363: Ben Rometsch — From Side Projects to Industry Giants
Download MP3Hey. It's Arvid, and you're listening to the Bootstrap Founder. Today, I'm talking to Ben Bromwich, the founder of Flaksmith, a Bootstrap SaaS business in the feature flag space. From open source software to on premise deployments, we will dive into building a single purpose tool for a wide variety of customers and how to scale a bootstrap business along the way. This episode is sponsored by paddle.com, also a tool that will help you scale your bootstrap business because they'll handle all things payment related.
Arvid:So you can focus on growing your own product and focus all your attention there, where it belongs. I use Paddle for all my projects and you really should also check them out. Now let's talk feature flags. Here's Ben. You have Bootstrap Flaxsmith, which is really cool.
Arvid:It's a software built around feature flag management, which is very interesting, and controlled software releases, which is something that I have no control over in my own businesses. Just saying. But tell me more about that. What is FlackSmith?
Ben:Yeah. So, FlackSmith well, it's it's a number of things, as a sort of entity, it's a commercial open source, sort of project and business. It's a software tool and technique that allows, products and engineering teams to, yep, control who sees features when they see them, how how features behave, and basically allow so the the the principle concept is it allows you to decouple the action of deploying your software with the event of releasing features. It decouple those two things. And normally and up until, you know, 10 years ago, there was you know, they were they were intrinsically linked.
Ben:When you wanted to release a feature, you did a build. Right? Like, that's kind of, like, what you did. And then, yeah, I don't know how long it's getting on now. Maybe 12 years ago, there's a couple of papers, 1 by Martin Fowler, a blog even, like, one by, the photo sharing site Flickr, and they were like, this is how we're releasing features.
Ben:And I remember the Flickr one, everyone was like, wow. That's crazy. Like, that sounds really dangerous and prone to problems, all that sort of stuff. And over time, the segment has has really matured. There's, like, a bunch of providers out there of all different shapes and sizes, which we're 1 in, you know, our quadrant.
Ben:And, like, as an engineering concept, it's become I wouldn't say it's like I don't know. Like, this is a question we often wonder is, like, what percentage of engineering teams in the world are using feature flags. Right? And I actually we think it's quite low. It's still not, you know, the the accepted wisdom of how you deliver you know, develop and deploy modern software, but it's definitely getting there.
Ben:And, a lot a lot of a lot of the business and a lot of the the open source projects is about, like, discussing those engineering techniques and educating people about what the pros and cons are because it it's yeah. Engineers always love to talk about the Engineers always love to talk about the technology and the sort of the software, but it's actually much more about the engineering and the process, the people, and the kind of stuff that that doesn't have ones and naughts associated to it. Yeah.
Arvid:Yeah. Maybe that's the problem. Right? Maybe it's the fact that it's that it's not just code. Right?
Arvid:It it takes a a slightly different mindset for an engineer to adopt the idea that not everything has to do with, like, entering a git push command or something, right, to deploy. It's interesting to see because, I I've seen in my forays into software engineering, which I've been doing for, you know, a couple decades now, that with the frameworks in particular, and I use Laravel in the PHP world right now, it is becoming almost a first class citizen now that that these things are now part of the the big packages or at least they think about it. Right? The it's a little little bit of it.
Ben:Yeah. Well yeah. Whereas you think, like, 10 years ago or 5 years ago, that would have been unthinkable. Actually, there is a there is a plug I wanna give. So, there's a project called Open Feature, which is a CNCF project.
Ben:So I'm on the governance board of that. Is kind of progressing through the CNCF sausage factory at the moment, and, it's trying to define and implement an open standard for Feature Flags. And the reason I mentioned it, well, I should mention it anyway because I'm on the board for it, and, we we we believe in it. But the where where, you know, where you go get a framework like Laravel or or Rails or Django or whatever, there's there's discussions with those some of those folk around putting open feature stuff into Laravel. And so the idea would be that you could create a flagsmith account or an account with another another provider and define your flags, and then you wouldn't have to do any dependency stuff in Laravel or whatever.
Ben:You could just say, right. I wanna use FlagSmith as my provider. Here's my API key, and you don't have to you don't have to muck around with anything. So that that's definitely like as a as a sort of a signal of where the the kind of the concept is in terms of its maturity within the industry. That's where it's at.
Ben:And so, yeah, like, that's been something that we've been working on since its inception. It was an idea brought up by Dyna Trace who I have to give a shout out. They've they've been really, really good, you know, providing engineering time, resource, and and pushing the project and is now going, yeah, going through the CNCF process. It's definitely something that I think most or if not all the providers are kind of, like, on board with. For those that are aware of OpenTelemetry, it's very analogous to OpenTelemetry, but feature flagging.
Ben:Yeah.
Arvid:That's really cool. I love this. Like, I I was already amazed that it was a commercial open source project, but now also seeing you push for open standards. Like, that is awesome. I I work in in podcasting with my own thing.
Arvid:Right? I I aggregate and transcribe podcasts, and there is a lot of open standard work there too, which even just makes it possible to work with a whole ecosystem. So it is it is really cool to see you push for that and still include everybody else. I think that that is just such a inclusive quite inclusive way of approaching the thing.
