Brian Sierakowski (@bsierakowski) has been busy over the last year: he started working on ChangeBot and TRMNL, and both projects are taking off.
If you ever wondered what a good changelog looks like or why you might need one, this episode is for you.
00:00 – Arvid (None)
Hey, it’s Arvid. You’re listening to the Bootstrap Founder. Today, I’m catching up with Brian Sirachowski. If you ever wondered what a changelog can do for you, this conversation is for you. Brian shares his indie hacking journey of 2024, building a SaaS in the changelog space as well as a hardware startup at the same time. Ever the true indie hacker, he shares behind-the-scenes insights from this zero-to-one stage of two amazing businesses.
00:33
This episode is sponsored by Paddlecom, my personal payment provider of choice. I get to work on my business full-time because Paddle takes care of all the money stuff. They have great APIs, a massive analytics and insights platform and Paddle takes care of all the money stuff. They have great APIs, a massive analytics and insights platform, and Paddle actually acquired ProfitWell for this a few years ago and, funny enough, Brian was the CEO of one of their main competitors back in the day. In the SaaS world. We are all connected and Paddle makes it easy to get paid. So check them out at Paddlecom, Brian. Welcome back to the show. 125-ish episodes ago, I think episode 238, was it In early August 2023 was the last time we talked and we had this very enjoyable conversation about what you had been doing up until then, and it ended with a cliffhanger that I really enjoyed about that chat. You had a few entrepreneurial plans, so let’s catch up and see what happened in the meantime. Brian was 2024. Good to you as a founder.
01:30 – Brian (None)
Yeah, yeah, I think so. It’s so funny. I’ve done kind of entrepreneurial stuff for a long time now. I’m also the entrepreneur in residence at our local accelerator, so I’m kind of like constantly talking to people in that zero to one space and I feel like every single time that I go through it I’m just reminded of how much harder it is and it takes a lot of time, and I feel like every single time you move into that space you realize how much you don’t know, Because it’s always from the outset I’m like, yeah, I know what I’m doing, I’ve built businesses before and I’ve been successful and I failed in the past.
02:09 – Arvid (None)
But I’m not going to fail again in the future.
02:13 – Brian (None)
How silly would that be to fail again? So yeah, I feel like this past year in particular, a lot of the things that I’ve had the belief around even around, I’d say, before this year even thinking about founding team, I was very kind of like figured out yourself first, and I think this last year I’ve been like, oh, we’ll build a team around you first, kind of understanding the value of co-founders and again appreciating everything taking time. And I’ve had a little bit of cross-training, because I’m usually like a B2B SaaS guy. I got a chance to work with a hardware company that’s B2C, which is totally different and in fact, way harder. So I’m kind of running back to B2B SaaS. I’m like get me out of it.
02:59
Hardware is like we need more batteries and there’s one factory in the world that makes batteries the way that we do and we can sell more product when that factory thinks it’s appropriate to ship and then we have to get it shipped. It’s all these things of the delivery of software is not like it. So, yeah, I would say this year has been a whirlwind, but it’s been great, and I really feel like there’s been a lot of and but it’s been great and I really feel like a lot of. There’s been a lot of culmination going on.
03:26 – Arvid (None)
Like a lot of the a lot of the lessons that the, the universe, has tried to teach me over the past couple of years are like actually, I feel like sinking in in the past, uh, in the past year oh man, that’s great to hear and, yeah, that that sounds like a lot to deal with and a lot of different things to deal with, like very opposing things b2b, sb, saas, b2c, a hardware thing let’s talk about it all. Let’s dive into the details, because those lessons that you have learned and are now applying, I would also like to learn and apply. So you know, and probably everybody listening. So, yeah, let’s maybe start with the SaaS here, because that was, I think, the idea that you had or that was kind of just brewing in the background back in the day. Right, yeah.
04:04
So what was the idea back then and how did it progress from there?
04:07 – Brian (None)
Yeah, you know, I kind of I did the thing that I feel like every SaaS founder, especially like multi-time founder, does of like you get to the end of a journey such as it was and you kind of think about the next thing, and you’re not necessarily thinking about diving back into entrepreneurship, but you kind of like at least this is what I do, and I think it’s pretty common of like you kind of look back in your recent experience and be like, well, what was really frustrating and what sucked, and maybe also, if you’re lucky, what’s changed in the world such that we can solve that problem.
04:41
And so I thought back to my time at Team Password and we were constantly owned by not updating our changelog, and I remember it happened a bunch. But I remember we had this one sales conversation and the guy I was working with was really nice and it was this enterprise that was really way outside the size of who we should. They went with another company. I think they went with LastPass, but the person I was working with was really trying to do a good job at bringing Team Password up in the best possible light, and so he shared the spreadsheet with me of here’s how we compared all the different tools and it was like this might be my memory, but I remember product frequently updated was the first item on the list and we had a big X through ours and it wasn’t the case that we weren’t updating the product frequently. So we weren’t updating a change log and we weren’t doing those extra steps. And then I also fast forwarding a little bit my experience at Bear Metrics. There’s a whole episode about the price increase. But I think back to that and it’s like we were really working. We were doing a ton of product work and we had this big bug backlog and when we did the price increase we got totally owned Again. Anybody who wants to revisit that? We’ve got a full episode on it. But I just think back of like well, what if there’s a world where we were like once a month updating that, or what if it looked like? What if we were updating that changelog like once a week? Or what if we had like three to five updates a week in that changelog for like a year? And then when we announce a price increase, there’s like 100 updates? That’s a much different like we didn’t have to do any more product work. We had to tell people about it. That was the problem that stuck with me. I’m like well, can we solve this? How do I make sure that Brian in the future never gets owned by a changelog-related problem in the future? That’s really where all the AI stuff came in.
06:44
That was the switch for us, because the idea before that was one of the problems with writing change logs is that it’s very unique work. You need to be technical enough to understand the changes. You need to understand the customers. You need to be able to understand what the product team. There’s a whole process that this went through, and then you need to have enough sales and marketing so that you can take that change and phrase it in such a way that here’s why we did it, here’s why your company should pay us more money, or here’s the business benefit to making this change, and everybody is super busy.
07:17
So the idea before we got to what we’ve landed on right now was well, we could centralize all this information in one place. Right now, there’s conversation happening in GitHub, there’s conversations happening in Linear whatever PM tool you’re using, there’s Slack conversations and then there’s actually code being written. There’s pull requests and diffs, actual lines of code changing. So that was kind of our first idea of maybe we can centralize all these things in one place so that a human could possibly do it. Not that they will be able to be successful, but it’s just literally an impossible task if you don’t have that information in front of you. And so then AI comes along and you get the audacious idea of what if the changelog just writes itself? What if we could train an engine? That the first version I kind of make the joke that we’re on update engine six right now, the first version. We’re like hey, what if we just ship a bunch of commit messages over to OpenAI and say write our updates? And that obviously sucks for a lot of reasons. And we’ve gone through tons and tons and tons of changes.
