You're like, are we friends or are we rivals? Because we do something fairly similar.

But you were like, no, we're best friends. We're best friends. Hi, everyone. You're listening to Scary DevTools. You have two guests this week.

You have Paul from Itty Bit and you kind of have me as well because we're gonna be talking about how Paul just bought my project slash startup StreamPop. People kept on signing up, paying for it. We got onto like front page from Hacker News. Like and I started to get all these messages like it was like not really working. What I was expecting that relationship was gonna be was different to what he was expecting.

It's like again and again and again. It's shocking like how often edge cases crop up as

soon as you add any kind of scale. And we would have some late night phone calls like that almost like commiserating Yeah. We did actually. What the heck is going on here?

Yeah. So I I last two years ago, I was trying to build a podcast clipping tool and it was like with I think it was still it was like GPT free,

I think.

It's before three point five. So it was like relatively early on, maybe it was even longer than two years ago that people can date it back there. And then I remember I thought the prompting and the kind of like AI stuff was gonna be tricky but like I found as someone that had done like a lot of like mobile development, largely like very basic like back end y stuff, I found it pretty tricky to actually do the video processing. So like clipping things up and stuff like that. And I looked at different solutions but most of them were like targeted towards like video hosting rather than like the actual processing.

And so the recommended solution was like Lambda and I set it up and you know, set up like an FFmpeg image with Lambda and stuff like that. And it was like one of those times where like I was just going through like the Stack Overflow, like answers on how to do this. And it felt like quite janky with the Lambda stuff and some people were saying like use EC two, I tried that as well. And I was just hoping for this like real like easy answer or like this tool that would be like really good for it And I just never really came across it. And it was like one of the first times I've had that where it was like there are a lot of people asking for this and it's not that easy.

So I was like, I feel like this could actually be a real interesting space if we could just like run FFmpeg really easily and and, you know, be able to do video processing easily. And and so I played around with it with my with a friend and we we started working on something. That didn't pan out so I kinda dropped it for a little bit. Just like I'm getting to this stuff as well but just like, you know, it just wasn't a good working relationship between us. And then about two months later, it it was just still simmering away in my head.

I had like a friend reach out that was like looking like he'd he was this really young guy, like amazing developer and he was looking for some sort of work stuff. He just moved to UK and like I was like, oh, you know, like he wasn't really doing anything. So I was like, okay, know, why don't you just help me that one one day a week like I'll pay to help me like on this project. And we made like really good progress of it and he to this day like amazing amazing engineer for especially for how young he is. I think he's like he was like 19 at the time.

And so we launched it. We launched this thing and yeah, maybe I can get into like how we did it. But for now I'll just say like Ash did like quite well for like compared to anything else I've done. So some it got onto like front page on Hacker News and the the project was called like open source FFmpeg API. So just like for people wondering what the hell does that mean.

It just means like, obviously, FFmpeg, you could say is an is an API but like already but like we just added like, you know, Fastify server plus like database and queues and stuff so you But

no, it's easy to run it locally. Yeah. And then it's incredibly annoying to run it like on the cloud as a service, hence the people who do the whole end to end thing. But if you just want specific commands, it's like, oh, I've got a Mac Studio at home. It doesn't scale.

What next? It's a really annoying problem. Yeah. And also I should say that if you use the words open source, FFmpeg and API, it's like catnip for Hakanusa.

Very solid title. Yeah. Yeah. That's exactly exactly. Because I think a lot of people like, they're like, oh, that's a great idea.

Like I'm always like editing videos at home and I'm like and it's like not it's not really for you. It's for like if you want to watermark every video that a user uploads or like you wanna clip every video a user uploads or something like that. So anyway, so we we put it on Reddit and stuff like that. I can get into that later of what went well because some of that did go well. And then someone posted on Hacker News, went really well and then we did have like a paid plan.

And this whole thing was just like running on like we had like a Laravel application plus this like Node. Js like actual processing thing and it was all running just on a Hetzner server which again we can get into later on like things to do and not to do. So yeah, we got like I I think like quite early on, it it got like a paying customer which was like crazy to me. It was like Mhmm. This is I think it was the first time I'd ever got like an actual SaaS paying customer for a thing that I built.

And so that was really cool. And then and then again, we can get into more detail on this but it didn't really work out with me and my friend I was building it with for kind of a complicated reason. We could get could talk about it because probably like some lessons there in a in a general sense. Then it became just me and I kind of stopped pushing it for a little bit. But people kept on signing up, kept on paying for it.

And then one day someone signed up and I I kind of like had this just like running. I get like notifications sometimes but some someone signed up like and I started to get like all these like mess I got started to get messages like that that was like not really working. So basically, I looked at it and the person that signed up was like running like a lot of videos through this. Mhmm. And it was just a single server and like a single machine on Hetzner and basically the queue wasn't empty.

So it just, you know, that the queue is not empty. So you know, I told this guy I'm gonna look into it and then kind of quickly realized like in the time frame that like you're you're running a startup like in the time frame we've got here like you'd probably just need to use something else because well, let her self host it because it's not gonna get solved fast enough. Especially because, know, my friend had set up a lot of the infrastructure stuff. I hadn't done that. So I closed sign ups.