Ben:I think that open standards have kind of they I remember, like, 15 years ago, there was maybe 20 years ago, it was much more common for their they're kind of a little bit of a curiosity now, you know? I think especially with, like, JSON and stuff, like, all these kind of schemaless or kind of, like, you know, sort of in a way schemaless things, which I I mean, I love I love you know, I I've come from I'm old enough to remember Korber and, you know, like, XML deep XML stuff, and, you know, all these, like, crazy, you know, Java b enterprise Java bs, all that sort of stuff. So I definitely believe that that that's the wrong way to go, and Jason's kind of, like, broken this free of that, and rest is broken this free of that. But the the concept of there being, like, yeah, like common standards is kind of it's it's kind of it feels like it's more and more unusual to come across. Yeah.
Arvid:I mean, barely, any big player has a any vested interest in in these standards. So it may be for interoperability, maybe for being able to have some kind of API integration, but that's where it ends. Like, their internal data representation is their own thing. Their API thing is their own thing. If you look at the Stripe API as well as it is designed, hey.
Arvid:Everything is kinda custom. Everything is very arbitrary and and opinionated. So, yeah, it's just rest. Right? It's it's just JSON data, but every everything else about the data, the connections between stuff, that is their own design, which is fine.
Arvid:But it's not very open. Right? It's well documented, but not open.
Ben:Yeah. I mean, could you imagine, like, Stripe and Adyen and a bunch of others getting together?
Arvid:Wouldn't that be awesome?
Ben:Maybe set green like a common, it's not gonna happen, is it?
Arvid:Yeah. No. It probably isn't, which, but if it were, it would make it so much easier to build integrations for these things. And, I mean, that that is probably yeah. If you want vendor lock in as a big business, like, you well, you know, it doesn't really make much sense to to offer this or even make it feasible.
Arvid:It's, it's still great that you chose the the open source approach here. And I I always feel when I see businesses built on open source that the founder or the founding team, they must be very masochistic about this. Like, they must really love pain because whoever I talk to, there's always dealing with people in the open source world. And I I don't mean, like, people who are really engaged and really enthusiastic. They're great.
Arvid:But dealing with people who are needy and entitled, that seems to be a common threat. Have you encountered this as well?
Ben:To be honest, not that much. I mean, a bit, but in the grand scheme of things like and we definitely get more we definitely get way more, you know, thank yous or how can I contribute or whatever rather than, like, you know, I can mind you? I mean, we do get some, like, you know, and it's it's actually it's it's not so much on the on the kind of GitHub side of the fence. It's much more on the, you know, I don't know, our chat support on our website. People are like, right.
Ben:I've got this thing running. It's the open source version, and I need you to basically spend an hour helping me configure it and getting, right, these these error messages fixed up on me. Like, you know, we'll help you a little bit, but, like, I'm not gonna spend, like, you know, more than 5 minutes doing this because, you know, you're not paying for my time. That happens occasionally, but not that often. And I mean, even then, like, you know, if you were if you were squinting, you could see, oh, you know, maybe, you know, they got they they sort of, like, bumped their head on this bit of the platform because we don't have the documentation for that.
Ben:And so then you go and you go, yeah. Fair enough. That's not clear in the documentation or it's not even mentioned or whatever. You know? Like, right.
Ben:Yeah. If you're using Laravel and it's on blah, then you might have this blah and or, you know, or fix it. So I always kind of think, oh, okay. You know? Because one of one of the huge sort of prizes that we that we that we really, really try and achieve is you should be able to get the platform up and running really, really quickly and easily and in a production state that can deal with production workloads, and you should be able to integrate it into your into your application without, you know, bumping your head or having to to message us.
Ben:And a lot of the architecture and infrastructure design around the platform speaks to that. So we we have, like, one data store, which is Postgres, and we have 1, monolith application, API, front end, or everything, and that's it. Like, there's nothing there's nothing else. There's no Redis. There's no Elasticsearch.
Ben:There's no yada yada yada, like
Arvid:No Kubernetes? Like, no no nice complicated deployment system?
Ben:We we've I mean, we've got Helm charts, but it's literally start Postgres, run this API, and that's it. So, yeah, that that's been a that's been a, I mean, I I you know, it depends on the project, but you see, you know, you see some projects where I mean, like, Century is the classic example. It's like, you know, 27 services, which is fine. Like, they probably need 27 services. We don't.
Ben:There's been a number of times where, you know, we've been at that crossroads, and it's like, do we use Redis, or do we use do you know? Let's let's introduce x service, and then we've gone like the task processor was a really good one. Like, it would really been a lot easier with Redis as as the as the broker, and we were like, no. You know, our customers, the last thing they want is another data store in their stack. Like, that's like the number one thing that they hate, and so we've always we've always steered the boat in that direction.
Ben:Yeah.
Arvid:I guess Postgres is a good choice for all of this. Right?
Ben:Yeah. I mean yeah. I actually saw a good I I actually saw a good, a good post the other day on, like, all the things you can do with it. Yeah. To have
Arvid:vector storage, like g GIS. Right?
Ben:Yeah. I didn't even know. It's got like a cron scheduler. You can use it as, you know, like it's crazy. Yeah.
Arvid:Yeah. It's it's just a it's funny because those are the tools they tell us not to build. Right? The the the everything tools. But then again, for a database, it it just feels like you just need to make it flexible enough at the core and give it all these little extra things, and all of a sudden, it becomes your schedule.
Arvid:It becomes your, your key value store and your transactional database, which is nice.
