty. You can't have the jury unhear something no matter what happens, even if it gets objected to, Right. Or it's removed from the stricken from the record. You heard it. That's what happens with the AI. It hurt it. So we have to replace it with something more important. And this is one of those things where this is why you will keep reiterating, redoing, going over some of the same topics again and again, which is only going to reinforce your search engine optimization, your keywords, your topics. And it's insanely correct to be teaching that way as we learned from this process. Right? Because it's a layered approach to teaching humans, not just AI.

There's so many things running through my mind right now, Tracy, but one of the things I think we want to make sure that our listeners understand is we're talking about training a private AI chatbot in a silo that isn't using all the information out there in the world. It's not going out and searching the Internet. When somebody asks a question, it's searching just throughout the data that you add for it to use. Correct. Which I think has an advantage, because if you're in business and somebody asks a question, you want to give an answer, that's yours. So you are the authority answering the question. And any references that it comes up for you to click and go read more information on a topic are yours, not other people's. They're not going to distract your person, engage with the chat bot and send them out to this other website and that other website and all that. Is that right?

Yeah, absolutely. Thinking about it, one of my favorite authors is Shane Snow. Smart Cuts is my favorite business book I've ever read. I think I've probably recommended it on like, dozens of different podcasts before in interviews. And what it is is that too often everything tries for the shortcut. Right. They try to come up with a fastest way to do something, the fastest answer, the shortest answer, the quickest way to get an answer. But sometimes that's not going to get you where you want to go. In order for me to get what I want in the fastest time possible, there has to be a foundational understanding between the two of us as to what that is. And that's not the case in the general AI because it's getting a new question. Remember, Tom, how you used to say that one in five questions that were asked to Google, they had no reference for, never heard of before, didn't know.

How to answer, nobody had ever. Yeah. And I don't know what the percentage is today. I'd have to Google it to find out. But, I mean, the last time I did it was maybe a year and a half ago or something, and it was still 20% of everything entered in Google search bar every day is a question or phrase Google has never seen before, which is kind of staggering to think about, but that's why there's so much new information constantly going in, you know, to Google. Yeah. So it's probably similar today. So what are you saying about that?

So what I'm saying is, look, I just did this actually, so I had to go outside of our system because Gemini knows us too much. Like, it's really hard, so I can't use that. So I went to ChatGPT and said to ChatGPT, what makes a compelling show description?

Podcast show or just show?

Podcast show description. Yep. And so Craig Constantine over at Podcasters Community, which I absolutely love, they have a great board over there where they message board where they talk about it. He was talking about realizing that he wasn't really, he didn't really cover his topic well enough. Like the keywords for his topic, like how people would really search for it. And we all know how crappy the podcast apps are at searching. Right?

The search in the native apps is. Yeah, very Internet 1.0. Very limited.

Yeah, right, exactly. So, well, so I was going to answer him back and tell him my advice on it. And you know, we have our how to create a great show description and we have all these tools and things in our own portal. And so I thought, okay, let me do two things. I'd ask ChatGPT what it thinks makes a great show description. And it says at the beginning, a great podcast show description is essentially an elevator pitch.

Wow.

And it goes on to say it should be concise. And it does at some point say, well, some platforms allow you to use 4,000 characters. A good description is concise and gets to the point quickly while still being informative. And that is absolutely the worst advice ever.

I would agree 100%. I'm cringing just hearing that. I mean, the elevator pitch is great for the first two sentences maybe, but that's all anybody's ever going to read before they decide to click and give your show a try. But that's not the only purpose of your show description. And we've talked about this ad nauseum on other episodes. But anyway, continue.

So then I went over to me to my chatbot.

You went into our chatbot. Okay.

Yeah, right. And I asked it what makes a great podcast show description. It says it is all about clarity, connection and discoverability. It needs to hook your ideal listeners. The first two sentences are gold. Focus on what they'll gain by listening, not just what your show is about. Avoid ego driven intros like I'm the host of. And instead of instead lead with value or transformation, your show delivers. Second, make it SEO friendly but natural. Use keywords your audience might type into a search bar, but don't cram them in. Exactly the advice. So, like, I was about to write all of this to Craig in the message board and instead I hit copy and paste from my own chatbot. And it gave all the references that I needed so I didn't have to go look up the links. So it said, here's our podcast show description exercise and here's a way to boost the SEO naturally for your show. And like, you know, all of these things were all right there and ready to go like it was. I mean, I wouldn't have had to do it. I could have had an assistant to it, and it would have been my words and my answer. That's why we don't want to just use this general model. You're the expert in it. In fact, when we were talking to some people about this the other day, we were saying that the biggest value is someone with a very large show catalog. Mine's well over 5 million words. Someone like our friend Scott Carson with the no Closer show. It's so deeply in the niche. No one has more information on note closing than Scott Carson does. His information is more valuable than anything that is contained in the large language model somewhere else out there. So if anyone needed to create, like, let's say they had a note closing business of some kind that maybe was like a sub segment, the best thing that they could do is license Scott's content to create their bot on their.

Website to answer questions for their customers. That's pretty brilliant, right?

So this is the future for us podcasters. This is where we are going to see the value in the future for us. Because we have a feed, because we keep it updated, because you must keep podcasting without the continuation, without keeping it current. You're not helping yourself, you're not helping the AI and you're not helping your future listeners, your future business contacts, your connections, your sponsors, whatever it might be in the future, whatever way that goes. You're not helping them. When you quit your show, a catalog is going to be as outdated as the. As soon as the next guy produces an episode, you're done. And so those things mean that this body of content, plus it being conversational, because that's what we do here on the show. And conversational teaches that bot. That's why when I was. I was literally reading you what my bot said to me. That sounds like something I would just say to you.