I think I closed that maybe slightly later but I which maybe I should have done immediately in hindsight. But I closed it relatively soon after and no one else had this problem. But I should have just closed it I think immediately at that point. And then I tried to start to think, okay, so how do I get this so like it actually works scalable? And again, I'm dive into this a bit later.

But the long story is this is kind of where I kind of lost a bit of momentum. I first tried to get someone to like help me do it, to like shift it over. And that was just like it it's quite hard to hire like infrastructure people especially like if you're just running this in your spare time like I don't have like I can't be paying people like As a as a funded founder whose entire like remit is

to hire those people, it's still difficult. So yeah, to do it as like part time, it's very difficult.

Scaling DevTools is sponsored by WorkOS. If things start going well, some of your customers are gonna start asking for enterprise features. Things like audit trails, SSO, SCIM provisioning, role based access control. These things are hard to build and you could get stuck spending all your time doing that instead of actually making a great dev tool. That's why WorkOS exists.

They help you with all of those enterprise features, and they're trusted by OpenAI, Vercel, and Perplexity. And for user management, the first million monthly active users are completely free. Let's hear from Utpal from digger.dev, a dev tool using WorkOS.

How it's designed is that you can start as early as day zero. But for us, it wasn't day zero. It was closer to when we first started monetizing because we didn't have a sign up at all. People could just anonymously use our tool. So it was a little later.

It coincided with when we wanted to start monetizing and, like, we needed a nice enterprise feature set. If you're open source and you're doing enterprise first, the minute you think about monetization is when you should think about Work OS. To be honest, if we do that again, I think we'd think about that on day zero, to be honest. Because, like, should have done it on day zero ideally. Anonymous usage should be permitted, but you should know who's using your tool.

It should be optional, 100%. It should be opt in, 100%. But it'd be great to have auth from day zero. You don't necessarily think about these enterprise features, but they still lead revenue. And it kinda is a no brainer in that sense.

So, yeah, I highly recommend.

It it it would just like it would just be like perpetually 90% done and there's always like a reason why we can't actually Mhmm.

Do it

and stuff. And so in the end, I was like, okay, that's not working. I'm just gonna learn how to do it myself. Probably another mistake I would say or not kind of a mistake because it just took me ages and in the end I I set up this whole like Terraform script so I could just scale up the number of machines. But what I realized was that it's also just being able to like adapt like be able to see what's going on and stuff just became like much harder as I did that and just like the actual all the other things that went around that just because I could provision a new machine.

Didn't like kind of just kick the can down the road which maybe was okay but like I think with video it just happens so fast because it's just like so heavy. Yeah. As you know as you know, we're gonna get on to Yeah. And so in the end, I moved it to railway and I feel that that was definitely a better option because it just like then it's like they're doing a lot of the observability, a lot of the scaling. But you know, still I think like, you know, you've still gotta look at it and monitor it.

But if I'd have done that first and but this was over the space of like a month months because it was just like alongside scaling DevTools and stuff. And yeah, so recently they the FFmpeg Fluent FFmpeg package was like deprecated that it was like built kind of on. And it just like I was kind of I've been so slow of this. I felt like it's just time to like wrap it up and then that's when I reached out to you. Still has some some users, paying customers like, know, otherwise, I'm just gonna like probably close this thing down like what is there something that's like, you know, can can it kind of transition onto the Yeah.

Is there is there literally any better option? It's kind of framing.

No. But So

I don't for for context, like during those rough times, we've like so we we met at DevTool startup

Yes.

Sorry, DevTool meetup. And you're like, are we friends or are we rivals? Because we do something fairly similar.

Yeah. And But you were like, no, we're best friends.

We're best friends because it's such a like strange niche thing to do. And actually, like our customers come from, they've typically tried to do it themselves because it feels like it's not that complicated. Yeah. And then varying degrees of stubbornness of like how far do you get before you just like give up and look to buy a solution. But you'd gotten much further than most people do, which I'm like, that's awesome.

Like I really like it. And we would have some late night phone calls like that almost like commiserating Yeah. Over We did

ask you.

What the heck is going on here. Yeah. And the reason video is just so difficult because it's like it's what, five orders of magnitude bigger than your typical kind of text streaming type thing. It just maxes out memory really quickly, especially in serverless environments. There's like memory limits Yeah.

And videos can easily and so. Yeah. They can easily sort of smash through those. So kinda built this relationship up, I guess, of we kinda mutually understand the problem even though we're kind of going after some of the same use cases, etcetera. So it was a like, it was a shame to hear that you were kinda closing down, but also I'm like, okay, cool.

Like, you've definitely got you've definitely got some users who would transition, I think. Mhmm. That's difficult when you come to you've seen lots of acquisitions, right, where and we'll come to the what this means by acquisition because it's a very broad spectrum. But it's like, okay, are those users gonna like carry on? Is it just gonna be the same company but now it's owned by an umbrella company or whatever?

For me, when we were talking, it felt like StreamPot's users were kind of our ICP really. And like whatever volume they're at, we're in a position where the more people who are doing cool stuff we've learned, especially video processing, like the better from our perspective. But if we hadn't built that relationship of of, I'd say like mutual respect somewhat, just understanding the problems. Yeah. Not always necessarily Trust.

