Welcome to about podcasting, a show for podcasters. We talk about podcasting practices, tools, successes, and failures mixed with interviews and music. Hosted on Podhome dot f m, the most modern podcast hosting platform. Welcome to another episode of About Podcasting. And this episode is about a podcasting tool. Today, I'm talking with Michael Katz who is the founder and CEO of Flowsend dotai.
Now Flowsend is a clever AI tool. You can upload audio or video to it, and then it will generate lots of stuff for you like chapters, title suggestions, show notes, clips. Kinda sounds like a Pothome AI, right? That is in Pothome, my hosting platform. It does all that, but it does a lot more as well. So it also creates an outline for a newsletter, for instance, for a blog post.
It can create TikTok clips, all that type of stuff. It's a really interesting tool, and, I would encourage you to play with it a little bit at flowsend.ai. Now without further ado, let's listen to Michael talk about this product. I've been, I've been playing with it, flowsend.ai. I've been trying it out. Jeez. It does it does a lot. So it's it's very broad. I uploaded, an episode to it
and then said, hey, create a transcript. And then, hey, create a content, which is all in in, like, a couple of minutes. And then it does titles. It suggests titles, episode description, timestamps, which it does a really good job at. And I know how difficult that is, actually. Topics and show notes, also very good. Very good. Clips and quotes, which is very useful. See SEO keywords.
LinkedIn episode release, also very good newsletter, suggested newsletter, suggested blog post, x or Twitter Twitter thread. And I can choose to create, a lot more things like a YouTube description, YouTube shorts finder, TikTok clip finder. Yeah. Clips for TikTok and and all sorts of stuff. So I am I'm impressed. It's cool. It's good. Thanks. Thanks.
Yeah. That's the hope. And, you know, it's one of those funny things where I'm simultaneously really proud of what we've built already, but also, I don't know. It's hard not to see your own product as the product that it will be in 3 years, 5 years, whatever it is, even 3 months, and to be frustrated and to think they're not even seeing a, you know, a fraction of what it could be. And so that that's kind of a funny funny contour where you're saying, yeah. It's so it can do so much.
The funny thing, of course, is it can do even more today with how much customizability we put in there, but then you're thinking, oh, imagine if you could see what we're yet cooking up cooking up right now. So when did you go live with this? You know, we we technically went live to the public with this, about 2 weeks before this recording. Oh, cool.
And, you know, we'd been in kind of a private beta before that with a bunch of podcasters, creators, producers, agencies, just kinda to keep it scale, but but limited, and make sure we could handle all the feedback and didn't, you know, inevitably in the early, early, early days, there's a couple bugs here and there that you want people to squash for you. And so you say, look, I'm giving you the product for free.
And you're getting value. And what we're getting is, you know, the occasional forgiveness if there is something wrong in a very specific education. Yeah. Definitely. That that's definitely hard enough. And and how has, has it been received? And it's been received great. I can't really ask for more. We we technically launched, at the podcast show in in London. And, you know, in our experience, we
operated primarily in b two b environments and done a lot of events in the past. And, you know, events are this really, really special force of function. They they make you do something. You can't really negotiate, timelines with someone else because you gotta go there. It's not getting rescheduled.
And so we decided, you know, I'd heard heard about the conference about 3 weeks before it happened, decided to take the last sponsorship spot and said, basically, if that's when we're going live, that's when we're gonna start putting pricing up and, and and really, getting after it to the public. So it's been, it's been really special. And then I think the most interesting part since launching has been seeing some of the the interesting use cases that people use it for.
You know, we have an idea, okay, here are the ones we're providing out of the box, and that that'll keep getting expanded. But then you just end up seeing fascinating hacks that people use the platform for that you never would have expected. Yeah. That's the thing. Right? So once you go live, then finally, you find out what people actually are using it for, or what's the use what's the what the problem actually is that your prop
product solves, and it might be different than what you actually thought it would be. And maybe you change the products based upon that. Yeah. At that point, you need to decide, is what this person is doing breaking the product? Like, is it something that is, you know, fundamental for them, but for no one else? Is this something that a lot of people could use? And that that takes some call to the judgment too. I think it's lucky.