08:25
Where it’s like, oh well, you just like. I think that’s one of the cool things about the AI development cycle is like it becomes very apparent where the gaps are. It’s like, oh like, this update’s just like doesn’t have enough information. So, okay, cool, we need to bring in more context. And then it’s like, oh like, this was technically correct, but it doesn’t understand who our customers are. It’s like, okay, cool, we need to load in context about the business.
08:51
And then like, oh, this update is correct for the customers and correct for what the changes, but it doesn’t understand the history of the product. So, okay, we need to build in, kind of that. So it’s like pipelines on top of pipelines and eventually you get to the point where, hopefully, I think we still have, like I say, the update engine’s 80% of the way there. But you get to the point where it’s like, oh, I remember it was earlier this year the first update was written and I’m like, oh, this is like I would share this with people. Holy crap, the computer did this for me and doing that through a bunch of different changes there. So, yeah, sorry, it was a little bit of a ramble.
09:25 – Arvid (None)
That was exciting, Accelerated yeah, journey.
09:28 – Brian (None)
But yeah, it’s kind of where we got to today, of getting to a product where we’re still in the phase of like, I want to get to the point where this is a tool that’s indispensable to people and something that also can kind of operate hands-free, and so that’s kind of the phase we’re in now of kind of going to people and trying to convince them. I’ve also learned a lot about AI. An interesting insight that I’ve had recently is that people are pretty burned by AI. I was talking to a guy just earlier and he was like yeah, I think your product’s really cool, but I just don’t believe it. And I’m like what do you mean? And he’s like well, I’ve signed up for a bunch of different AI stuff and they all make all these huge promises and none of them ever worked out, and so we’re actually building weaker versions of our product, not because we can’t automatically.
10:25
We can read the code changes, understand the diffs and actually write the updates, even if the commit messages are terrible, which they frequently are. So we’ve actually had to take a step back and be like okay, we kind of need to. People are just maybe not going to believe that the product can do that. So we need to walk them into okay. Well, what if we just build a really nice change log that is hosted for you and integrates and does all this distribution stuff for you? But that’s been a really interesting learning as well. We’re in the stage of the AI hype cycle where people have tried AI stuff and they are chat GPT wrappers that don’t actually deliver on the promise, and so now we need to handle that from like a marketing and from a product perspective.
11:10 – Arvid (None)
I’ve heard this a couple of times. I’ve actually seen this with several people that I talked to recently is that people just do not trust AI because they’ve been burned, and to be able to convince them, you effectively need to give them access to the product for free, because otherwise they just do not believe it. Like even if you show them a demo, they’re like ah, this is just a demo, this could be like wired together in the background. They want to have this magic moment that they so desperately it’s still kind of internally believed can exist but have been burned away from. And that is actually a big expense for founders. That makes it really hard to calculate stuff like what is the cost of acquisition here? Do I need to give this person access to the full tool? And they need to be able to spend $50 worth of open AI credits just to be able to get that one first magic moment. That is a problem. How do you deal with this kind of stuff?
12:01 – Brian (None)
Yeah, I think that, well, one of the interesting things is in our strategy to stair-step people into the product. All of the most expensive AI stuff is kind of at the Our full imagining of the product is also the most expensive from a credits perspective. So the initial phase zero is like, well, if you don’t have a changelog, that’s public. I think that’s really important to us. At the outset of this project, I wrote a 92-page blog post about why updates are important and all that sort of stuff. So it was supposed to just be a one-pager to. Anyway, you know how that goes Somebody who’s written a book before, maybe by accident. But yeah, so that was the of the initial thing of like, well, why don’t we just go into the space and provide a change log that’s easy to update, and that one of the initial ideas is like we undershare these updates.
12:59
So kind of building in a lot of the distribution stuff. We’re like, okay, people are barely publishing enough updates. But it’s like, okay, well, are you sharing it to social? Are you resharing it to social? Are you capturing an email list? We have this full suite of distribution-style features that really have nothing to do with AI but are necessary to achieve the goal of the 92-page blog post and then stair-stepping people into it. So the next level up the middle tier, and this is how we’re doing our pricing as well, at least thinking about it now. The first one is just a change log and I don’t remember the number we put on it. It’s like $40 a month or $30 a month or something like that it’s not too high.
13:43
Yeah, yeah. And then the next one up is like okay, now we’re going to put the power in your hands and so you can connect your GitHub account, you can connect your linear account, but you can go through. And I don’t know if we should be sharing this analogy publicly, but it’s really made sense to us. You can create a shopping cart of like, hey, I want to write an update, here are the linear tickets that are relevant, here are the pull requests, or here’s the commits that are relevant. And then, as we add more integrations in, it’s like hey, here’s the Slack conversation or whatever. And then I want you to, I’m going to direct the AI and say like, hey, this is what we want to write the update around. And so that’s kind of like the middle ground of like there’s a human in the loop, but there’s that frustrating thing when you go to write an update, you launch the page and it’s a blank sheet of paper and you have to write it. So that’s the middle tier. And then the tier, that where we started and what we thought people were going to want, but it turns out people are most skeptical about. There’s this autopilot feature of yeah, we’ll just monitor all the changes that are happening within your company. We will do an evaluation of whether or not this is worth writing about, and that really comes down to do. We think there’s customer value on the other side of this thing, which is obviously a little bit fuzzy and something we obviously there’s obvious room for improvement there, and then we will tell you. We’ll serve as your junior product marketing associate, and this helpful little bot is going to bring you updates and say hey, if you want to edit it, you’re welcome to edit it, or you can just approve it through, and so that’s the premium tier, that’s the $4.99 a month version.
15:22
And so we came to that idea, not through a means of monetization, but just from like, well, if we can’t get people to trust us with writing the updates at all, then it just feels like it’s just a big leap to be like hey, it’s just the kind of startup problem too. It’s like, hey, we’re this startup. I’m Brian. You might’ve seen me online. You might’ve seen some positive things or some negative things, but here I am and, if you don’t mind, we’d love for you to sign up for an account, create a password and oh, by the way, we want you to connect your GitHub and we want to connect your project management tool and your CMS and then maybe we can deliver you some value.
16:00
It’s exactly what you were saying. It’s like that moment of epiphany is behind. Oh wait, I have to give you access to all my source code. What are you doing with this? Or a lot of times we have people like, hey, can you just read the commit message and not the diffs? And it’s like, yeah, we tried that. Man, I’m sorry, we wish we could do that too, but we actually realized we needed a step in the process to actually review the diff and understand the nature of the code being changed in the context of who your business is, in the way that a human would. They wouldn’t just look at the message and be like this is what it is. They would probably have a meeting which sucks with one of the developers and be like what is the nature of this change? Why are we doing this? And so we kind of have to do that stuff.