Solving them, but just trust. Yeah. It was there's so few people doing this, like Yeah. There's not and it's not even like a secret. Think the the moat is running all of this infrastructure and hiring those infrastructure engineers Yeah.

Is really annoying and takes a long time. Yeah. You can't sort of, you know, get clawed to spit out what to do there. It's not quite it's not really code so much Yeah. Even the package that you talked about.

So, yeah, you called up and I was like, I think there's some better option than just closing it out. Yeah. Yeah. Could I before we get to that point, I think that's a good place to, you know, tend to hooks of what does an acquisition look like, but Yeah. I have to go backwards and talk about the, I guess, co founder or like the relationship aspect.

Yeah. You said there were some mistakes made Yeah. There. What do you think those mistakes were? So yeah.

I mean, if it it yeah. I'm not gonna say his name just because in case I don't know if he wants to be out there. But anyway, if anyone wants to get in touch with him, he's seriously brilliant engineer. But and if he is listening, then like I have nothing but respect for him to be clear. And like he's amazing guy.

And definitely hold all the fault with me. But I think it was just not I think what happened was we were just very different. So I mean, big age gap for one thing, just like different experience stuff, different skill sets, which I think usually is good. And that that thing was that was definitely a good pro was that, you know, to be fair like like he was far better engineer than I am at 19. He is really really really good.

And I think that there was at at first, you know, I think he he was he's like of wanted to be a founder and it started out as like I was paying him as like a kind of contractor and I saw it. And then I think that was kind of a mismatch in like expectations on like at some point when you said I he wanted to be a founder and I said, okay, we can do it as co founders. And that was kind of like a mismatch of like expectations of what being a founder was and I think like what I was expecting that relationship was gonna be was different to what he was expecting and it just wasn't working. And so I kind of was like, okay, this doesn't work. And then what happens I think sometimes if you do that, then you kill the what was working and it could have worked out as it was before, maybe.

And so I think the lesson is like I I honestly, maybe I don't actually know what the lesson is here to be honest. It's just like I think it's just really hard to find co founders because there's just so many like dimensions to it of like skill set, like just like ducat on, like life stages, what you're looking for, like principles, like and it's just so it's so hard. Like what area you're interested in, what you wanna build like.

And co founder's such a weighty title. Yeah. It's it's kind of a one way ratchet. Yeah. Can't demote from co founder.

Well, I

don't know if can you ever really demote someone in a sense? Like it's really hard I feel like. Even if they're an employee, I would imagine it's like

I've I've had experience where people have said, you know what, I'm actually not like the right person for that level up, like hire somebody like above them in a way.

They demoted themselves in this

Kind of, yeah. And it's not when there's like only two people, that it's more or could of course, yeah. But that being careful about deciding on is actually a found relationship because I think the way you described starting off with this can be ideal, which is we've not worked together before, start as kind of a contractor type relationship and maybe graduate up to it, maybe just going a bit too fast. That was like a non reversible decision. Yeah.

Maybe was a bit too fast perhaps.

Yeah. Yeah. I think maybe I was a bit rash on it. So I was like, you know, yeah. Because I I think I felt that the relationship would change if it became co founder and then Mhmm.

How much do think the podcast influenced where you've obviously spoken to a bunch of founders, sub solo founders, but I guess probably slightly more weighted towards people who are a founding team. Do you feel like having done this number of episodes now, that influences how you're thinking about working on your own product? Yeah. Definitely.

Yeah. For like especially on I I mean for instance, for like co founders, I don't think so because I I think that's just a very like I mean I think that's probably you learn more just from like hanging out with your friends and like just like I don't know being in a team or something and like who you like working with and stuff like that. But yeah, like marketing, how to like talk about products, like launching it.

Because you mentioned that went quite well

to start

off Yeah.

One was

Definitely took some lessons there and it was, you know, applied and it seemed to be effective. Yeah. And then you you hinted to come back to probably some things you may change if you could go back again. Yeah. Like what were you hinting at there around the kind of marketing growth story?

Oh, yeah. So I mean, actually, I don't think, like I think we didn't make that many mistakes on the marketing growth just because we didn't do that many we didn't do that many things. It's just like like I felt like the first thing we did was basically the main thing that we did which was just and that worked and then it was all from then on it was just like we just didn't really do any more marketing because we just had this like Because

of what? Top tip, turns out marketing is like kind of useful. Yeah. Like people will come to your website if you talk about it Yeah. And they won't if you don't.

Controversial.

Well, no. Well, no. Was saying like, people kept coming. No. Was kind of like almost like the opposite.

It was like, the marketing when we just like, I think got, I don't got lucky, the first method we tried worked quite well.