Ben:Yeah. We're definitely we're we're really happy with that that we feel that that that decision's really, really paid off. The other really interesting thing about that is, like, once you've introduced a new service, you're not you're not taking that out of your stack. Right? Like, that's just not gonna happen, like, realistically.
Arvid:Yeah. And you you add more things to use the service to make it useful. Right? That that becomes a psychological game way then, oh, we could totally put this new feature out there. I love the simplicity of this.
Arvid:And I I do wonder because, like, in in most software businesses, like, when people start a business and they try to find their first customers, usually, it's a SaaS approach. Right? They would they build it on their their own system. And and then they get into on premise, and then they get into, you know, like, private cloud or whatever other ways of deploying, these things is is feasible or makes sense to them. And then they they start to build these packages that they can easily deliver and easily deploy without causing too much fuzz.
Arvid:What was your story there? Did you start out with all 3? Because you have all 3.
Ben:No. No. No. No. So we we, well, we originally started the projects.
Ben:It it was just an open source thing. Right? We we I've been running a software agency in London for, like, 18 years at this point, and we always used to come up with different crazy things to do in our spare time, and none of none of them could have been legitimate businesses by their own right. There were always crazy ideas. Like, my favorite one was, we had an Instagram clone that where they called life is terrible, and you have to take photos of it when your life had gone wrong.
Ben:And the filled the filters made the photos look worse. Okay. Yeah.
Arvid:That and and I don't wanna be insulting, but that sounds like a very British thing to do. Yeah.
Ben:Yeah. Is. Yeah. Oh, funny. Yeah.
Ben:No one in America would have used that.
Arvid:No. Probably not. That's funny.
Ben:So so we we'd always come up with stupid ideas, and then and then we decided to be more rigorous around, what sort of things we built in our spare time. And, so we we wanted to use feature flagging for our clients, for the agency. And at the time, they just we went to GitHub to look for because we'd seen that LaunchDarkly had had sort of was was getting on its ascendancy. That's the kind of big player in the space, and, we didn't wanna kind of pay. It was quite expensive.
Ben:We felt it was quite expensive. People think that that it's trivial, serving flags, and it is it is quite simple. You know, it's not very transactional, etcetera, etcetera, but it isn't it isn't trivial, especially when you're serving them at scale. And so we we were like, okay. Look.
Ben:What what's on GitHub? There wasn't really anything on GitHub. There was a couple of, like, little projects that were like they were either language specific or framework specific. And we were like, look. This is obvious.
Ben:Like, there there was, you know, at at the time, there was, like it was very common that you get, you know, a commercial venture backed company that would that would push the space or or create the the segment, and then you'd see, like, open source companies or, commercial open source projects, sort of, like, filling in that gap. And there wasn't one feature flag, and so we were like, hey. That's a nice opportunity. We're really the perfect people to build it because we'll be our own customers. We know exactly how we want the API to feel, what the SDKs would look like, and all this sort of stuff.
Ben:And, also, it's a great project that you can build an MVP. I think we had an MVP out in, like, a week where like an internal one where it's like, you know, I had an admin in his face, and you can add you can add flags, and you could toggle flag values. And it had a I think the first I think the JavaScript SDK was the first one that we built, and, you know, that could connect to the API and and do all that. So it was a great, it was a great project to start with, but we only did it we really did it for ourselves, and we put it on GitHub, after we spent ages worrying that there was gonna be a API key in this source code history that we'd forgotten about, but, I think there's good tools for that now. And then at first, like, I I we just sort of thought we didn't really, well, at all, analyze or research the market.
Ben:So my experience was that, you know, you do that, you do a SaaS platform, away you go, you know, and then step 4 is profit. And we did it we did that. We built we built out, you know, we we started hosting it properly, you know, in a fault tolerant manner on on AWS, and we put a website together. And at the time, it was called Bullet Train, which is a terrible never name your projects. If if someone Googles you, you shouldn't be competing with things like actual trains.
Ben:So definitely, that's that's something that we we discovered. Yeah. So it was called Bullet Train. I kinda like the name, but for commercial business, it's not not a good idea. And we got a bit of, we didn't we didn't market it really.
Ben:We weren't like, right. You know, we need to get 5,000 GitHub stars in a month or a year or anything like that. We were just kind of doing it for ourselves. And yet the SaaS business wasn't picking up at all. We didn't really care, because we were running the agency.
Ben:We were working on it in our spare time. That's one of the big learnings for us was having a a kinda well, depends where you are in the world, consulting business, software agency, getting paid for your time. When when you're an agency and you don't have any paid work, that's kind of an existential crisis situation because you're like, that person is burning up money because they're not billing. But, actually, what what the big discovery for me was that that's an opportunity that is free for you as a as a as a business because that person's salary has already been paid. They're really you know, they're an engineer.
Ben:You know everything about them. You know what they're good at. You know their their strengths and weaknesses. You know what they can work on and what they can't. You're in an office with you know, like, that's an amazing, amazingly valuable opportunity.
Ben:If you've got, like you know, so we would be like, oh god. You know? Like, we've got one front end developer and one back end developer and one designer not not billing. And now I'd be like, that's that's absolutely brilliant because that's the team. They they can build go and build something.
Ben:Right? And they're they're it's free because you've already sunk that cost. I mean, the rest of the money the rest of the agency needs to be making money. Otherwise, you go bust, but, that's gonna happen anyway. Right?