16:47 – Arvid (None)
Yeah, and then the developer would translate the code that they have in their mind into something that can actually be communicated, which is what the AI does too. It brings me to another point, because the first thing we talked about was people being skeptical of AI in general, of the promise of AI and the results that it can deliver, but sometimes does not if people don’t really put the effort in which you’re clearly doing Like. This feels like a very, very interesting like I’m, as a software developer and entrepreneur, I’m super excited by this. I’m trying to contain my excitement for this conversation so we can actually talk about you know, please, I don’t want to blush too much, you know, but no, it’s really, really cool because I also struggle with this. So it’s a pain point that I feel right now.
17:27
But I also feel the other side of things, the entrepreneurial side, where people are like AI we don’t trust it and AI you can’t trust it. That’s the other thing, the whole privacy angle of things, and I talk to a lot of people who run AI-assisted or AI-enabled businesses that have, for that reason, went into local LLMs and on-premise deployment or private cloud kind of stuff. So where’s your perspective with ChangeBot there? Is it going to stay a SaaS or are you going to go into the enterprise and kind of, oh no, you run your own AI, kind of thing?
18:03 – Brian (None)
Yeah, that’s a really great question, and so there’s kind of like two different directions on that. So, just in the sheer choice of tools, it feels like we are. I don’t know it changes every month, but it feels like using models specifically trained around a specific task will get you the best results. So, for example, we should use a model. If we’re doing code comprehension, we should use a different model than the model that is writing the update and taking all that data and writing prose, and so I think that’s kind of one big piece of when you talk about getting where are we going to host this and where’s it going to be and what’s involved. I think a big part of that is how, again, our very first goal was the updates need to be well-written, because if we don’t get there, then it’s like nobody’s going to use the product. We have three product pillars. One, the updates are great, they’re well-written. And then the second thing is that there are a lot of them, because if we can write a great update, but it’s only once a year, it’s like, okay, we can beat that. And then the third thing is that the updates are everywhere, and so it’s like, well, you have all these great updates that are well-written and there’s a ton of them, but if nobody ever gets a chance to read them then you don’t get the value. So that’s kind of a part of it.
19:22
From the technology perspective, we want to be really I don’t know, and again it’s like the next version of OpenAI or whoever next state of the art could come out and maybe the code evaluation in that new model is better than all the other ones and you can kind of go back into one. But I think that’s kind of a core consideration from a security perspective and from a data perspective of which tools are you using? You get some benefits also out of using these big models, because it’s like some of the things that if you’re going to locally host an LLM, then you need to be worried about leaks and you need to be worried about things like your model sharing data from one customer with another customer, and you kind of run into that area. So we’re kind of of the perspective of we’ve done a lot of work in this space, but we had one of our potential customers be like, hey, can you share a document with me of what are you doing from a security perspective? And I’m like, oh yeah, that totally makes sense as far as why you would want that and we should write that up. But we’re kind of going through the process right now. Maybe by the time this launches we can have that to point to as an example.
20:31
But there’s all this stuff you can do as far as like, well, what do you encrypt and what do you store? And for us, we don’t store any of your code because we don’t need any of your code. And we also have kind of the special benefit of like, because we’re writing change log entries, the end result of what we’re generating is meant to be public. So, like even as we are going through the process, like we’re not asking for anything, that’s like we’re kind of almost intentionally turning down any sort of tech talk that might indicate what’s going on behind the under the hood, because it’s like that’s also just just not relevant and it’s like an error case. It’s like a failed eval for us if the update is too technical, which sometimes it will do, if the update is too small and the LLM is just looking for stuff Like oh, we have to talk about this a little bit more. But yeah, I think it’s really important to do all the normal data security stuff. So it’s like don’t store more information than you need, obviously, and anything that’s sensitive, it needs to be encrypted, and you also need to make sure that you are conscious of all the new LLM stuff. So leaks are a really big thing of data from one customer. It’s kind of funny to consider this because this was the whatever. 10 years ago in web development, it was really common for data from one customer to get into another. The idea of having a multi-tenant web application was a new idea, and so it almost feels like we’re back there and I feel like the same lessons that we learned 10 years ago on the web. We need to do the same thing.
22:06
The one thing that I might we haven’t done this yet, but it’s maybe to your point of joining those two topics together of the right tools plus the right process is we thought that once we have a bunch of customers and we have people editing the updates so we’re delivering updates and maybe they’re editing them it might make sense to do a unique, fine-tuned model for each of those customers, just based off of the style of that. It’s wholly outside of who your company is and what you do and anything else. It’s just like this is the style of update that we like and that’s kind of like a nice final surface layer that we can apply. That will just. This is like we know what the inputs were and we know what you want the outputs to be, so that’s kind of like a nice fine-tuning situation to be in. So that might be a situation of like maybe that final mile goes with the customer. Yeah, even where to host it is kind of an open question. But I’ve also kind of learned too it’s kind of the classic thing of like people don’t want to deal with this stuff, which is why they’re not doing it now. So I think for the foreseeable future, regardless of what the infrastructure decision is, we will manage it.
23:18
But we have gotten a lot of it’s funny you mentioned. I’ll say one last thing. You asked one question and you just set me off man, one of the big questions we had is who is this product for? And those were the two paths. I spoke with one woman who runs a. Their company has 1,000 employees and they have eight people in the product marketing and that’s a huge product marketing team. It’s like them versus the world, and so that’s where we really got a lot of the insights, of just centralizing For them.
23:56
It’s like if we could just pull, even without doing all the AI stuff, if we just pull all of that information about what’s going on inside of the organization into one place, that would be super useful to them. That’s probably an open opportunity and that kind of takes us in a top-down product strategy where we’re like okay, well, what systems do you have? Where is that information? What format is it in natively? What format do you need to be in? And we kind of custom build a product based off of that enterprise and there’s some common shared core across all of those different products. But it would be like a very like integrations and kind of a. You know, there would be like a three-month onboarding, you know, it’s like it would be like an enterprise sales thing. And so we’ve actually decided to move more for smaller teams that are. You know, maybe they don’t have a thousand employees, maybe they don’t have 1,000 employees, maybe they have five.
24:52
But the thing that I went back and forth on this for a long time, the thing that we did is that we realized that this would allow us to do a bottom-up approach of like okay, well, what are the core features that all of these businesses care about and how can we automate as much of this as possible? Not that we’re not trying to do customization, because that’s the name of the game. It’s like hey, we have somebody. They’re like, hey, we love this autopilot plan that you have, but we don’t use Linear, we use Jira. We’ll be like, great, we will build that Because other companies use that as well. But that was the key strategy shift. We wanted to be much more focused on the bottom-up approach of really discovering what this product is and what it should do. Even all the distribution stuff came directly out of customers.