Right? You know, you're you're you're kinda lucky to actually have people using a product and to be pushing it in certain ways for their own need. That's a blessing in its own right, but it also forces a lot of really interesting
conversations and debates and and and prioritization, of course, in life is one of the hardest challenges that we all deal with on a on a regular basis. And no more is that more apparent than in building a product with a nimble team that you're saying, look, we don't, you know, have unlimited resources. And what's the original problem that you're trying to solve? Was it your own problem? It was my own problem. Yeah. You know, before this, it's interesting. My background was in b two b software.
I founded a company called Cowen that built the software for sales teams. So very different from from what we're doing today and, you know, pure enterprise sale, large teams. And at some point, I realized I needed to create this marketing and content engine to spread Cohen's message, software building. And I decided to create a podcast
that was like, you know, it's called the Windwire. And I brought on industry executives who are my target customers, and it had this very specific concept that was tied to the kind of software you're building. I was so nervous to start it. And, but I've been set up for lunch with a successful investor who actually has this really big podcast and a little podcast network. And he encouraged me just to do it. Just like, what could go wrong?
You know, it's it's your brand out there. It's your voice out there. Does anyone want to listen? Is it going to be good? And so I started it. And despite its success, right, I think we actually had a lot of listeners and amazing feedback. I found the production process was time consuming,
especially if I'm, like, a little bit of a perfectionist and need to learn my own voice and how I wanna run this. And, you know, I was using chat g p t and, you know, as an expert user there, it still took so much time, so much iteration to kinda get to what I wanted to get to, and to really imitate me. And I wasn't generating that content engine
that I wanted. The reason I started in the first place is I don't want this to spit off content. Because by the time I finished production and created some copy, I was on to the next thing. I think that's what most people are. We turn in the paper and we're just like moving on. We got other stuff to do now,
or next episode. And so I tried all these other podcasting AI tools, and they didn't really meet my needs in terms of quality or matching my brand voice or customizing that workflow, getting more out of it. And so it led me to wonder if others face the same challenges. So as we were selling off our sales startup, I talked to some companies and individuals and thought it was a pretty common problem.
And so on that really base layer was okay, this is taking a ton of time. On the next layer was I'm not really getting enough out of this. Right? I need to I want this to be really accretive to me in terms of whether it's bringing in customers, bringing in listeners, establishing my brand and thought leadership. And, you know, wasn't necessarily doing that in podcast. Growth is so, so hard. You kinda have to throw a lot of stuff at a wall to make sure it works. It's a huge problem.
So how could we basically generate all that content in the same amount of time as it would take to generate 1 piece, right, or less, and allow it to be really customizable. So it learn new. And so that's why we built it, and we we leveraged some of our existing AI expertise to to create a tool that simplifies and enhances that post production process, marketing, content generation, repurposing. And,
yeah, so scratch your own itch. Basically, it's really the best way to build a product in the first place, I think. Yeah. Totally agree. I I did the same thing with Pod Home. I created podcasts, and it's difficult. Right? Hosting is old school. The UI's role, not what I wanted. And I wanted just to make it easier because it takes a lot of time, like, creating chapters, creating good, titles and descriptions, all that type of stuff. So I have a similar but, much smaller
kind of similar capability than you have. I call it Podhome AI, which is an part of the hosting platform. You upload your episodes, and then we generate also a bunch of stuff, not as much as you generate, then it's right in there in the podcast episode already. And then your thing is different as it's much broader. Right? You can also create newsletters, blog posts, create specific clips for TikTok, YouTube, all that type of stuff, and also for video, of course, which mine is only audio.
So that's different. Well, and it's interesting you say that because, you know, there's so much overlap and collaboration happening across the space in really interesting ways, and everyone's gotta carve out their own niche. Right? You know, I've seen other hosting platforms embed some AI capabilities into theirs. And, you know, we're in talks to collaborate with some as well. But for, you know, I don't know, 50%, maybe more of podcasters
or creators, that's actually sufficient. Right? Like, I wanna generate these titles, description, show notes. I wanna get this out there. Right? Just like just give me something, And it can be pretty good. And those people can use our product, but they also don't need to. Right. Maybe it's sufficient for them, what exists out there. And that's that's actually, you know, fantastic. We support that.