So it had so much momentum, it

was just like, then people kept coming just from that first initial kind of like launch through like the articles and stuff. So you

were being overwhelmed by that

even without Yeah. Exactly. It was like that was the that was the thing was just like I completely it was completely a technical cock up. I would say most of the most of the things I did wrong was just like technical stuff, which is like kind of backwards to what most people say of like but

This is where and talking own book, so caveat, but this is where I think video is like genuinely different though. It's not possible to do a kind of, you know, crappy MVP, like the the minimum viable part of it is still quite has to be quite well engineered, it's difficult. Yeah. Like technical challenges like bottlenecks and things like that can can flare up quite quickly. So you mentioned the architecture was sold by somebody else and you were gonna have inherited it.

Yeah. That something that if you were setting up on your own, you maybe would have made different choices?

Yeah. Think I would have it's hard because I think they were good choices if he'd been staying around and like running it in some ways. But I think maybe we should have just at least thought about like, okay, what hap Well, you know, I think I think the main thing was just Yeah. I I think if if he'd have he'd build it Like if you build the code like if you write the code like it's it's kinda easier to to like come and you know the architecture. But I think once you go live system, people are using it and then you're trying to learn how this thing is set up and like how the infrastructure was running.

And for someone that had like primarily just like stuck stuff on like DigitalOcean and just like never had any issues with

like Yeah.

Scale. That was like a big learning curve. And so I think it's I think the main thing would have just been like knowing the strengths and weaknesses of the team and thinking like, okay, if if this person leaves, you know, are you gonna be able to like maintain this? And then also like I felt like just the bar. Yeah.

As you said, like the bar for video is just like so high that like being honest with yourself like it, you know, I mean so many dev tools have issues with like uptime. Yep. Even though they've got like lots of funding and they're not doing video. They're doing you know like things that probably still like very intense and they get a lot of users and stuff. But I think it's just like, yeah, it is actually quite hard.

I don't know like not for like some people listening like, no, it's easy but like, I don't know, for

Yeah. But that's the great thing about being at DevRite, you see, starting things with like, because you thought it would be easy and you just don't give up. Yeah. It's it's fair.

Yeah. You always like, just

give up. Trolling. That's exactly it.

But I knew when to quit. Yeah. But like

Trolling through Stack Overflow answers though, it like doesn't really resolve it. It's it's different than, oh, I've got, you know, useEffect, how does that work? Yeah. A billion blog posts. Yeah.

Then you get like slightly more niche about, you know, maybe how to set up a Hetsna server. But it's still quite like generic, so there's like hundreds or thousands of posts. You get into the real weeds of video encoding and, you know, we had to hire a specialist who's like on the FFmpeg team at one point because there was just it's possible, it turns out in video for like every there's not like a guaranteed number of frames in like a chunk necessarily. There's like variable frame rates. Even within that, you have it's called like a group of pictures.

Not always the same number of even group of pictures, that sounds like it's gonna be the same number every time. Right? Turns out not. And like, there's just endless, literally endless. Which I guess,

generally maybe you don't notice, but when you start doing it at scale, it's like it happens. Yeah. It's Enough.

It's shocking like how often edge cases crop up as soon as you add any kind of scale. It's like again and again and again. And I just say for us, we've also spent two years, you know, building a product that's now robust. But during that time, like we've had us just hell weeks where the entire pipeline is blocked, like you said, by one file, just you you unpack it and explodes and like uses like the entire memory of the whole system. It's just random things like that.

Yeah. And it's not malicious users. Yeah. They just uploaded like an iPhone video or something. Yeah.

They've chosen some obscure settings in their iPhone menu.

Oh, yeah. I mean, even if you just tap the thing in the top right and make it, you know, whatever it is, Ultra HD. Yeah. You know, an hour of videos going over a 100 gigs. Right?

Oh, yeah. And then Insane sizes. We had one user a customer who was just uploading stuff in like ProRes. And like they didn't I guarantee they know what ProRes is. It's basically it's like it's not quite uncompressed, but it's like it's really really high resolution, high bit rate format.

And then they were like not sure why the like storage build was so high. And how to like explain to sort of not video experts Yeah. You know, they're they're just trying to build like a chat, basically. Yeah. Usually not.

Yeah. They don't know they don't know what ProRes is. Like, you don't know what you don't know. Yeah. And it's only by being basically a dork about this whole thing for like Yeah.

A decade that you start to feel like you maybe know some of this stuff. Yeah. But I still get surprised at least once a month by something I've literally never seen before. Mhmm. So effectively, you just unknowingly opened up that door for yourself.

Yeah. Which is fun and, you know, it's I love I don't wanna leave the media space. It's incredibly fun. And that it feels a lot more like unexplored in a sense compared to I mean, people know how to build like crud apps and stuff like pretty well. And it feels like even with like the AI stuff, there's always like there's like the chat stuff.

And then the voice stuff is like kind of and and like images and videos like always like few steps behind I think because of

there's layers of complexity. Yeah. So I'm gonna ask you a question about because you mentioned about hiring is difficult and you were doing everything on your own and aside from having other things. The the solution to that for a lot of early stage teams is to raise some money and Yeah. To build a team to then even pre product if you've got, you know, you can have a compelling pitch, you can still raise some money.

Yeah. You've got good connections like in the space and people probably would fund you as an individual. Did that cross your mind at any point?