25:40
The feedback over the past eight months was like update sucks, update sucks, update sucks. And it’s like, oh, update’s good, need to be able to edit it. It’s okay, can edit it All. Right, now I need it to be everywhere. It’s like, okay, great. And then now it’s like, okay, I need to be able to capture an email list and send an automated monthly update. It’s like okay. So it’s like as soon as you clear the one hurdle until you clear the hurdle, that’s all you hear. It’s like. It’s just like update sucks, you know which is which was fair feedback and they did suck, um, but then once you clear the hurdle, it’s like every single time you jump the next one, you get, there’s another, there’s another one coming, and I know once we build all these features, there will be, there’ll be another, uh, another request on the way, for sure.
26:22 – Arvid (None)
Yeah, and it does sound like you have found several, if not very, very many, points of complexity where going into the complexity just opens it up for more complexity, right? The example of you’re getting no’s, no’s, no’s and then you get a yes and after that yes, all of a sudden it’s a whole new world with another set of surprises. Man, that sounds like a lot of layers that you have to work through and I work with AI tooling as well. I know those layers and the weird, idiosyncratic kind of stuff that happens in there, plus then trying to also solve this for enterprise customers who have this very we want this and this is the only way it’s allowed to be mindset.
27:02
That is also a challenge for a startup, right? A startup that has to deal with custom requirements on every single level, the integrations that you talked about. Man, that must be so hard. How do you deal with this? Because it feels like if you want to go for smaller customers or smaller businesses, that still probably doesn’t solve the problem, with people choosing very different tools for similar jobs. Are you going to just take the top five product management systems and the top three code repositories and deal with that? How are you going to approach this over time?
27:42 – Brian (None)
Yeah, that’s a big change in my mentality. I think especially maybe not in the past year, but especially five, ten years ago maybe not in the past year, but especially five, 10 years ago I think that I definitely had much more of a headstrong mentality of I know what the market needs and I’m Steve Jobs or whatever.
28:00 – Arvid (None)
And I’m wearing turtlenecks.
28:01 – Brian (None)
I’m doing the whole thing and I will tell you market, this is what you’re missing out on and this is what you need. And now I’m just kind of like the thing that I always come back to, because running a product and building a new product is very challenging Both the actual what to do and how to do it, but also just keeping motivated when someone tells you that your updates suck or whatever, or like, oh, we don’t need this, or whatever the case is, the reality is, even if you get to a million customers, there’s another billion companies in the world that said no to you. The thing I always come back to is I never want to run another company again without having a constant stream of updates for the marketing team, for the sales, whatever. So because of that, that’s kind of my core and I have those three pillars. Those are the non-negotiables. And then everything else is like if a customer comes to me and be like, oh, I’d really like to do this At this stage we have few enough customers.
29:01
I can have one-on-one conversations with everybody. My CTO, my co-founder’s name, is Ara, and so he’ll probably listen to this and slap his forehead, but it’s kind of like we kind of just do it, as long as it’s in line with those product pillars, it doesn’t take the product in a different direction. It’s like, yeah, if we have somebody, we’ve already gotten a request for GitLab instead of GitHub, which is insane to me because it’s like, how many customers does GitHub have? I thought it was all of them. But it’s like, okay, cool, and maybe there’s an argument there’s even less tooling in GitLab. But it’s kind of like, yeah, once we have this system set up and we understand the nature of Git, it’s like, yeah, we’ll do GitLab. It’s all about prioritizing them.
29:48
But I kind of feel like, especially for us, and maybe it’s also this type of product where it’s like, the more information that we can pull in, the more context we can get in, the better the result is going to be and also the more time savings is going to be to the person. So it’s going to be really great updates and they don’t have to really edit it at all. So it’s kind of almost like, because of that, we’ve sort of resigned ourselves to like, okay, we’re just going to be building a bunch of integrations and it’s hard enough to convince somebody that they need more updates. And once we find somebody who understands that pain, to be like yeah, we’re the person to help you with that. We’re not going to be like oh well, you should also switch off of GitLab.
30:28 – Arvid (None)
Don’t force them to do these kind of things, right? You just use what they use.
30:31 – Brian (None)
Totally so. I think it’s just kind of like and we can only move as fast as we can go, and we also my mentality is sort of working from the center out or the bottom, whatever Pick another analogy that I’ve made so far but it’s like we want to make sure that our current customers are as excited about the product and are using it. We can see how frequently they’re publishing updates and that needs to be happening, and if that falls off, I’m going to reach out to them and be like what’s going on and why is that? Or like, oh, I was going to get to it but I got busy. I’m like well, why is it that big of a task that you need to?
31:09
So, starting with those people, and then we have people starting trials and coming in all the time Basically everybody. They first connect a marketing site instead of their product because they’re like what is this thing? And so that’s something for us to work around and maybe that’s a feature, maybe we provide updates around the changes to your marketing site, but then it’s just a function of like, okay, or maybe they don’t have permission to the repo, all those sorts of stuff. So that’s kind of all the stuff for us to work through. But yeah, we just kind of make a backlog and I’m sure at some point in time we’ll need to say no to an integration. But I don’t even know if that’s going to happen, and it’s like never know, it’s like not in 2025, I don’t think.
31:50
But it’ll be something where I think we’re just going to have to get a lot better about having sort of more transparency around, like where Someone asked me today like what the roadmap was and I was able to kind of like rattle off a couple of things and then I forgot one thing and I’m like, oh, this is actually in position three and everything else got pushed over and so could get to 50 requests pretty quickly. And then at that point we need to probably do a better job of just being like, hey, we can’t support whatever. We don’t support WordPress. You can’t publish to WordPress right now, and with what’s going on in WordPress, maybe we might wait a little bit, but that’ll be something in the future as far as how we can communicate around that.
32:30 – Arvid (None)
That makes sense, and integrations always are a lot of effort to put something into the system, because you always need to deal with the little tiny issues there and the slightly different ways that they do it compared to somebody else. It kind of makes you think as a founder, that if you built a tool like what ChangeBot is, that has a lot of data sources it needs to pull data from and has a lot of destinations it needs to send data to, your integration kernel like the thing in the middle that allows you to integrate into and then distribute from needs to be highly flexible. Right, it needs to be very, very flexible and be able to deal with all kinds of weird little issues along the way, and I think that’s something that every SaaS founder can take away from this. Like, if you don’t know what your customers use in terms of tools, be super flexible to be able to integrate them along the way.
33:20 – Brian (None)
Yeah, absolutely. I think it’s also about like kind of like transforming each of the different like integrations into a format that’s like native, useful.
33:28
So, for us, the end result of what we wind up storing is the summary of changes that’s very easy for us to include in doing all the RAG stuff and all that sorts of stuff, and so I think that that’s how we look at everything, should you know, kind of transforming all those data points into sort of a similar format so that it almost like and maybe it will like maybe, you know, maybe next year we’ll revisit this conversation. I’m like Arvid, I was so wrong about I was so wrong about that, but I don’t think that it necessarily matters that like, oh, you know, this conversation was in linear and this conversation was in GitHub issues. It’s like it’s all of that just turns into context around the nature of the changes and it’s timestamped as far as when it was happening and it’s like related to. You know, we’ll do like clustering and stuff like that to understand like what things are connected. But you know, I think that that will allow us to get to the point where it’s like well, again, I mentioned Slack offhand.