We basically said, look, we we're kinda targeting a specific market, and it's our ability to really care about that content for people who wanted to sound like them. Right? It's my brand voice, and I like this format.
And I need these outputs specifically for me because I need to put these things out there, and I need this platform to learn from what I need. And that also forces a lot of focus constraints on us, or we're not gonna go there and build the hosting distribution and kind of some of these, like, very, very specific capabilities and, you know, that there's a really big market for. Because we're basically saying, look, today, we are focused on extremely high quality or end of line written content,
and getting a lot out of that. And I think it's a good thing. I think it's a good thing that, you know, we're providing table stakes, and platforms are embedding it because it's fantastic when someone uses Pod Home for that, saying, look, I have it uploaded. Just give me this stuff. I really need it.
And then it's great if they say, look, we wanna use Pod home for our hosting. We wanna use flow send for our content, and that's a beautiful relationship. And you wanna go build integrations around that stuff and say, yeah, we actually have a really close relationship. And I'm all about that kind of stuff. Yeah. That would actually be, interesting to, collaborate,
on that. For instance, if somebody, generates a bunch of metadata, at yours, then we can say, hey, create a new episode in, in PortHome, for instance. Or if you were to, expose some APIs that I could use, something like that, that would also be very, very interesting. And, of course, I'm very, very interested in how you actually do it. But I don't know if you're willing to share. Well, I mean, I know I'm willing to share plenty of it because I think, you know, part of what's going on here is
there is a secret sauce, but it's very public secret sauce. Right? Which is, you know, the average user and even, you know, sophisticated users don't want additional complexity. AI is really complex. It's nondeterministic, as you would say. It doesn't produce predictable outputs a lot of times, and it's hard to control. And there are ways of doing it. It's like an expert horse rider.
Right? They know how they'll tell tools. It's not gonna be easy for your average show. You can just ride a horse along the path. Sure. But can you do the really complex stuff? No. So how do we actually abstract away a lot of that complexity from the user? And that takes a lot of time,
but I think it's kinda like the duck paddling and you don't see how hard it's swimming under. It looks very calm. Kinda like that, which is you actually don't see simple simplicity hides. I mean, you know, it's it it looks simple on the surface. There's a lot of complexity in the back of how many iterations we've done to actually get to something that we do think is really reliable.
You know, we are the only platform as far as I know in this space who's using multiple AI models, and kind of almost choosing those for the user for a lot of our out of the box outputs to say, yeah, we've actually found that this model produces better blog content as an example. How do we choose that for the user? Right? And give them customizability as well, but we we wanna take away that complexity because there are really hard problems under the surface here. And so,
you know, like I said, I'm actually happy to kind of explain that because it actually gives us an opportunity to talk about, hey, how much time do we need to spend to get you something that is really reliable and simple? And that's something we're testing constantly, right, both in human ways and automated ways. And basically, you know, what's interesting is so much of that customizability we've built in and quality focus we've built in.
We built it for individuals, but some of the design choices and infrastructure choices we made turns out lend really well to people who are doing this at scale as well, you know, who are managing multiple podcasts or videos or webinars with multiple different out output types and brand voices. And, it's so tempting to be like, yeah. I'd love to just serve the enterprise. It's really tempting.
They have more money and they're doing more volume. But I think at the same time, I I try not to forget about the individual that I was struggling with this and, you know, basically
wishing I had something that, like, what I built, and I don't wanna forget about that. I don't wanna forget it for that person. So when people come up to us at the conference and saying, well, that's cheap for me. I'm like, you know, I cringe. I shiver for a moment, and then I'm like, well, you know, okay, Michael, don't get carried away because you need to serve that person and make it reasonable for them.