Ben:Like, whether you've built a side project or not, you're still gonna go bust. So that was a hugely valuable thing. So we kind of didn't really care too much. Probably we probably should have cared a bit more, and things would have started or would have would have grown a little bit, faster, but, you know, we weren't you know, I didn't really have regrets about that. But then what happened was, yeah, what what you're saying about people running on prem, we started getting contacted from massive companies saying, you know, I really like this platform.
Ben:I and that's one of the other benefits of open source. Right? They can just well, now, like, you know, Docker pull, Docker compose up, get it running on their on their local machine, play around with it, and say, right. Okay. So I've got this running on my machine.
Ben:I could see how this is now gonna be running. I I can see the path to that running in our data center and us using that because I'm not allowed I'll I'll get sacked if I use the SaaS product. That's another interesting unknown percentage, which is, you know, what percentage of cost companies or large companies out there have that, like, predication of you can't put this sort of data outside of our control. Yeah. That's a different discussion.
Arvid:That's probably quite sizable. Right? There's probably a lot of these businesses that have this.
Ben:I think it's more than people I think it's more than engineers think, is is my answer. I think engineers would underestimate what that number is, but it also depends on what the type of product is as well. Like, we're potentially dealing with data like per you know, you can send, what we call trait data about your users or or or the or the the thing that's running the application and then control what flags are flag values are received based on those traits. So you can say, you know, if they're on iPhone, then turn this feature on, or if they're in Canada, then they should not be able to, you know, I don't know, buy guns or whatever.
Arvid:Oh, yeah. Okay. That makes sense.
Ben:Yeah. So so those those companies then wanna be able to run all that stuff on premise, and they started asking us, how do we you know, how much is it? And that was when we were we were really in trouble because we that at that point, I was like, I have absolutely no idea. Like, how much is how much do I charge for this? And at the time, we were charging remember, we we got the first, like, inbound lead, and it was kind of like they were really, like, sort of almost aggressively wanting to buy the software.
Ben:Like, we really need it. How much is it? And at the time, I think our most expensive SaaS plan was like a £100 a a month. And so I was I was sort of I I kind of got anxious and I was just like, I just shot out a number and I said, oh, it'd be £400 a month. And, so those listing who don't have have the same amount of experience as I did at the time about enterprise sales, no one pays £400 a month for anything.
Ben:Right? Like, it's not they would be like, these guys aren't serious. They don't know what they're doing, which was true. That was an accurate, analysis of the situation from my point of view. And so yeah.
Ben:So, we started to get more of an uptick, and at that point, we realized that we needed to get some other people to join the team who had experience of speaking to those people. You know, at the time, I didn't even know, it's called go to market. Right? I didn't even know what that I was like, what is that? Enterprise sales, go to market motion, you know, get all of your pricing organized, all of your, you know, your legal support, all of that sort of stuff, yet it's kinda dull.
Ben:If you're an engineer, it's, you know, it's not it's not writing code. It's not pushing commits or anything like that. But but in our case, that was the business pretty much. Like, it's it's like maybe I think now it's not all the business. I think I think our SaaS, non SaaS revenue is about 25%, 75%.
Ben:So, you know, it's it's a lot. And they're they're they're, you know, they're servicing those 2 types of customer are very different. And, you know, the the the economics of of of serving those 2 types of customer is very different, but the business wouldn't be and the and the project wouldn't be where it is today if we hadn't made a a sort of like a a an admission to ourselves that, like, we we don't know what we're doing at this point with regards to this aspect of the business. We don't know how big it is either, and it turns out that, you know, there's there's a huge addressable market for people who wanna run on premise feature flags in our in our case. So, yeah, so we went out and we found not a venture capital company.
Ben:We didn't do that, but we did find 3 guys who had formed a business recently before we did that deal called Polychrome Capital, and they came on, and we kinda did us an equity swap. So they they worked on the business to get it, like, 20 times 20 times the MRO that it was at the time. And and for that, we gave them half half the organization, half the business, which, you know, I I remember I remember telling my brother, I was thinking about doing this deal, and he got really angry. He was like, what the hell are you doing? Like, why would you give him half the business?
Ben:And I was like, look. I you know, this could be the enterprise part of the company, could be the majority of it. Right? And without those those folk, we're not gonna you know, we we we could be selling stuff 8 times cheaper than, you know, than than than we could be. And so I've I've said before to other people, it's like easy the best deal I've made, in my career in terms of making that admission and and realizing the moment where, you know, yeah, you know, like, if you love something, set it free kind of thing.
Ben:It was very much that moment of, like, this is this is time for me to right? And and the team to say to ourselves for for it to get to the next stage and and to grow to the next point, and we need, other people involved and and other areas of expertise that weren't just engineers or or, you know, hackers basically yet.
Arvid:Mhmm. A lot of founders have this fear of of that moment because they know they have to give something up. Right? Not just, like, that they have to give up being the technical founder and knowing everything about the business, which is probably more what it is than the technical part. But giving up agency, giving up, like, the that just being able to do whatever you want on a whim with your business.
Arvid:Do you sometimes regret this? Because you took that choice and you made it and you you got out better on the other side. Is there still regret?