34:31
I don’t know if we’ll wind up doing that, but there is kind of this recurring thread of like, well, what if ChangeBot was kind of sitting in the room while these conversations are happening and I don’t even know how that would work. It’s one of those things again. It’s like I guess that’s the other fun thing with AI of I just never would think to do a feature like that. But there’s also a world that’s like okay, well, let’s figure out, let’s coerce those conversations into that same format, that same context format, and then let’s just include it and see what happens. That would probably be how we would roll that feature out.
35:04 – Arvid (None)
Yeah, it’s really interesting because then you could take out of all the conversations that happen, find the ones that may have to do with new features and suggest them to be part of this. Or maybe people say, no, this is not one. Maybe this is one. Maybe there’s a human step in between or an AI smart enough to figure this all out. Just present the results that. You’re right, you don’t really think about these things because they just would have been impossible a couple of years ago. Right, it’s bizarre.
35:30
Hey, I really like the technical side of things and I could talk about this for hours. I probably shouldn’t, which is fine, but I’m going to pivot this a little bit to the actual impact side of things, Like when you have a change log. First off, maybe, why should you have a change log? That’s something I personally often struggle with. Not to have one, but to fill it, like to actually communicate, to push information to my customers. Like, what is the benefit of this? How often should it happen? What should you talk about? Can you give me, like, can you condense your 92-page thing into maybe 92?
36:00 – Brian (None)
seconds statement here. Yeah, I think that’s fair and that’s a very common request that I get. So I understand that. Yeah, I think that, and for me, I’m coming from personal experience so that feeling of having a potential customer comparing me to other companies and not being able to put my best foot forward and showing what we’re doing and that we’re actively engaged I’ve heard the story from other founders of like because they’re not active on social or wherever somebody be like oh, I thought you guys were dead. That’s like a very common phrase and it’s like no, like. We’re actually like working 80 hours a week and you know it’s just finding this excess time.
36:40
I think people always find themselves in the spot, especially small teams, where they’re like well, you know, I can spend eight hours and build a feature, and then I could spend another four hours and write about that and do a really great job of writing about that change I just made. Or I could be halfway through another feature, and so that’s the trade-off, and so I think that’s the thing of like. And also there’s the barometric store too. If you want to make a big change with the business, it’s one of those things where you really wish you were doing this all along, and this is one of the challenges that I have with marketing the product is because I see this from a three-dimensional perspective and I look at each team. I look at from the top end of sales, giving them this resource, where it’s like, oh, you’re comparing us to this company. It’s like, oh well, we’re updating our product much more frequently and we used to compete with ChartMogul back at Parametrics, and that was also one of the formative experiences, because ChartMogul has a fantastic marketing team and they would again, in my extremely biased opinion, they would make a big deal about a feature that was not necessarily a big deal, but I was on the other side and our customers would see that and I feel like there’s just like, it’s like death by a thousand cuts, it’s like even if they see it, see it, see it, and then maybe something comes up in the future and they’re like, okay, well, maybe we’ll try out Termogul because they’re doing all these great, amazing, revolutionary things. Again, not to downplay their product, I’m insanely biased. You should not listen to anything I have to say about ChartMogul, but our original tagline was like make your competitors hate you, and so that was really a very sales and marketing-focused message For the marketing team.
38:25
It’s like finding important things to talk about. That is like you want to be kind of present always, but it’s like you can only share so many memes and I know I’m like the worst for this. I share too many memes. But it’s like even those little small updates, small improvements, just gives the marketing team something valuable to talk about. And it’s like, hey, even one of the biggest shifts is how small of an update are you willing to talk about? And I think there’s different levels. I don’t think you should do an email blast to all of your customers for changing an email form on the page, but I do think it kind of should go somewhere. Maybe changing a form is too small, but it’s like, oh, we fixed this issue. When you update your email or whatever, something like that, it’s like, hey, there’s this little frustration that somebody reported to us and we fixed. Just having Signal on that is just really valuable and it could be a little tiny social thing that goes out a little bit of Twitter juice, blue sky juice, whatever, linkedin, and so yeah, so you kind of take it all the way through the funnel. So it’s useful for the sales team, useful for marketing.
39:35
I think it’s useful from an operational perspective internally to know what’s going on here, and I think about churn prevention and kind of upsells. So at Barometrics we had a big problem of upselling because people just didn’t know of all the features that we had. I know for a fact I would have conversations with people and I was like why aren’t you using this? They’re like oh, we don’t know why our customers are canceling. I wish we had some insights around why our customers are canceling. And I’m like, oh, did you know? We have a product called Cancellation Insights. They’d be like, no, I was like, well, shit man. And so there’s an upsell thing.
40:10
But I also kind of have this anti-churn path as well, where it’s like I think you should be sending your customers more updates than you’re sending them invoices. If you ask them for money more frequently than you’re letting them know about new things that are happening, it’s like anything, just like, hey, we’re alive, because they will get the invoice. Even if you don’t send email notifications, which you should, they’ll see on their credit card statement that you charged them. And so I think there’s just a huge anti-churn, churn prevention technique of like hey, we can see that the product is improving and I think that feeds in the product as well. If you have the impression that the product is not being worked on and you’re like I don’t know what they’re doing over there, I don’t know what they’re focused on, I don’t know if they’re doing anything, people are very unlikely to reach out and say, hey, I really need you to.
41:00
I think that’s been really huge for Terminal, the hardware product that we mentioned. They’re terminal, the hardware product that we mentioned. They’re just so active and so reactive as far as fixing things and there’s a whole developer Discord community. I’ve really seen that of like. It’s accelerated because people are like oh, they do care, they do stuff, they react to questions that we have, or they react to problems and they ship. And it’s like, oh, especially for a hardware company, it’s incredible. It’s like, oh, this thing wasn’t working and now it’s working and I’ve noticed that the rate of engagement has gone up dramatically. Yeah, oh for sure. So it’s almost like what is the value of writing a change log? And I’m like well, it’s like a shorter list of what benefits it doesn’t. It’s kind of all the stuff that you want to do as a business are come from the ability to like both have that content and get it out in front of people.
41:50 – Arvid (None)
I love this. To me, the term that comes to mind is value nurturing, Like the idea that you kind of grow this seed of value perception in your customers by constantly watering it. That’s what it is right.
42:01 – Brian (None)
It’s content marketing. You know, I think content marketing happened in an age where people were very like top of funnel, like new customer focused, and I kind of in the back of my mind, I’ve had this thing of like I think this like product customer update. It’s like kind of like the new content marketing focused at retention and you know, and you know, expansion Now that everybody kind of realized like, oh, it’s not enough to get new customers, we need to keep the ones that we have like oh, it’s not enough to get new customers, we need to keep the ones that we have.