You can get, you know, some of these larger companies to pay more and and go figure that out. Yeah. And you can do both. Right? You can have an enterprise product, some for for scale, for for integrations with, with hosting companies or something like that, and also serve the individual. That could just be the the same product or the same same base. Right? That can still work. Yeah. For for sure. Although, what I would say is really hard about this kind of space,
podcasting in general that I've found so far. And, you know, like I said, my background is in b two b. So I love selling companies. You know, it's got its own hairy problems too. But at least you kind of know what you're getting a little bit and you know what they have available to them. And you can talk to them more generally about how to use the product When you turn it over to the masses that produces its own really big problems, everything needs to be really intuitive.
And you watch a session recording and you see someone fumbling around missing a feature, and you're like, no, it's right there. And that's it's so natural. Anybody who's built a product has seen this before and knows this feeling of, like, well, of course, how did this not be intuitive? But to someone who didn't build it, that being this is totally confusing to me. And so much of the what makes us really different beyond the quality is what I would say.
You know, that is sort of, like, subjective, but I think, you know, at a certain level, you you recognize it. So much of the other kinda, like, features that make us different. Some people would never even touch them because they just know how to do the very, very simple workflow. And so you think about constantly, how do we expose those better so they can make it good enough that if they don't, it's very good. Let's figure out those ways to to to embed them
and and get them using it. And and that's a whole new challenge for me. It's like the customer acquisition part on, like, the cut quote unquote low end or individuals, and those are totally different motions. And then number 2, you know, controlling the product for that person. And I think, you know, doing something that is really user friendly also benefits you in the enterprise as well, but it's still just that much higher level of rigor. Yeah. Definitely. Consumers,
the yeah. It's special. You need to make it just very intuitive, very, everything just has to work very, very well and has to be findable, like you say. We have this, in in Pod Home, we have this thing called dynamic text. If you upload an episode and you do, your episode description, you can type it in and you can format it a little bit. And then, you can also have text that you either append,
in front of it or in the back, or you can just insert it anywhere where you say, for instance, I'll insert all my chapters in here, right here. And then we do a little chapter list.
But it's very difficult to find, apparently. It's right there. I know where it is. I I always click on it because I built it right there. But it was very hard to find. We now expose it in a slightly different way, but nobody could find it. Nobody was using it. And I couldn't understand why, because, you know, it's so simple. I always do it when I test it. It's click, click, click. And then, of course, I know where it is because I built it. But it is very difficult to, see the product from,
the mind, from, from the eyes of, the consumer, because you're not a consumer, you're the builder. It's very different. It's, it's really hard. And I, you know, want to be critical of those people. I wanna say, I can't believe you don't know this,
but then I look inward for a moment. And I look at all the products that I've used over the years and tried, and I'm someone who, you know, for most of my life was constantly just trying new things just because they were new, and I was curious, and I wanted to learn and see what was cool about them. And I realized how often did I do the walkthroughs,
and how often did I go through the support docs or, you know, And and I knew even intuitively, yeah, I'm barely scratching the surface here. I got learned the product a little bit, and yet I wasn't willing to. And the way I put it sometimes is
people don't have time to save time, or they don't believe they have time to save time. They know, you know, if I just spend an hour here, I'm gonna spend 10 hours in the next month. They don't care. Our brains don't think that way. It's like, I'll get to it later, and you just don't get to it. And so
I it's something I think about constantly as we build. And, of course, like I said, it breeds frustration of, like, okay, well, this thing is working really well, but people aren't using it as much as we want to. How high up on our priority list is changing that? Is it, you know, we gotta fix maybe other mission critical problems first, you know, and so that's those are the trade offs that you deal with every day. But that's also the fun and the challenges then. Kinda. So
Yeah. If it was easy, you Yeah. And everybody would do it as well. So so tell me about the configurability, because that is a very important aspect of your product line. How can, how does it work? Yeah. So, you know, what we have is basically like a, like I said, the suite of content that comes out of the box. And I've seen platforms like this that give you a 100 different things or more and, you know, of varying degrees of quality. And that's tempting because we could very easily do that.