It did a little bit when I was like when so as I mentioned, like my friend was is an incredible developer. And I think when when things were going well there, it definitely did because I felt like, okay this guy has potential to be like one of the best engineers in the world like definitely. And so on the basis of that and I felt like, you know, I could be like one of the best the go to market side of things in DevTools. So between us like, yeah, and it's a big important problem and we could build like a huge business. But then when that kind of didn't work out, then I was like, I don't think just this is such a technical area.

I think that this needs to be built by someone like who's like very focused on that for me and it just didn't feel like. I I like working with other people as well. I don't think I'm like, I can found my own like bootstrapped business I'd say. But like to do VC backed business, I think I would like to work as we're in a team. And so I it wasn't a strong strong strong like thing that I considered.

Maybe another time, business, something that I like I felt like had to absolutely absolutely have to be that have to be in the world otherwise like And then maybe I would if if I felt like that was the way to do it. But yeah, I think that unless you're trying to build not just a billion dollar company but like many more time multiples of that, like I felt like VC doesn't seem to be the way to go because I just like a lot of the time people I mean, think if you sell for a billion, you're gonna make a lot of money. But like, if you sell for like a lot less than that, sometimes people make a lot of money, sometimes they don't because based on how much they've raised and stuff. And it just seems like it's probably a bit easier to just like bootstrap it. I don't know.

So yeah. I I don't know. I I didn't. Seriously, I guess is the way to answer that. But maybe I should have.

I don't know.

I think you've hit it there though, what your motivation sort of is and was is not necessarily to build that massive organization that's, know, and and go into a scale like this somewhere in between that's an amazing outcome, but it's not returning the fund for like a major VC firm, right? That's a pretty big spectrum and there's a

lot of

steps along the way there. Yeah. If your motivation isn't to go and build that like enormous company, it's probably not the right choice anyway.

Yeah. I guess maybe the exception I can imagine myself I I don't really have a desire to be like the like a to be like a billionaire. Like I I don't that's never been something I've really particularly cared about. The only exception maybe would be like if there was something I really wanted to build and I just felt like we needed a lot of money to do it otherwise it just wasn't gonna be possible. And then I would I'm sure I would do

it but Yeah. So I have to ask about the outcome. Yeah. I'm here.

You're currently transparent on it.

I don't mind that at all. But you're currently writing massive angel checks into all the new London founders. Right? Is that that kind of acquisition?

Or giving them Itty stock as well, writing it.

Itty Bit stock. Will be worth a lot of money to But be yeah. So we spoke about what our options are. Yep. We, as a company, Itty Bit, we could definitely get some value from the user base, I think.

Transparently, your advice, like we've already been tapping you up for free anyway, you know, through that relationship. Now there's like no you can wear a hoodie, for example, without feeling like you're advertising a competitor or such these things. But it would definitely user base is like somewhat useful. It's not enormous, but, you know, it's the right kind of profile for us. And I felt like it definitely would be worth having those people come over, would help us out if we can.

And then there's some interesting technology there. It's not super robust or anything like that, but there's there's lessons there. With constantly trying to solve similar problems, the exact same problems in fact. So that that was an aspect of it. But a big deal for me obviously is we've built our relationship like Hope with Yourself.

We're not great at go to market, like we're improving, but we've been very technical and focused on product for too long. Mhmm. Like we're at the next stage of our journey, so like adding users. Here on a plate is a bunch of users. Right?

Like, see what saying? And then, you know, 50% of our ideas for marketing come from like this podcast anyway, so we might as well like get Jack involved formally. I think it was probably part of the Yeah. Part of the mindset. So do you want to talk about kind of what we've agreed?

Yeah. So well, I mean you I think we were just like dancing around. Or maybe we just talk about what actually happened because maybe it's interesting to people. I've just like I think I what I think I was like literally like drafting like shutdown emails. And then I was like, I should probably actually just like text Paul and just say like

Right. What what Just to see.

Just to like say like and then just seeing like if you'd you know, like if I tell you and then like if you were like, oh, maybe we could do some or something like we should talk. And then yeah, so I whatsapp'd you and then I think we got on a call the same evening. Right?

It's like I had a customer call and I got like I mean, as important as that is, I kinda really was like, oh, don't wanna do this now. I gotta find out what's going on. Like what's the tea?

What's the drama? Yeah.

What's the drama? But so was about an hour later or something. Yeah. Same evening.

Yeah. Very soon. And then I think I just was like I think I just sent you I think we had five or six people paying like $20 or $60. I think most I think one person paying 21 per and then five paying 60. And some of these are like active, some of them are like lapsed.

And like 350 ish like users, like signed up users Plus a waitlist of like I think current voice is like 200 people that had signed up on the waitlist since I closed it down. And then plus like we were still getting like maybe like 40 people, I think 30 or 40 people joining like this month on the wait list which I I actually feel like it's probably the most valuable part of it because like new people coming in. From our place, people with the the need and

the active intent, like they're not just been to the website, but they've like opted in to this seems like it solves my problem. And we solve the same problem Yeah. Essentially. They could just build back I've looked a few other things like that's kind of our core market is people who need to do a lot of video encoding like in the cloud, but are like fussy enough to want to like choose their own settings and to kind of have dev level of control over it. So I feel like it's a really perfect like slice of people who need that same exact thing and it's relatively recent.