Ben:Well, the the the interesting thing is that that didn't actually happen. I didn't I don't feel like I I feel like in any if anything, I was sort of freed up more. And the reason for that is that, first of all, I didn't have to deal with with selling all this doing all this stuff that I wasn't qualified to do. They did all that. Right?
Ben:Like or they knew people who could come on board and help, like, you know, I don't know, accountants or lawyers or people like that, people who could build, you know, CRM systems and get all that up and running and build that engine. I didn't have to do any of that anymore, right, which was great. And then the other thing is it wasn't a VC thing. So it wasn't like, right. You know, we're we're basically gonna give give you a bunch of money, but you're basically gonna give us control of the board and our decision on strategic direction is final and, you know, blah blah blah blah blah.
Ben:We added we that's I was part of the idea that was like they they have final say. So Polychrome have final say on matters of commercial, you know, I don't know, pricing or if we say no to a customer or or things of that nature, and then technical, stuff and and and product, we were gonna maintain control of that because, you know, like like, I you know, like, decisions around what was put in the open source repositories and what was proprietary, all of that stuff. So, I mean, you know, in a way, pragmatically alright. You know, not legally, but in terms of, like, pragmatically my and and and the the the originating team's sort of day to day lives, we were freed up way more, on account of that. I'm I'm conscious of there's there's an analogy around Brexit.
Ben:So for those of you I mean, I'm still very angry about it. But, you know, there are a lot a lot of people in you know, when when when the UK voted to leave the European Union, a lot of the argument was that, you know, we'd gain this sovereignty. Right? But that's just a that's just an abstract you know, that concept of freedom is an abstract thing. And, you know, legally, the UK or, you know, us might have less control, you know, seeding that with other people.
Ben:But, actually, they state, you know, pragmatically what we can and can't do. We, we have much more freedom when we did that. And that's definitely something that I kind of wasn't really you know, I just saw I just saw it as, you know, there's an opportunity that it can get to the next stage of of its of its growth. And quite I've know I've noticed this with other businesses that I've worked on as well. When when you're scaling a business, you'll quite often hear a kind of a ceiling, and then sometimes you know?
Ben:And and some you know, that will happen again and again and again, and sometimes it will be obvious what that ceiling is or what's causing that and what you need to do to then break through it, and sometimes it won't. The deal that we did with Polychrom is very, very, very clear what that ceiling was, and it was very, very easy to go, right. Well, you know, we just need to give half the business away to sensible, smart, well intentioned people. And and, you know, luck well, luckily, that's what happened. I mean, it might not have been sensible, smart, or well intentioned, but, I mean, you know, that's then you gotta check that they are before you do that.
Ben:But, yeah, like, it was definitely, yeah, it was definitely the the the single most important decision in the life of the company. And, yeah, it's it's it's been really interesting watching that grow from that point. Yeah. Because at the at the time, I was the only one when we did that deal, I was the only one working on the on the project full time.
Arvid:Oh, wow. So how long into the journey of the business did you go all just by yourself? Like, when did this deal happen in the history of the the business?
Ben:So, yes, the first commit was, I think, around 2017. And, I started working on the project full time just before COVID. And then we did that deal, what like, it was annoying because we couldn't meet we couldn't meet each other. So we did the whole deal. 1 one one of them was in in Cologne in Germany, and the other 2 guys were in America on on the West Coast at the time.
Ben:And, whatever he did, you know, none of us knew how tall we all we all were. So we concluded that, like, a little bit through, yeah, when when you couldn't fly anywhere, basically, whenever that was in in 2020. So, again, like, I mean, that was several years. Right? Like, the business now in terms of, right, the code base is 6 or 7 years old.
Ben:So it's been a lot of, a lot of work that that's, you know, it just carried on. We haven't really, yeah, the the the, you know, the the growth curve has never it's never hit that sort of, like, asymptotic thing, but it's it's definitely like it's gone through different levels of gradient, and there's, you know, there's a for for when when, we have new people onboarding onto the company, I or when we have a we have an annual off-site, and, you know, I show that graph, and you it's because it's kind of a mad graph because you can, you know, ask people at what point did BodyCream come on board, and they it's they're, you know, painfully easy to point that out on the graph because, of the effect it had on the business. Yeah.
Arvid:Good choice then. I think that's, that's that's a miraculous choice and and a good one.
Ben:Yeah. I mean, we we were very fortunate. It was very fortuitous to meet them because, their model is extremely unusual. And I would I've been having calls with VC firms for months, and I'd actually sort of given up, and and I was like, there's no point doing these anymore because we were too small. You know, this everyone there there was a lot of activity around b to b SaaS.
Ben:Developer tooling as well was super hot. Like, the GitLab story and things like that, that was, like, super hot at the time. So I was getting inbound from VCs all day. I was taking calls from some of them that that you know, the ones that I'd the ones that I'd heard of or the ones that sounded like they had an interesting take, but, you know, I was like, this is just isn't gonna work for us. Like, it's not the right it's not the right path for us.
Ben:Yeah. So I don't quite understand why the VC industry is quite so homogeneous in its approach. I mean, I guess I do because, you know, they if they're you know, they hit a couple of home runs, they get fantastic rich, but there's very, very little variety in in that space. And for us, like, the perfect thing wasn't like I think the worst thing that you could have done to us at the time was like, here's $4,000,000. Spend it as quickly as you can.