But I think that that's really overwhelming for most users and they don't know where to start. We've all had that problem before. We're we're looking at an infinite menu, and really, we can't even choose because even if they're all really good, we just we don't want that menu. We want what's essential. And so we try to be really judicious about adding things. And so, like I said, that means we need to add customizability.
And for me, like a good example is I was running a podcast, you know, a business podcast. And so, yeah, I wanted to post on LinkedIn and various social platforms, but I want other kinds of automations. And, you know, I'll go from really basic where I said, I really want, to send a note to my guest after the podcast. And, also, I wanna send a note to my guest
on how to share it. Right? And I wanna give them text to share across multiple platforms that actually adhere to the constraints of those platforms, whether it's Twitter characters or LinkedIn and the formats that work on those sites. And so those really start to add up. You're like, well, I can't can't give them all. I'm barely able to do that for myself. And so we we we wanted to be able to do is allow people to add these additional outputs in
1 by 1 or multiple at a time and say, give me this. Right? And so I add that in. We have a bunch of suggestions and we'll continue to add more there until, like, that growing menu. And then just full customizability where we'll kind of add some constraints around it to make your prompt better. But, you know, once you add it in one time, that learns over time. Right? So that'll become part of your effectively, like your workflow, your template.
You know, for me, I wanted to post social posts about the top advice that the guest gave or the top 5 uncommon insights they provided. Okay, great. I want those too. Right? I want these other platforms that I wanna post to in TikTok and Instagram. They all have different constraints and different things, different ways to optimize.
And so I started to realize this really adds up and you just don't do it unless it's dead simple. And that's kind of what we wanted to provide is that dead simple nature of letting people add things in as they want and as they you know, it's a problem of imagination, maybe a little bit sometimes.
And that's specific to that project or that podcast. Right? Whether it's their account or one of their folders. Right? Which all can have different workflows depending upon what they need for each. And so that, I mean, that was a really core tenet of us building it as saying this needs to be customizable. And while that's risky in its own right,
it just needs to cater to what you need. And what we've seen, you know, in some of the best cases is that as people's time gets freed up from having to do the core stuff, they're able to imagine more things and get carried away with it. And that's when it's the most fun is when they're saying, well, what if I could do this too? And I'm like, yeah. You can. That's amazing. You got time now. It's no incremental time to do that now. So it's exciting. But,
but, yeah, it's got a customizability is a is a is a funny word because it's it's a beautiful thing if you care. If you don't care, you're kind of like, I don't I don't wanna think. And so we wanna give people that option to not think, if that makes sense. And then, of course, as I mentioned already on the on the examples of mirroring brand voice, how do we make it sound less robotic?
And we've done a lot of things to do that generically, but then also the ability to embed examples of what you generally do, whether it's like the titles and descriptions are shown that you generally publish and imitating that.
It's a custom AI. Right? And then, you know, everything else. Here's how I write a blog. Here's how I write a newsletter and being able the the platform stores those for you. That makes sense. Yeah, definitely. And that is very clever. You know, there's, many types of, let's say, episode titles that you can do. Some people do, episode number first, and then include guest name, always.
Some people do it completely different and just say all the SEO words that, that they wanna say or just exactly what it's about or just a clever title. Yes. Things like that really are very, very helpful, I think. Also Yeah. It's interesting you say that because we're we're definitely adding in some of those capabilities around SEO optimization beyond generically, but on a podcast specific level of what you need. And then naturally, as you already you know, like, I'm I'm
all about partnerships and ecosystems. I mean, that's kind of my background as well. And I find that open ecosystems tend to win, because it allows also everyone to specialize and and really focus on even if there are overlap areas.
And, you know, we're we're equally looking to collaborate with folks across the industry where maybe they are telling their, you know, they run a course, their creator, and they're telling their people, what are the outputs you should be looking for and how should you be customizing them? And, you know, maybe they get some benefit from that as well, where, yeah, it's, this person's expert template. Right? And so we wanna provide the technology. Okay. That's very clever.