Yeah. We have tons of people from our waitlist from like two years ago that just never respond to anything. It's like, that's dead. Like, it definitely has like a half life of usefulness. But the fact it's still getting active traffic was the main thing from our perspective.

Yeah. I feel like that's the most valuable.

And just transparently for in terms of value, on a level similar level to like if we sponsored a podcast or a YouTuber or like a creator, it's kind of roughly what we'd expect to get in terms of traffic from a similar kind of spend I think. Mhmm. And be a bit more generous because, know, we're friends. Yeah.

So well, anyway, so so I think I think we danced around it. Probably we're still dancing around even saying it on the podcast. But we think we danced around it for a bit and then I think you suggested £20,000. So it's like dollars, probably like $25.25. Ish thousand dollars of stock options in Itybet.

Yeah. Which people can write people can write in and tell us whether that was like if people can write it and give their opinion on like whether this is undervalued or overvalued on so

Yeah. I'd be curious to see what people think about think about that.

Yeah. But I feel to me that seems like very good, very fair and very and as I said I was this was something that was gonna be shut down. And actually I think it's a point to say to people because I think David Kramer actually made this point but it was like kind of on a different way. But like if you're just gonna shut it down like try to maybe reach out. I know the bigger point is probably like you should actually build relationships with people that might someday like acquire you or like in your competitors.

They might be your competitors now but maybe they'll buy you or like maybe you'll buy them. And if you're like kind of had a very adversarial relationship like It seems like maybe that's not always a good thing. Mhmm. Mhmm. Because I had this before where like I almost sold my last company and in the end it didn't come through but they were a competitor and we've kind of built a relationship with them.

And definitely if we had just been random people messaging them or like we were adversarial then that conversation isn't even gonna happen or maybe they're gonna screw you over.

Yeah. But It because if you're truly competitors, then you've got actually the same user base that like Yeah. Definition, like, right? And it's like, well, whatever volume that's at, I've never met a founder who doesn't want more of that kind of perfect ideal customer Yeah. Like to come and check out their product.

So it's obviously, there's gotta be some value there. If you're, like I said, adversarial, but some people are just as a marketing strategy, like aggressive, right, like online, this is on Twitter all the time, stuff like that. It's like, can you see through that and it's just marketing? Yeah. Like generally, I I don't know.

Is that the kind of user you wanna attract? Like, just there's too much drama in the world. Okay.

So you're saying you don't you're not a fan of like the kind of drama based marketing?

It definitely can work. Yeah. A, I think it alienates more people than it attracts. Or do you specifically want to attract the users who are like, you know, up on the drama? Yeah.

Or do you just want like chill people who have no time for that? Because they're just building a cool product. Yeah. And most things are not direct substitutes. Yeah.

So my general approach is to be pretty enthusiastic about competitors. Mhmm. We have customer calls quite often where we're not a great fit Yeah. And there's similar adjacent services that that are a better fit. Mhmm.

Why not recommend that? Because that person's gonna churn anyway, probably. If we're truly not differentiated and we're just doing the exact same thing as another company, like, why why do we exist? What's the point? Like, we have an opinion.

We have, you know, the way we design our API, for example, or the way that we, like, talk to customers, the way that we handle invoicing is all with an opinion. Mhmm. And that resonates with some people and it just doesn't resonate with other people. Mhmm. We're gambling that there's enough people in the world who resonate with our ways of things that can build a big business, but it just feels disingenuous to say like, no, there's no other alternatives, you have to use our things.

Yeah. It's not true. And I think the one thing you've got is like, can you be authentic and and trustworthy? If I say that a competitor is actually really solid, I I think it builds more trust and potentially we get more customers because we're pretty upfront by that. You talk to a lot of developers like marketers and you'll hear again and again, developers kind of allergic to like traditional marketing, right?

Like you book a call with a salesperson and they're gonna like ramrod you with like a fifteen minute presentation. It's like, oh, no, I'm just not interested. Yeah. That's not that's not how to sell. Most of our customers will have researched.

You know, we're a smaller player at the moment. They would have researched two or three alternatives. So we're not having to educate the market on our competitors. We might have to talk about where we're different, but that should be honest.

Yeah. If we just have

a chart where it's like all the green ticks

Oh my god. That's the worst thing you can ever do, I feel like.

You know, it's it's just yeah. It's inauthentic. Yeah. And it doesn't work very well. So building relationships is like really important, but also gaining sense for like who's our ideal customer and who's maybe Yes.

Competitor's ideal customer. Because you're gonna go to eventually, you're gonna go to conferences, you're gonna go to events, you're gonna bump into these people. It's like you've got much more in common Yeah. Than Yeah. You're both like Almost anybody.

You could

have been building this together.

Absolutely. You could have

been co founders of each other if you'd have been in different space and time.

Yeah. It makes sense, I think, build those relationships more so than to, like, attack them. And if you're building in a space that's so small that it's truly like, you know, zero sum, you're probably not gonna build a big business anyway. Like it has to be like a really big market that's growing really quickly.

Yeah.

There's far more people coming in to like for us like multimodal kind of video hosting stuff

Yeah.