Arvid:Do whatever you want, but you still don't know what to do and how to do it, but here's money. That is a problem. Yeah. I see I see this a lot too. It's like people all following the same rule book, the same playbook, yet nobody actually wants to do the work.
Arvid:Right? Nobody wants to go into the business and actually shuffle things around for it to to be better or teach the founders how to do it right. I I suffer from the same problem. Like, I'm not a salesperson, not a big marketer. I can do it kinda mostly, but I'm also completely out of my depth when it comes to enterprise sales.
Arvid:Right? For that, like, I've just never done it. And I'm trying to learn it now with my business at this point, but it's it's a it's a tough journey because if if first off, not a first language English speaker, that's always a problem, then I I just don't know how to talk to people in that enterprise setting because I'm not from there. So there's a lot of different things that can stack on top of each other to make this harder. So good for you to have phoned the right people at the right time.
Arvid:And what I find very interesting because you your journey is one of, like, absolute bootstrapping, yet you were open to these calls. You were open to talking to people about this and giving up 50% of your company to just improve it. That, I think, is such an interesting mindset in a world where everybody is black or white, either this, either that. Right?
Ben:Yeah. I mean, that's that's I mean, may maybe it's just because it was COVID, and I just wanted to have someone else to talk to. But, yeah, I think the learning I've got is that, yeah, keep yourself open to things. And then when the right opportunity comes along, then you have to grasp it really, really hard. Right?
Ben:And then we did make a bet on the business at that point. But I I I kind of also I I sort of foresaw, you know, there were at at that point before we did that deal or, you know, like, right. There's there's 2 different branches here that the the life of the business can take on. I obviously don't know what either of them look like, but the the non polychrome one did look like, to be honest, you know, because of the because of my and the team's lack of of experience around, like, building businesses like like Flagsmith now is, is it kind of just sort of, like, pottering along, not really you know what I mean? Like, never really kind of, like, getting to that critical mass where it can bring on, like, you know, have a have a reason to reassigns team of in engineering and and and marketing and and sales and stuff like that.
Ben:And it just kind of would've, I think it just would've got pretty boring, to be perfectly honest with you. I mean, a lot of businesses doing the same thing every day. Right? But, if you're doing the same thing every day and nothing's really changing in terms of the, you know, the revenue of the company, you know, who you're competing against, the sorts of clients you're talking to. Then, you know, and then on on the the other the other side of that branch was like, you know, the growth just sort of started to to really come alive.
Ben:We have we have money to bring in engineering, which meant that that was more interesting because, you know, we weren't just sort of, like, firefighting customer issues and dealing with infrastructure problems, and we had people who could work on features in the platform and and all this sort of stuff. So, yeah, I mean, you know, like, you you try and make a calculated a calculated risk on that. Right?
Arvid:Yes. It's and that's what it is. I it feels like, it it found a choice to make at this point. Do you wanna keep slugging along, or do you wanna try something else? And I'm I'm glad you made that choice.
Arvid:I think it's very inspiring, inspiring to to see somebody give up something to then get something. I think that's that's that's quite noticeable.
Ben:Yeah. I mean, it I I think it's I think as well, like, I think there's a bit of a a perception that a lot of these deals are 0 sum, and, you know, maybe that's I don't know. Maybe it's is it maybe that's a slightly American mindset, and, you know, like, I was definitely of the opinion that, yeah, this this doesn't have to be a 0 sum thing. And it's also you know, I mean, one of the best pieces of advice I ever got given was, like, you have to sell into a growing market. Right?
Ben:If you're selling into a contracting market, then you know?
Arvid:Or or, you know, if if
Ben:you're selling into something that's growing, then then it by definition, it's not zero sum for those those entrants in that market, right, those those players in that market. And so, yeah, like, that that was definitely, you know, at the top of mind when we when we did that. But I think also, like, you know, you you get the the sort of the companies that have these almost, like, celebrity founders. Right? Like, in in, you know, like, the Hacker News folk.
Ben:Like, there's, you know, there's there's probably 10 or 15 or 20, you know, founder CEOs that that that people, you know, in in in a world could name and, you know, kind of adulation towards, and they're kind of rock stars type thing. And I think I think, you know, like, it's not just those those people running those businesses, and there there's definitely, like, there becomes a little they they become a little bit of a marketing thing themselves. Right? And so that that then fills down that if you're much smaller, you're like, oh, you know, like but, yeah, but this guy's doing it all by himself. You know what I mean?
Ben:And that's obviously not the case. Right? They're they've got, like, 8,000 people, pushing behind them and, you know, they might spend all of their days doing podcast recordings and photo opportunities and stuff like that.
Arvid:Right? Just like you are right now.
Ben:I don't I don't spend all my time to keep track.
Arvid:No. But it's it's it's a it's a valid way of doing some kind of brand adjacent marketing, like building a personal brand as a founder and then kinda letting that, attention leak into the business as almost a lead lead generator, a funnel. That works too if you have the personality for it. If if you can deal with people on the Internet, right, that's another thing. Like, if you if you can handle the trolls and stuff.
Arvid:And if you want to spend your time on that instead of doing technical work or doing customer outreach or all of this. So it's it's always this kind of balance, a choice that you make in that point. But, yeah, I think what you've done is just very consciously, very intentionally over many, many years putting in the work. That is quite obvious with us. And it feels like that's the way to do it.
Arvid:Right?