Yeah. So how do you see this? How do you view your product if you compare it to others in the industry? Like for instance, Descript or Descript or wherever you you say it. It's not their core thing, but they have lots of AI actions where, you know, your your episode is transcribed. And from there, you can also say, well, give me a chapter list. Give me some titles. Give me, you know, similar things. How do you view your competition there? Yeah. Look, I'm a happy Descript user,
and I have been. Right? And so, so it, you know, of course, it's like a a question right in the wheelhouse. Yeah. Well, I guess the simplest way to put it is I used it and that didn't work for me. Those components. And I'm sure they'll get better there. But, but but yeah. Look. I think there's an interface problem, and there's a focus focus problem that most companies face, which is the interface you're tied into.
You know, Google dealt with this problem with AI for a long time. Right? Is, you know, their business became so dependent upon search and ads. Well, they're kinda missing the boat right now on the AI growth because they didn't wanna cannibalize what they were doing or they said, look, this is our interface, and they don't wanna con con conflict on that interface.
And so I'm not even trying to come at the script at all. I I love their product, but, okay, their interface is primarily this editing interface, and it's kinda complicated sometimes to use that. And so, different different platforms have various degrees of complexity there, but that that's one area, which is like, you know, focus really matters on that front in terms of how do you make this easy for people. People say they want everything in one platform.
I I actually don't think they do at the end of the day. And it takes a really special UX to be able to to execute on that effectively across different areas. And then the second thing, you know, versus a lot of the platforms today is, like I said, there is, like, table stakes, like, how do you just give outputs? And then there's, like, I really care about these outputs, and I really care about quality, and I want the best. I'm not trying
to puff out my chest and say that we're perfect. We are not, like I said, I'm embarrassed every day that we're not where I want, you know, want to be in 3 months or a year. But that is our core focus and anyone who's ever worked at any company in their life knows, yeah, the more places you have your hands in, the harder it is to focus on that. And also there is costs, you know, using cheaper AI models. Well, it's good for your profitability.
And most users will never detect the difference, or a lot of users will never detect the difference. But for those that do, we wanna be the best thing for them. Right? And so there are other platforms out there like what we do. I tried them all, still trying them all, and I wouldn't have built this if I didn't think that there was still, you know, advancements to be made even in the short term. And so,
you know, we'll continue to think about other ways. You know, we know a lot of other ways that we want to make it super obvious if I were better. For the time being, some of those elements around sounding like you, sounding really high quality out of the box, managing all these different assets. Those are those are ones that we, can already differentiate on in really, really simple ways. And like I said, we want to abstract away that complexity. Yeah. No. That makes sense.
And and how do you see the, eventual competition of AI models themselves? So for instance, g t p four o, I think it's called, Leitzwing is, now out. Multimodel, also takes in voice slash video slash imagery. We'll see how that works out. But the the complexity layer that you guys abstract for us is moving is becoming thinner, let's say. So it will become easier for users to just say, hey, chat GPT, here's an episode file.
Go do all these things and create it all. And even, I don't know, even post it on, on all the platforms for me. How do you see that going in the next years? Yeah, it's always, yeah, it's always a risk. You know, I actually, I met a competitor of ours at a high level at the the cut the show in London, and he was telling me about the competition and race to the bottom. And I said, you know, my opinion is that our biggest competitor is doing nothing.
And the 2nd biggest right. And I think that's every company. Right? Status quo doing nothing. And the 2nd biggest competitor is the AI models. And he was actually kind of a little bit surprised I said that. He's like, no, they're not. And I'm like, no. That that is the comp we're not each other's competitors. Like, yeah, we are, you know, at a high level and sure, but we're not. The the they are they are our biggest strength and our biggest risk.
And sounds like you kind of aligned to the same perspective, but I still feel, those companies, the AI models have so much to try to get better and compete with each other. And,
the market is still so big for people who haven't used it. It's like techy people, of course, have used it. Also, there's a use case problem. And I give you, like, example. There's so many platforms that emerged in the past 10 years, you know, where software are really proliferated that you can do anything with, like an air table or Notion or all these different, you know, kind of, like, database products or even like a Salesforce CRM.