Than we could possibly service Yeah. And spread across all our customers. Yeah. Like there's still gonna be increasing demand for the next, I think, five years Yeah. Until AGI shuts everything down.

Well, hopefully they'll

But your stock

will be worth a lot

when Yeah. Saw Yeah. Yeah. Good.

Yeah. I'm banking on it. So can I ask you what's next? Yeah. So I think well, I I am thinking about doing a job as well as scaling DevTools and pushing it forward from like because I I feel like scaling DevTools is valuable if I understand what questions to ask and that comes from like doing the thing rather than just like reading about it.

And so I learned a lot doing this but I also think I have a lot of gaps on like as the companies, as DevTools are growing, what things matter more than I think I've asked a lot of questions about the very earlier stages. But I think I would like to be able to like ask a lot more questions from my own problems, challenges that are like based on the things that I've had, I've experienced. So I think I'm gonna do that and like I think it'll make the show better to be honest. Because I I don't think I don't know if people can write in and tell me differently but like I feel like the quick cuts and like clever edits probably stuff is like less important than just like me asking like good questions and actually like being like scaling it up to all myself. Mhmm.

Yeah. Don't know. That's why

was saying you're like zero to one for too long, there's so many different more interesting topics beyond that that you you go join a company that's gone from 30 people to 300 people in a year. That's different set of challenges.

Yeah. But you don't know what

those questions are if you've not Yes. Experienced it. Exactly. That makes sense.

And even like a lot of the stuff that I was a bit quite naive about until I started ShroomPop because like a lot of the stuff like how do you do like quality and thing, I don't know like I just didn't really understand how difficult it is to just stay on top of like, if you've got like an open source project and a SaaS, just like manage those two things. Or like be staying on top of like GitHub issues or like deciding what features to build and stuff. I could like maybe I would hear those things and say, okay, I hear like that's important. But like I think it's different when you actually you experience it and you're like, this is actually really hard. And like you could actually get you could probably be like the best company in the whole DevTools space by doing stuff that's like nothing original in like just doing things like really well for instance in most areas I felt like because it's just so hard to do the basics really well.

Yeah. It's so hard to identify patterns amongst the products that really break out. Right? It's it's there's so many useful, like I said, free open source pieces of software that even are very popular. Yeah.

But they don't break out into big companies. Yeah. It's a sort of different thing. Yeah. And of course, some of it is like founder mentality, founder goals, like what are they aiming for?

There's a lot of founders who would like to do that thing, who just who don't. And I I think one of the things that this podcast is really good at is unpicking somewhat like the founder psychology or the Yeah. What how do they think about decisions? And that does like scale up. It's like these these different frameworks and things like that, which is pretty cool.

But if you don't know what the next step is, so it's useful for me as a founder who's we've raised a pre seed round, we're growing at this point. We want to get to that next level of like series a, series b, beyond. Yeah. You can talk to other founders, but to listen to like, you know, how many of you you've interviewed? What?

A 100 people probably, somewhere. Yeah. It's over a 100 episodes. You've done some some Some repeats. Some repeats.

Yeah. Over

a 100.

Yeah. It's it's useful. Yeah. But you don't wanna just be repeating and retreading the same ground

No.

Over and over again.

No. Yeah.

I get that.

Yeah. You gotta make it fun for yourself. Right? This is this podcast is for me. Like, that's it's I'm the I'm the in many ways I am I am the audience.

I just ask questions that I'm interested in and so I need to keep finding new things that I wanna get answers to rather than Like because I sometimes like I was like, just feel like you know all the stuff, you just need to do it. So I need to do more stuff so that I have more questions. Yeah. Because right now I feel like I just know. And and not that I know everything but like I know enough.

I just like need to do it because there's difference between what you hear here. Unfortunately, I wish you could just like learn it all from a podcast.

The the academic aspect of it as well. Like there's so much bad advice. Yeah. I wanna sort of say, as a founder, you should ignore, like, pretty much everything that you hear from, like, people who are not founders, and that includes, like, probably your investors or, like, customers can have good opinions, but they definitely understand, like, the full picture, etcetera. Yeah.

There's really something unique about hearing from a bunch of other founders who have had to make the same, because there's always trade offs. Like, it's Yeah. It's a simple decision if there's only one good option. Yeah. Yeah.

Well, cool. Everyone can make that decision. Yeah. That's very rarely the case. There's almost always trade offs and hard decisions and Yeah.

You sometimes have to shield your team from them. So it's like that's comes back to the sort of co founder question a little bit, like, who can you bounce that off a little bit? So other founders who've really got and then you get into the weeds and ask like tactical questions and quite

like How do you do like news post or something?

Yeah. Yeah. To be fair though, if it was the tail out well, he's just like just he looks at every still, every PR personally. Yeah. Despite sort of Latter of Heralds like sailing like crazy, I think.

Yeah. I I have would just stopped the podcast there and just kept asking how over and over again. Yeah. That seems like baffling. So that's inspirational though from I'm like, okay.

Yeah. Every investor was like, you shouldn't be working in the code. You should be like, you know, build a company, not the product. Yeah. Was like, Yeah.