Ben:Yeah. I mean, I I think there's no, you know, there there there have been situations in in the industry. I remember when, like, Google AdWords launched, and there was there was, like, ad tech companies being launched and then, like, 9 months later being bought by Google for, like, $700,000,000 or whatever. And there was loads of those. Right?
Ben:If you if you look, if you research that, that happened a lot. Those those days, you know, yeah, those those days are gone. May maybe there were those opportunities with crypto, but then that came with all the nonsense around it and stuff. So, yeah, we've we've definitely been, the businesses that I've worked on have have been in the concept that, yeah, you just you know, you you can't I mean, we, you know, we we we had the SaaS platform live for, you know, well over a year. And after a year, I think we had a few £100 in MRR, which, you know, like, was like, okay.
Ben:You know, it's gonna pay for a few beers after work and maybe some some of the AWS cost.
Arvid:Yeah. But that's where it starts. Right? That's where it starts. Like, you have to have a the proof of thing, and then you make it better, 20 times better with the right people.
Ben:Yeah. I mean, it's it's kinda weird as well because the the number goes up so gradually, you know, and you don't really notice. And then, I actually so there's another business that I work on. It's a virtual office service called the Hoxton Mix in London, And we did Y Combinator or we applied to Y Combinator, and we we got an interview. We we we almost got in.
Ben:We had 2 interviews, actually. And I always wonder what would have happened if we'd have got in. But, anyway, we didn't. But I remember when we were at Y Combinator, we were, £18,000 MRR, which is quite you know, it was it was a bit more of an established business. And that's always been interesting because that that's been a marker of a number.
Ben:Whenever I look at the MRO now, I'm like, oh, okay. Yeah. That's great because it was $20 when we went and met, you know, Sam Altman and blah blah blah. And, actually, it's kind of annoying because, so Fliespin has, I guess, the MRO that it had when we when we did the Polychrome deal. But the yeah.
Ben:The thing is the number increases so gradually that it's often really hard. It's like something about the human brain that isn't able to stand back and go, wow. Like, that's a massive achievement. Do you know what I mean? Like Yes.
Ben:Even even if at the end of the year, like, you know, we we had to get together as a as a team and go, look. You know, we've had this year. It's gone from this to that and blah blah blah, and we've launched these things. And even if you literally sit down, write it out, and then read it back to yourself, you it's still very, very hard to just conceptualize, that growth.
Arvid:Yeah.
Ben:It's or maybe it's just me. I don't know, but it's it's kind of annoying.
Arvid:I think I think it's pretty common. I think it's it's it's another form of hedonic adaptation, kind of. Right? You get used to it. Like, it's the the the hamster wheel of things, and it's like, okay.
Arvid:So that's 40,000. Now it's at 45,000. No. That's alright. I mean, that's still a quite significant improvement.
Arvid:Right? It's like, yeah, what is it? 12.5%. Like, it's it's it's a lot of change, yet the numbers are so big. I think that's our brain.
Arvid:The the numbers are so big. Even over a thousand numbers are almost incomprehensively huge. So it's just big and slightly bigger. It's all big. So I I mean, it's a fun problem to have that your MRR is so high that even an increase in MRR is like, oh, whatever.
Ben:I mean, it's not it's not the size of it. It's just the it's the kind of the linear growth as well. Like, it that's the thing. It's you're kind of like, you know? But, yeah, like, that's that's what I'm getting at.
Ben:Like, it's you know, I mentioned it's not it's never got that kinda asymptotic line. But the thing is, you know, if you spend 7 years doing something and it's growing at a linear rate or or close to a linear rate, and you can do things to make that that angle of that slope slightly different, you know, that time is a an immovable force in that function of that that maths. Right? Like, that's, you know, and I I think, there's nothing wrong with that. Right?
Ben:Like, that's the other great thing about not going down the VC route is we we don't care. Like, you know, like, we we we actually don't care, because we've got no there's no clock ticking in the background around anything to do with a business at all. Right? And that's, that's that's the thing that we avoided selling when when we decided not to do that path. Yeah.
Arvid:Yeah. Being able to compound at your own rate, like, at your own accepted time, that's that is a that is that's almost a gift to you as a as a business owner. Right? Like, that you can make your choices without feeling this anxiety inducing pressure of some random arbitrary choice that was made 2 years ago. Right?
Ben:Yeah. Right. Yeah. You know, you you've done your seed round. Right?
Ben:Now you need to grow x, month over month to get to the next stage and then y and then you know, like, we we don't we've never had that ever. And, you know, so, like, I'm happy, you know, yet may maybe maybe the business could be bigger.
Arvid:Right. You always can.
Ben:But, I mean, I I who cares? Like
Arvid:Right. It probably makes it much more interesting to look at the future of the business. Right? If if that's not just growth numbers that you're looking at, you can also look at completely different things. What does the future hold for Flaxsmith?
Arvid:Where do you wanna take it?
Ben:Yeah. So that there's some having a sustainable, like, open source project, that's been super important to us as well. Right? Like, leaving leaving a a a really, well engineered, usable, reliable platform that people can use for free has been, like, really, really, important. And we've never been shy of saying, you know, we're gonna make the SAML component closed source because that's how we make our money, and that's how the business stays itself, and that's how we have 5 engineers working on the product full time, you know, making the product better.