42:29
That’s kind of the wildest. Before I go to sleep at night, I put my head down on the pillow. What’s the wildest? What’s the biggest that this product could be? And it’s like, oh, it’s like the HubSpot, but instead of focusing on top of funnel, it’s focusing on from sign up through the entire life of the business. If we can 3x your LTV on every single customer, especially if it does it automatically, that becomes a very valuable product and a very valuable company.
42:57 – Arvid (None)
It does feel like a changelog, and not just a changelog, because that’s the artifact, right, but the communication strategy, the deployment, distribution method around it, that is the relationship builder. That is something that retains a relationship, obviously like a financial relationship they stay customers. But it also builds reputation for you, because when I look at the change bot change log, it’s a mouthful and it goes back to like what is it a mouthful? And it goes back to like what is it? March 2024, with the first example posted. And then you scroll up and you see all these individual changelog entries several for certain days and then it’s just so cool to see this whole development of a thing and the effort that goes into it. And that’s the reputational part, right.
43:45
You see, this is not just a tool that somebody cobbled together. This is something that grew. That grew from conversations, that grew from feedback, that grew from involvement with a customer base. Just having that as a place that people can just be astounded by what went into the tool itself, having that for my own SaaS, is something I want. It makes me want to communicate more, because every single part of this chain of communication, then, is a relic for people to kind of pull up and say oh wow, this is an old log, but they already, like three years ago, they already put in so much effort and it’s still going. That must be something in there.
44:23 – Brian (None)
Yeah, and I haven’t been able to quantify that and I think probably it would be very valuable to my marketing if I could. But there is kind of like when you go to a very old building or something, there’s just a different mode we go into as humans of like, oh, this is something that’s been worked on for a very long time and a lot of effort. So much of the effort that we put into our products just goes nowhere. Most of the people will build a feature and then not talk about it at all, which is again, it’s like the tree falling in the forest. But then it’s like we also. This is a total tangent, but something I heard just right at the right time.
45:01
There’s an artist called Tyler the creator, and he went on a tirade about how people will release an album and then tweet about it once and he did nothing but talk about the album that he created for a year straight and he’s like I’m proud of it, I worked really hard on it, and it’s like why wouldn’t I, why would you just write one message that, especially with social, it’s all algorithm. It’s the likelihood that even your followers saw it. You would get 10% of your followers. So that was very formative to me when I heard that. I think I heard that back in February or something like that.
45:35
I’m like, oh, yeah, you built this feature, you did this hard work to build this thing. And maybe it’s not like, oh, it’s a brand new version 2.0 of our application. You should size how you talk about it to the size of the change. But it’s like, oh, you fixed this bug. Again, it’s all comparison. Most companies don’t even fix bugs, so people are embarrassed to post bug fixes on their change log. It’s like no, you actually fixed the problem. Yes, you cared enough to fix this and you’re letting everybody know.
46:06
You know it’s like you cared enough to fix this and you’re letting everybody know. And especially, it’s like okay, if I look at your change log in the past six months you’ve got one bug fix. Maybe that doesn’t look so great, but yeah, if you’ve got two to three updates a week and some of them are bug fixes, some of them are new things, some of them are just tweaks. Oh, customers couldn’t find this button over here. We moved it over there and we’re always kind of prompting for customer value, like hey, we think this is really going to help you find mentions of yourself in podcasts so you can create better brand deals, whatever Kind of doing those.
46:37
Add in those little do, a little bit of the sales and a little bit of the marketing for you in those things. I think that the net effect of zooming out and seeing like oh, in a couple of months this company has posted like over a hundred. I think we’ve just crossed a hundred updates. And if it was me building a brand new startup and like writing some of the code and doing the marketing and things like that, like there definitely wouldn’t be a hundred updates in the past, like year, it would be like I would try to do it monthly and I would fail.
47:08 – Arvid (None)
It would be like one every one every six weeks. That’s why I’m at, that’s why I’m at right now. With PodScan, I sent and this is this might sound self-congratulatory but yesterday I convinced myself to actually send one out and I did. That’s great. I have this list of things that I really wanted to talk to people about, because these are important features that I built for them. Like, people ask me and I built them and I then I tell one or two individually hey, it’s now live, and then they use it and they’re happy, but everybody else is like I don’t know, right? So I put, I pushed that email out to the list and I got feedback. I actually got people hey, this is cool, and I was so happy. I was so happy I, because the features have been around for two weeks.
47:45 – Brian (None)
I just haven’t talked about it. You should let people know about this stuff.
47:47 – Arvid (None)
I’ve been trying to get myself to do it weekly, fail horribly and then do it at least monthly. I think I’ve sent out well how old is PodScan now? Like six, seven months. I sent out six, seven emails and I know I probably should have sent out 30, if not 40, with all the little things in between, because every single thing is something useful, something meaningful, something people asked for and, like you said, it could just have been. This problem that everybody had until now was fixed. Thank you, this guy, for reporting it.
48:14
And you can also shout out people in this context very easily, because the community of your customers might know each other because they’re in the same industry. So you have this opportunity to actually pull people in and give this kind of another reputational gain at this point. You’ve convinced me in these short 40-some minutes that we’ve talked about this topic, that I should really take this seriously, not just because it’s a conversion inducer or a retention retainer but those are all great things but for the product itself, it’s a meaningful part of the business to have the history. It’s kind of keeping records of your own journey, and that is to me an important part to building in public. So why do I not automatically have something gather this information for me, or at least have a platform to put this on? This is really cool. I’m glad you went into this field. It’s really cool. Yeah, thank you, yeah, and as I kind of look, this field, it’s really cool yeah thank you, yeah, and I kind of look at it.
49:09 – Brian (None)
it’s like, well, what was the point of building the feature and doing the work if you don’t talk about it Like you need to get credit for your code? It’s like you’re not like you’ve already done. I mean, maybe I’m downplaying it a little bit, but I think the victory lap, where you get all these great feedback or maybe people give you more ideas and you get credit, and people go, they’re cheering for you and it’s like that’s the part that we skip. Why are we skipping? That’s the fun part. And so yeah, that was one of our ideas too.
49:44
Of like, hey, maybe the. So yeah, that was kind of one of our ideas too. Of like, hey, maybe the engine needs tuning and maybe we’re far away from people just turning on full autopilot, of just hey, just post this and email my customers and things like that. But the bare minimum is that we can give you a rough draft and be that reminder so that when you go to sit down once a week or once a month, whenever it is that you’re going to do it you don’t have to start from a blank sheet of paper. You can actually like say, okay, cool, here’s all these things I can, just I like this line, I don’t like this line, and then you know get it out there.