Your product ends up being a bit shit and it should be like really good.

Yeah. It's very hard I think. It's like there are like just like these giant figures that are like just really good Yeah. Engineers and Yeah.

And they they reject like those like heuristics that you'll read about in, you know, Twitter threads or something. Like, no, this bullshit just you need to like look at every PR or something. It's like, it's cool that most founders are just basically stubborn Yeah. Or certainly opinionated, like have a good vision. And there's not there's not a way to get hold of that.

Like conference talks aren't really the same thing as like digging deep and you know, you can't ask a follow-up question for a conference talk and things like that. So I'm just glazing you at this point but I'm like, come on, keep doing the podcast.

Yeah. Don't worry. It's not gonna stop. It's it's gonna keep going. We're coming towards the end actually.

So maybe we could just do we got some we got some takeaways like lessons. What what can people learn? What or maybe maybe you can sum it up on what do you what do you think people can learn from from the StreamPot experience?

Personally, I think you should have raised money. Okay. Once you started to learn about, like, the technical challenges of it, there's some and we could have a whole episode like what makes a really good bootstrapped business. We've got like mutual friends here in London who've like grown insanely good bootstrapped businesses. That's like a world that I have a lot of time for.

But I chose to raise money for it to be because we knew the technical challenge was just we had to spend a lot of money on Yeah. Basic product and infrastructure ahead of being able to solve the problem. Yeah. So guess to generalize it away from StreamPub is just make a decision, are you building a really technical product? Yeah.

You might need to hire more than one engineer and if you're in the fortunate position to be able to sell fund that, great. But most people probably have to raise Yeah. At least a small amount of money there. Yeah. I don't think it necessarily locks you into the billion dollar outcome if it's like accelerating money or pre seed money.

Like, there's definitely angel investors who are happy for, you know, five x, 10 x return. Mhmm. And obviously, if you've taken less money, the the 10 x is a reasonable Yeah. Like an achievable Yeah. Sum.

Yeah. So maybe there's a lesson for people who are currently not raising money, but maybe they should. That's the rarity. Think most founders probably shouldn't, especially you can get so far so fast now. But I used to be very like bootstrapper.

I don't think having seen both sides now, there's definitely reasons why like raising money would make sense.

Yeah. And that

would have helped you Yeah. Along the way Yeah. To some extent. The one way decisions versus like reversible decisions with the founder Mhmm. I think that's really interesting.

I'm thinking about loads of things I've done, like, oh, no. Can I roll that back if it was a mistake? Like, it's definitely something you just have to commit. Right? You have to like decide and make, you know, without all the perfect information.

But some decisions need more time than others, I

would say. That was one.

Yeah. That was one. You because it feels like it wouldn't have damaged the relationship for you to push that out by three months or six

Right? No. No. Just see what happens.

Yeah. And be friendly with competitors, I

guess, is

the main one.

Yeah. I think that's good. I think that's very fair. Yeah. Perhaps I I should have raised money.

But we will never know. Maybe there's a parallel timeline that like, I'm like, we're filming this in my Rolls Royce or something.

That would be a very nice way to do a follow-up episode. So I look forward to that. Have

a tacky as well.

Oh, yeah. So right. Self driving Rolls Royce around London. Perfect.

Okay. Where can people learn more about Itty Bit, which now I'm I've got some skin in the game on Itty Bit, so go people should go.

Four four follow-up point. Yeah. Marketing is good, it turns out, and we should do more of it. Yeah. Honestly, ittybit.com is, I think, a decent summary of what we offer.

If people have an upload form on their website, I would love to chat to them and kind of learn about, you know, what comes in there and how we can make that more efficient. But we have settled into the tagline of just automating all the boring media stuff. It is a rabbit hole. It's one that we've, you know, jumped into willingly, and we really think it's fun. But almost no team should be doing a lot most of the stuff themselves.

It's a little bit of a tarp hit for money and time, and it's especially it's catnip to engineers because it is like this the idea of, like, being nerd sniped, yeah, is really novel in treating problems. So if you have, like, a smart engineering team, they will convince you that they do need six months to build this stuff. And you probably don't. You should probably come check us out,

think is the the better option. Love from my experience. Yeah. Okay. Amazing.

Well, thank you, Paul. That was a very interesting episode, I hope. Well, it was for me. Was fun. It's like therapy.

I hope that people enjoyed it. Let us know.

So at worst, it was cathartic to you. Hopefully, the listeners found it interesting as well.

Yeah. They're gonna be like, this guy Jack is an idiot. I'm not gonna listen to him anymore.

What? No. If they think that, they should write in.

Yeah. I think tell me.

I would love to hear all the other advice that they listen in and they were like screaming, like Yeah. Just thinking Well, done that. Should have done that.

Yeah. But actually, you know, like one thing I will say about I I do wanna do a dev tour myself again. And I will join I'll probably join a job but like if people are interested in like brainstorming and stuff then I'm very open to doing like hack days or something. So that would be a cool cool thing.

And you're on the job market. Let's all finish with that.

Yeah. Open to offers. I think so. Yeah. Okay.

Thanks everyone for listening and thanks Paul.