Ben:So we've never been shy about doing that. It's also interesting as well because we've always wanted to have a very defined focus of what the product does, And there's always been a temptation to go like, oh, we could do analytics as well, or we could do, like, engineering analysis, and we could, like, look at, you know, I don't know, cycle time of flags and blah blah blah. Maybe maybe we might do something in the very distant future around that, but we've never we've never you know, that or I I don't know, like, make it really good for, like, AB testing. And so area of concern for what the product should the problems the product should solve has always been quite rigid and quite narrow. And so we're kind of getting to the point where we're we're sort of this this feature's coming down the line around, like being able to wrap a collection of changes together to to form a release and and doing stuff for larger teams.
Ben:Like, we just did a much more finely grained, like, permissioning system based on tags so that multi multi team single projects with multiple teams don't stepping on each other's toes and stuff like that. And that that's another thing as well that I think you never hear and maybe we should hear more often is like, you know, maybe, like, the feature set is pretty much like we're kind of you know, like, that's kind of right. That's fine. There's a constant sort of constant uphill battle dealing with framework changes, especially the JavaScript ecosystem. PHP is very very well behaved for that, I think, and Laravel is, like, you know, making it secure, making it performant, having it, you know, all constantly improving the the ergonomics of running it, installing it, upgrading it, all that sort of stuff.
Ben:But in terms of, like, what it looks like when you log in in 3 years, it's gonna look the same, I think. I mean, like, it might have a style refresh or something, but in terms of, like, what it's for and what you can do with the platform, it's probably not gonna be different. Yeah.
Arvid:That is so refreshing to hear. It is so refreshing to hear somebody say, hey. I like what we have, and I think it's fine. We might do, you know, operational stuff, performance stuff, obviously. That's just because there are so many externalities that you need to take care of.
Arvid:Right? Integrations and whatever. But that that's that's always out of your control, so you need to react to it. But proactively, we have what we need. And I I love this.
Arvid:Tight scoping, and then you you fill it with your product, and then your product is, for the lack of a better term, done. Or, you know, feature complete, like words we don't usually say in in software engineering, but it's enough.
Ben:And we're lucky as well because the concept of feature flags, just this dynamically controllable Boolean at its simplest is so powerful in terms of, like especially for larger teams that, you know, it's almost like you're kind of if you if you put too much you know, like, the dish is perfect. Right? Like, and then if you just start adding a bunch of stuff, then you're kinda like, what you know, like, you're almost kind of, like, encouraging bad engineering practices if you start putting fireworks and whizbang features everywhere because people would have lost sight of, like, what it's there for in the 1st place as
Arvid:well. Yeah. Yeah. It's it's kind of like a very specific knife that a chef would use.
Ben:Right. I like
Arvid:it's it's it's not meant to be the the knife for everything. It's a very specific professional expert level tool, And it should stay that because I think I mean, that's almost the UNIX philosophy of of small tools. Right?
Ben:Exactly. Yeah. Very much so.
Arvid:I I like that. Like, do one thing really well, and if you need to do another thing, build that other thing, but leave the first thing alone. I love that.
Ben:Yeah. That's that that philosophy has definitely been a lot of
Arvid:what what we've you know, it's
Ben:so easy to say yes to building a new feature, but it's so so much harder and and takes so much more control to say no for short.
Arvid:Well, I I love this. I love your approach to things. I love the choices you made along the way, the rationale behind them. It's really, really cool. And I love the whole journey.
Arvid:It's really nice to see something bootstrapping and then going the weird entrepreneurial path of doing whatever needs to be done. Right? It's just wonderful to watch. And I I hope that your journey still has many decades to come for for for anybody who wants to join you on the journey and wants to see, like, what you've been doing, what you've been working on, Where should people go to follow you and see what you're working on?
Ben:Yeah. So, GitHub.com/flagsmith, that's the the first place to go. We have a Discord server, which, people can join and hack around. Like, all all of the stuff is is is is findable from from there. Flowsmith.com is the website.
Ben:And then, yeah, the other one to shout out is, Openfeature, which is open feature dot dev, and there's a CNCF Slack that has an open feature channel as well. There's definitely a ton of work. Like, the open feature surface area of work is way, way larger than ours. And so I would yeah. If people are interested in feature fragrant or if they've they've used it and they wanna, get involved in, a growing CNCF project, which is kind of cool, there's also yeah.
Ben:Like, we'll be at the KubeCon in London in, I think it's either February or March, and there's an open feature, like, morning, I think it is, where people can all get in a room and nerd out about, like, feature flagging and and figure out what to work on. So, yeah, that they're the they're the main places to check us out. Yeah.
Arvid:That's exciting. And I love this. I I love that you're shouting out this this movement. That's really cool. Like, that that is really very wonderful.
Arvid:Thank you so much for sharing all of this with me today. That was quite insightful. It was an absolute pleasure. Thank you so much.
Ben:Thanks for having me, Avid.
Arvid:And that's it for today. Thank you so much for listening to the Bootstrap founder. You can find me on Twitter at avidkahlaviadkhl. You find my books and my Twitter course tattoo. And if you wanna support me in this show, please tell everybody you know about PodScan.
Arvid:F m. Leave a rating review. It makes a massive difference if you show up there because then the podcast will show up in other people's feeds. Any of this will help the show. Thank you so much for listening.
Arvid:Have a wonderful day, and bye bye.
Creators and Guests