50:22 – Arvid (None)
That’s. That’s the thing, right, there’s a job to think. It’s so much easier to say yes or say no, but then not even think about it. I think that’s what the problem always has been. For me is a priority problem my additional four hours of work that I could put into the product. They feel so much more urgent than the necessity to communicate with my customers feel urgent. I know that the necessity is just as urgent, right, it’s just as important for me to talk to my customers and share the things that might make them not cancel or might make them upgrade, right. So, from a builder’s perspective, the tech stuff is always burning hotter. It always feels like there’s an explosion somewhere in there.
51:05 – Brian (None)
And that’s where your skill set is. It’s like this work is hard, like it’s hard to write, you know, like being a writer is like an actual job and like you know marketing people, especially if you know the best of the best in like content marketing, it’s like there’s a true art form. It’s like there’s really like very little science to it. It’s extremely artistic and like very like you have to be very smart and it’s like a totally different side of your brain as well. So like I do that as well, if, like if I have like a task that I’m certain that I have the skills to do, and then I have a task that even if I know it’s like extremely important, but I’m not sure. You know this is not like conscious, this is subconscious, but I’m like I will just like kind of ignore or delay upon the thing that I’m like not. Or even if like even product stuff we have one. If anybody listens to this and is so kind to sign up for ChangeBot, we have the email, the screen you see like hey, we need you to confirm your email. I don’t like the way that that page is styled and it’s been on item five, six of my to-do list every single day for the past two weeks and I just realized I was thinking about it last night. I’m like this is taking so long because I haven’t designed what I want that page to look like. So I’m holding myself back from doing it because it’s uncertain.
52:13
I’m like, so what I need to do is just design what I want that page to look like and explain to people why we’re asking Like, hey, we need you to confirm your email because we’re whatever, we’re going to be spending AI credits during your trial. You know whatever, like we need to protect ourselves, or whatever the case is. Or you know, here’s like if you’re confused, book a call with me, all that sort of stuff. And so I think that if two weeks ago, I would have made my task, you know, think about what this page should have on it, articulate why you don’t like the way that it looks and the way that it’s performing, then I would have been much more likely to do it. And so it’s like that on steroids for updates. Because it’s like what do I write about? And like, oh, am I just bragging too much? Is this worth writing about and how should I phrase it? It’s like all those sorts of things. It’s just like I’ll just build another feature.
52:58
But, that’s dangerous. Then you get to the point where it’s like, oh, we’ve been working on the product for a couple of months or a couple of years and now we’re trying to get this big customer and whatever the case is, even, like you said, just getting that feedback, getting that, maybe you burn out, maybe you don’t have the motivation to keep going, unless you see people see like, hey, this is really great that you’re doing this, keep doing this. I am actually using your product. People hearing somebody to be like there are other people on the other side of this that are getting value. Please keep going. Like that’s really useful to hear.
53:30 – Arvid (None)
That’s kind of a yeah. It means to make people remember that you’re there right, that you’re not just a transactional tool, but that there’s actually a relationship in there. And, yes, I think every time you talk about this like you have to tell people that the product does this because they won’t find it out themselves. I’m always reminded of if you build it, they will come, and how. We know that that is such a folly in just in approaching anything right. Just building a thing doesn’t mean anybody knows about it.
53:58
We’re in a world of constant attention grabbing algorithms. You need to put yourself out there. Just social media is the exact same thing. If you want to get people interested in your business, you have to be present where they are and you have to talk about your business. A change lock is that. But for your customers, for your users, for your prospective users, for your trial users I’m thinking about this just now. If I don’t have a cadence of at least 10 days, then some of my trial users will never get a product update from me during their trial, and that sucks. Why wouldn’t I send them one every three days so they see oh, wow, I signed up to a thing that’s still getting new features. I need to stick with this. That’s just self-sabotage at this point, totally.
54:41 – Brian (None)
Yeah, constantly getting better and yeah, I think that one of the things that we figured out very early of people that we really wanted to serve are just people in SaaS. It makes sense to have the recurring benefit, but also there’s a subsection for people that we think are really good for our product, of people who are shipping, because we have had a couple people come to us already and be like hey, we want to fill our changelog and the reason why it wasn’t filled is because they just weren’t doing anything, they weren’t building stuff, and so I do feel like maybe there’s something down the road that we could do for those people. It just doesn’t really. It’s just a different problem. It’s like a repurposing thing. That’s a different stage too right.
55:21 – Arvid (None)
First you fill it up now and then you do the backfill later. Yeah, yeah.
55:24 – Brian (None)
Well, yeah, I also just feel like we just I just want to stay like super focused. It’s like, hey, especially like in your case, like you’re doing the work. It’s like it’s not like you need some magical system. You know there’s not like a some marketing framework that has some like initialism that you know you have to follow the whatever, the tower framework or whatever. It’s like you’re doing the work, you just have to talk about it and it’s like that’s the type of company that we really want to support, because there’s a lot of people out there that are just really grinding on the product and you wouldn’t know it. It’s not apparent that it’s like, oh, this person is spending legitimately 60-hour weeks, and they have been for two years. You know their product is good and you know it’s interesting, but it’s like you don’t really get a sense of scale, and so I’m like those are the people that I want to help, because it’s like they’re doing all the hard work.
56:15 – Arvid (None)
I bet you’ve convinced a lot of these people to look into ChangeLogs in general, if they haven’t already, and to look into ChangeBot right now. So if people want to follow both ChangeBot and Terminal we didn’t even get to talk much about this and you and all the other things that you’re going to be building over the next year where do you want them to go?
56:34 – Brian (None)
It says 400 page views are simple analytics for ChangeBot, so when this drops, this needs to be double that number. It’s going to look something like this, I think, when the episode drops. But yeah, terminal is amazing Use Terminalcom. It’s been actually totally like. I feel like this is the sales pitch, but it’s actually true of the idea of being heads down in maker mode. It’s something like that.
57:00
If I didn’t have a little dashboard to show me what our web traffic looked like, I’d be going to Simple Analytics. Sorry guys, I love you guys. I’d be going to Simple Analytics. Sorry guys, I love you guys. I’d be going to their website every day and then I’d be clicking on stuff and I’d be rabbit-holing and so having it in front of me always and if there’s an issue or when there was a spike, that’s an indication to me that I should go take a look. And then most of the time it’s just little SEO pops here and there.
57:24
Just having all those notifications and all those dashboards pushed into this form factor, this hardware form factor, has been actually really useful because there’s so much to think about and so much to work on. That it’s really great, and I’m a part of the developer community as well and seeing the stuff that people are building. One person has an international space station tracker so you can know when the space station’s over. It’s really cool. One guy actually space station tracker so you can know when the space station’s over. It’s really cool. It’s totally open source from a firmware perspective. So one guy hacked it so that it’s live updating with his playlist and so he’s got it on his speakers. So anyway, it’s really cool. Yeah, I hope it is, because mine is still in the mail.
58:00 – Arvid (None)
In Canada there’s been a postal strike over the last couple of weeks, so I have not received anything. Oh, that’s so funny.