And it becomes a problem of use case. It becomes a problem like imagine. You could do a 1,000,000 things. Sure. Is that good? Like, that's not always helpful for users. They need guidance. They need workflowification, speed, expertise embedded in there. Yeah. Sure. I could tell this thing to to to give me my assets. But is it really gonna be like mine?
Not really. It's not gonna be clear. I'll never even think to use it for that half the time. It's like an app store. I haven't discovered 99.999 percent of the apps in the App Store. And that's the case for these AI models is I will not discover 99.99 percent of the use cases there. So how do we provide, you know, a really well known brand for what we do, and serve our clients really well for their specific needs, and continue to accrue value for them over time and and learn from, you know,
become a really embedded platform and basically that it ingests their needs over time. Yeah. And I guess, you we slash have some time here, because you have once you know, when chat gpt can do this easily, that doesn't mean that that that means that most people won't use that for it. Because you could do it right now if you really wanted to do. And I did. Yeah. I did. Right? Do you know what I mean? I did. And I I'm a proficient user, I would say,
regular user of it for for many things. It's a second brain. And yet, I found that that took hours. Yeah. Still, at least just to just just just to get to kinda like standard, not even to get to, like, best in class and what was really, really helpful. And so, yeah, I experienced that. I tried it. I still am trying on a regular basis to just to know. But,
but, yeah, it's that personalization aspect, and it's it's the workflowification of it all where, you know, you're not just like an AI wrapper purely. You're a workflow product. And it's like you said, there's, different models and and APIs right now are, better or worse at different things. Right? So for instance, for, transcription, I use Deepgram
for that because it's very effective. It does a very good job over many languages as well, which is very important. And it is very, very fast, which is also very important for me at least and cheap. If I could also use Whisper, for instance, from OpenAI, which is very slow and doesn't do what I want it to do. So there there's just a difference. But the GTP API is very good, and I do use that, for instance. So different models for different use cases there. Yeah.
Yeah. I I I think it's a good thing. I think it's a good thing as they get better. It's the risk. But, you know, you end up serving a specific user. You build up equity among them and show that you care about the community and that you care about what they do and their needs and their workflows. And I think at the end of the day, you start to really learn them, be them better,
and be able to take in that feedback such that you're not just gonna be taken away by one specific AI that they can use. Right? Okay. So what is, what's next? What are you guys, working on besides growing your user base, of course? Yeah. I mean, that's, of course, the number one at all times. You know, we we definitely started in, like, the kind of podcaster space, individual producers,
and agencies. And I think we're starting to see some strong, inbound and growth from folks like in the larger segments. And I think that creates additional product needs even on like the administrative side and, some of those, like, enterprise ready capabilities. But then, you know, and so that's interesting. And then, of course, there's also, you know, a ton of inbound around other b to b use cases,
and other non podcast use cases that are really interesting. I you know, most podcasters out there are many that I know, don't just do podcasting, right? They're producing content across a variety of different spheres. Whether it's, you know, oh, we just another short, there's another video content, or maybe they're like a solopreneur,
who's doing other kinds of stuff out there and wanting to purpose other kinds of content. And so we're trying to make that a lot easier, just in terms of, template templates, ease of use, embedding that expertise into the product for them as well.
And then, you know, we're working on a lot of partnerships right now, as you already mentioned, how do we embed into their products to provide value to them as an ecosystem partner and, to, you know, give them the best in class capabilities of what we do. You know, I know that those collaborations are gonna be really key to achieving, growth and scale, more AI enhancements, and then, you know, other elements around SEO,
and growth. Because I think at the end of the day, yes, saving time is fantastic. Sure. I know. Great. It sounds like me too. But the next level of that is podcast growth or content growth. And we've already seen some really interesting and encouraging things from existing customers that are realizing, wow, this is really helping me out on that front. But we wanna really double and triple down on that in terms of what do you wanna rank for and where do you want your content to reach to?