Well, it's because you did.

Because I did say it, right? And it just, it found it amongst all the concise long witness of me, right? This is me, only better, only faster. Only a smart cut, right? Like, that's what I want you to think about. This is me in the smart cup mode. It gets you your answer faster.

So, well, plus it can be there serving your customers, your community, wherever they are, 24 7. You don't have to be alive there to answer the question. They don't have to wait for a person to read an email and write an answer back. I mean, there's so many advantages for this. And I also want everyone to know, obviously this episode is not how to create your own AI chatbot Right. We're talking about the opportunity and that you have as a podcaster, if your podcast is in alignment with a need out there that people have or you know, that your business has to create a much better customer experience. I mean, I'm even I have other things we want to create one for to assist on the sales side of our business, all sorts of things. But we're still researching, perfecting and testing all the aspects of the chatbot and as we get there, we'll share that well.

So, so many people just want to use the chatbot to be a sales person, right. Like an extra sales or a customer support person. Right. And so they're taking their FAQs, they're taking their past emails. This is where I'm saying you're making a mistake because it's not going to create the robustness you need and your ability to update it and monitor is difficult. So instead what I'm recommending you do is you create a podcast. Now, you don't have to create a podcast that you syndicate out there and is on Apple and Spotify and you know, wherever else you listen to podcasts, it can be completely private. But the feed model allows you to keep it updated easier. It does allow you to scrub some things because you can just pull it out of the feed and it's gone. Right. So it's. You can scrub some things that are totally inaccurate or that you don't want to be representative of in your model anymore. From a reference point, it still trained on it, but from a reference point it's gone. So those things are really wonderful. So, Tom, your sales content is an absolute nightmare. It is a content nightmare in the large language model space into the creating of a data center. Because a data set, because it has no rhyme or reason, no order to it, no value adjustment. Like you could have had a great sales call on one hand and the next one was really sucky and didn't result in a sale. So like there's no reference points. Just because you have all the recordings doesn't assign value to it. That's where taking those and putting them in a feed in a natural way. So we have feeds that we create for guesting opportunities. So like when you guest on a somebody's show and you want to create a playlist that you put on your PR site or your About Us page.

Like a player on your website or. Yeah, or on a media page.

Right. We have a way to do that without syndicating that show. So it's not out there, but it's now on your press page. And you don't have to do that much work. You just copy the link from whatever podcast you were on and drop it into the player and now it's there. That's what you need to do with your sales call. So when you get off a great sales call, you need to be able to take the link from the video, drop it into the portal and be able to take that. And now it's in an order of what it is. And then put in its episode description why it was good. Give it context. Give it like as if you were giving it a title. Because how much of our stuff is titled Date and a person's Name.

Oh, for sales meetings. That's probably all it is for just.

About everything in the podcasting space. Anyone who isn't our client or hasn't been working with us or hasn't gotten our advice or most of them put the worst episode titles on earth out there. They're just a number and a name or a cutesy title that like mirror sitcoms. And that's great if you're in Hollywood, but really bad if you're creating context for stuff. So thinking about that continually is like if we're going to participate and receive reward, whether it's business directed to us or it's actual monetized licensable reward, we have to be structured with our content and we already know how to do it here as podcasters. So let's do it more. Think about all the things you should just add to your podcast because you can.

Very exciting stuff and opportunities. Another really good reason why if you are authority in your niche, you should be podcasting and it is to fuel AI for these types of tools. Very, very exciting. And there's a lot more to come on this because this is a big initiative for us. We're going to continue to be doing more of it and we'll be sharing that with you as time goes on.

Yeah. And I think we'll have a future episode coming up sometime this summer for those of you who know when you're listening to it. But it'll be coming up in the next month or two where we're going to talk about the voice side of it and why your podcast is actually even more valuable for that. I really funny experience Dom, that you're going to laugh at. I recorded a voice script for Zoom and it told me it wasn't good enough. I used my microphone. I did everything. It was probably just too good and it didn't understand. It was very Odd. So that we're going to talk about some more things like that, where how you record and what you're recording, how it can really create a much better voice bot on the other side. Like mine is. I mean, mine is sounds like me.

So does sound like you. And again, the conversational aspect of podcasting is so important. Most people's FAQs on their websites are not conversational at all. They're straight question answer.

Yeah. So the last thing I want to leave everybody with is this idea that there are estimates between 2030 and 2035 as to when the large language models are going to run out of content that's not duplicative, not regurgitated. So in other words, it's AI created content. So thinking about that from this standpoint of where your value lies, what we create every single day here is going to continue to be valuable. And as everybody else runs out of data. I'm not running out of data. You podcasters, I know you can go long. You're not going to run out of data. You'll just start a new show. You might get bored with the data you're talking about, but you're not going to run out of it. And that means that we have the greatest opportunity for continued value in the future world. No one can write at any magazine, right? Faster than we can speak. Keep that in mind.

That's for certain. All right, well, let's leave it there. Thank you all for listening. You can take a look at any of the links to anything we talked about today and get a lot more information on podcasting with a great search feature we have on the blog post and the page for feed you'd brand on podcastersunited.org, which is the home for this show. So please check that out.

Heard something valuable today, but didn't get a chance to write it down. No worries, we've got you covered. We partnered up with podcasters united.org where podcast hosts like you can get free resources, free masterclasses, the largest podcast tip library assembled, and free promotion opportunities. While you're there, don't forget to follow us on YouTube and on social @United Wepod so you can get feed your brand live. Until next time, keep podcasting.