58:08 – Brian (None)
Yeah, last couple of weeks so I have not received anything. Oh, that’s so funny. Yeah, I saw that we’ve had a couple of people complaining about slow delivery and I just saw because my wife does e-commerce stuff, you know here and there and so we got a notification that it was something like USPS was like, yeah, deliveries are going to be a little bit slow Because, like Canada’s not accepting any of our mail and like that’s had carryover, like it’s like, oh, we have like an entire country’s worth of mail, like that we have to hold on to and we don’t know what to do with it. So I think it’s yeah, hopefully that gets worked out.
58:33 – Arvid (None)
Well, they stopped. I think they were sent back to work, and there’s a whole political angle to this. I don’t definitely don’t want to dive into this, but I that you just don’t have to think about. With a SaaS, you don’t have to deal with postal strikes or anything like it. That’s cool. I’m looking forward to having mine and I got the developer edition as well, so I’m looking forward to getting my pod scan monitoring into that system as well, just to see how my system is performing.
58:59 – Brian (None)
Yeah, that’d be really cool to have a little. We have like Hacker News and Reddit and things like that. That’ll kind of show you those headlines so you can feel like you’re informed, but you don’t like click and go through, so that might be cool for you to build a similar one. It’s like, hey, like here’s like your most recent mentions, and so you don’t have to think to go and you just kind of look and if you see one that’s like, oh, it’s, we were mentioned on whatever CNN, it’s like, okay, maybe look into that um, that’s, that’s.
59:22 – Arvid (None)
That is one of the the customer facing. Definitely I’m excited to hack on that. And then for my own monitoring too, right Like what’s the system doing? What are my servers doing? Are the GPUs burning? That kind of stuff. There’s so much monitoring we could do and so much updates that I need to know for my own business. No, I’m excited for Terminal. I think it’s crazy to think that you have to work both on something that is as like AI-centric and SaaS-centric as ChangePod is, and then you have something that is so almost like deep tech, even though just hardware, but you know like it’s just. You have to deal with firmware we have to deal with like display logic and updating screens, getting on people’s Wi-Fi. It’s the universal issue.
01:00:09 – Brian (None)
We really tried to build this in a super secure way where most IoT things are like the company can drive and push updates to but terminals they request and so we can’t proactively push something. The terminal has to ask for packets, and so that’s caused us some pain points. But I feel like so many IoT devices are just vulnerabilities waiting to happen. It’s like, hey, let’s put this not sufficiently secured device onto your network that the public internet can access.
01:00:42 – Arvid (None)
It’s just an insane thing to do. That’s a bad idea.
01:00:44 – Brian (None)
Yeah, it’s a really bad idea and it’s like 99% of those devices. So you know, we’ve also had to kind of work through that. It’s like, hey, because of our choices, like we can’t like, it’s like a little bit harder to troubleshoot because we can’t just like push, we’re like we can’t see what did we send? We have to go back and, like you know, reverse engineer from that perspective.
01:01:04 – Arvid (None)
Funny how that, like privacy and security concerns, that’s something you take into every single business, like the moment you have to do with the internet, the deal with the internet, you have to just make sure that what you do, or what your devices do, what your software does, is beneficial and not harmful to your customer. I mean go without saying, but it’s still. It’s a concern that may be overlooked in the zeal and eagerness that some people have for building businesses right Totally.
01:01:33 – Brian (None)
Yeah, you can easily build a Trojan horse when you’re like you’re literally delivering something physical into somebody’s house. So it’s like only so many people are going to like eventually we’re going to probably snap back away from this, but people getting hacked from their refrigerator and stuff like that I think we’ve only got so many of those left where we’re like okay, we really need to rethink how we’re handling it.
01:01:52 – Arvid (None)
Yeah, it’s weird that my light bulb can be an attack vector into my home.
01:01:56 – Brian (None)
Yeah, totally, it’s bizarre, exactly.
01:01:59 – Arvid (None)
Well, let’s drop a couple URLs and social feeds here so that people can follow you on your journey. We have Terminal, we have ChangeBot, we have you yourself. Just throw a couple URLs out there that people can mentally click right now.
01:02:12 – Brian (None)
Useterminalcom for Terminal. You can buy them now and, depending on when you order it and depending on the state of the manufacturing in the world, we will get it. We’ve got an update on the website so we let you know when it’s shipping. So we still have units in stock right now, but if there’s any delays, that’ll update. So we’re really transparent about I think we’re one of the few Kickstarters who actually delivered a product and continue to do so, so we try to be really good about that, and global shipping strikes notwithstanding. So that’s Terminal.
01:02:45
Changebotai is Changebot, and you can go up there and you can sign up today and we have the whole self-service flow where you can book a call with me. And then I’m on Twitter, mostly BCRakowski, which will be linked. You don’t have to worry about spelling that and I’ve been trying out BlueSky as well. I’ve been kind of dipping my toe in the water. I feel very much like a noob over there and in fact, I don’t even know. I think I’m B Sirikowski, maybe I’m Brian Sirikowski. I don’t actually know who I am, but we’ll find the link.
01:03:13 – Arvid (None)
We’ll find it out. That’s how new I am.
01:03:19 – Brian (None)
So, yeah, I understand that I’ve been given the advice that I should be begging to be on starter packs, and so request if you have a SaaS dude, saas bro starter list, please include me and then yeah, we don’t really do anything. That’s ChangeBot native. Right now, I think. We have a Twitter account, we have a BlueSky account, but mostly we’re just using that to reshare while we’re in the founder-led stage. It’s like you want to talk to us and R is on there as well, so that’s one way. If you want to follow us on either of those platforms, you can follow ChangeBot and then you’ll get all the tweets and blue skies from myself and Ara as well.
01:03:55 – Arvid (None)
All the updates, all the updates, literally and figuratively.
01:03:57
Literally all the updates. Well, I really appreciate you sharing this. I’m so excited. I’m excited that we had this conversation today, obviously learning all about these things. But I’m excited that we had this conversation today, obviously learning all about these things, but also that what you wanted to do back in the day actually came to fruition over this year, plus then Terminal and all the other things that happened besides that. That is so really cool to see.
01:04:17
I think, founder-wise, you did pretty well. That’s really awesome. I’ll give myself a B plus. That’s awesome, man. I really appreciate it. Thank you so much for sharing all of this with me today. It really made me think differently about how important change logs are, so I appreciate that, and I can’t wait for next year when we have another chat and see where it’s gone. That’s going to be awesome. Thanks so much, brian, really appreciate it. And that’s it for today. Thank you for listening to the Boots of Founder. You can find me on Twitter at arvidkahl A-R-V-I-D-K-A-H-L. You can find my books on my Twitter course site too, and if you want to support me on this show, please tell everyone you know about PodScanfm and leave a rating and a review for this podcast by going to ratethispodcastcom. Slash founder it people’s feeds and that will truly help the show. Thank you so much for listening. Have a wonderful day and bye-bye.