And how do we kind of even embed some of that in our onboarding to make that out of the box? And like I said, we are we see a lot of platforms out there. We're focused on a lot of different content outputs. A good example is a lot of these platforms are doing video clipping. I've seen a lot of bugginess of those that are doing that. It's or not as quality. I would only be passionate about releasing a product I think is really high quality,
that I would use. That's core tenant for me. So we're not getting caught up in that right now. Well, we're automating the identification of those, and helping you save a lot of time on that front as an expert analyst. But that's another focus area where, like, yeah, maybe in the future, maybe in, like, a done for you type of approach, but, we just wanna make these written content outputs as high quality as possible. And so there's a lot of different areas to explore around that. Okay. Very cool.
And in a way, you have all this content, you wanna put it somewhere, like, on the social media, send it in a newsletter, all that type of stuff. You mentioned integrations. Are you guys gonna, integrate, with things like social media so that you can directly post or directly schedule, or are you going to make something like that into the product?
Yeah. The the reason we didn't start beyond any kind of, like, com you know, prioritization and building complexity was there's such a diversity of content types within the platform. And once you get into the custom outputs, I mean, you're really cooking at that point. And, you know, we
provide, like, a staging document in essence. Right? It's almost like I manage my, like, spreadsheet in there, basically. It's like a full spreadsheet based workflow. Like, okay, here's the checklist of things I need, to release with this podcast. But, because of that diversity, we're basically saying, look. Well, not every one of these can be even be identified to be funneled to a certain platform. So let's first just focus on
giving them that full customization, and then we can kind of create those sections. We're saying, okay, well, this one's a LinkedIn post. We know this one's LinkedIn post. So the here will identify with the social scheduling. And a lot of the businesses we talked to candidly were saying we already have tools that do the, you know, where we use for for those platforms. And so we didn't wanna build something
additional. But, yeah, it's definitely on the roadmap. Just given at the end of the day, you do want to increase actionability, right? Of like, okay, take away my next thing. Make it really easy. Right? Okay. A newsletter as an example, we're not gonna ship a newsletter platform, but we're willing to integrate with, you know, with these little platforms to put that in there. We're not going to,
you know, I've seen some other tools actually do this, but we're not gonna build a website builder yet, but okay, fine. We're willing to, you know, embed in yours and try to make that really, really simple. Yeah. That's a good strategy. Because if you have to, reinvent all that type of stuff, that's, it's completely gonna take away the focus of of what you're actually trying to do here.
Yeah. And maybe for a very, very basic podcaster, I still think we're a great option for them, but they kind of are looking for just, like, you know, let's say you have every single meter as, like, 0 to 10. They're like, I need everything to at least be like a 3. Do you know what I mean? And just, like, I would I would take one platform where everything's a 3 kinda thing. And that's fine. Right? That's beautiful. Right? I I've I've been there before. But if you want everything to be 8
or or more, you're gonna need to use a few things. Yeah. And that's just how software is built. Okay. Cool. So, where can, podcasters and people go to find out more and start trying out your product? Yeah. We are located at flowsend.ai. So that's fl0 w s e n d, flowingandsending, dot a I, and you can try the product out for free. That's also a core tenant that we, you know, we it it's just beautiful when you see people use it for the first time. Like I said, it can be frustrating,
if they don't do the things you, you know, think will really make it even more magical. But just seeing them be surprised and wowed is great. So, yes, we wanna keep that, allow for free trials. Go ahead. And then also just given where we are and we want people to be successful, helps us with feedback. We offer kind of a free custom onboarding.
So, you know, you can schedule a call with us and say, hey. Here's what I think I want out of this. Do you have any suggestions? And we wanna help. Right? We get to we get the benefit of seeing what everyone is creating so we can help you emulate that, you know, the best of what they're doing based on the kind of content that you're creating. So find us there. Find me on LinkedIn. I'll be posting plenty, plenty of my thoughts and learnings and,
you know, I learn a ton every day. So hopefully, you get some value out of that too. Okay. Excellent. I'll put all that stuff in the show notes, of course, so that people can find that easily. So everybody go and check out flow sent dot a I. It's really great platform. Let's, try it out. Thank you. Thank you so much. Alright.
Thanks for tuning into the show. For more episodes, go to about podcasting dot show, and remember to host your podcast on podhome.fm, the most modern podcast hosting platform.