Episode 85
· 39:51
And so that's why organizations get boring and crappy when they're big and stop building new things. You already have to like be encouraging, especially as you get bigger and you have lots of users and stuff. You have to go out of your way to encourage ideas rather than everything going a bit sensible and boring. And then you're hitting some local max and, with what you're working on. Like a matter of pride starts motivating people.
Jack Bridger:Hi, everyone. You're listening to Scaling Deltals. I am joined today by James Hawkins from post doc, who is one of our first returning guests. And, James, you've been quite busy since we spoke a year and a half ago. We were just catching up.
Jack Bridger:Could you share a little bit about Posthog and and what you've been up to in the last year and a half? Sure.
James Hawkins:Yeah. So Posthog is a platform that lets engineers build successful products. We try and give you all the tools you need in 1, where we started building in 2020, kind of just before COVID hit. And I've forgotten the second part of your question.
Jack Bridger:My second part was, what you've been up to since that? Well, I can actually say partly because I can say that just today, the project I'm working on, the offer engineer was like, we need oh, we should install post hoc, and there wasn't even, I was like, I'm interviewing James from post hoc later as well. So I'm using it today, and it's really good. We're using session recording. And then the company I do contracting with are also using it completely independent of me.
Jack Bridger:I'm not using it. The other team is using it. And they're, like, getting on really well with it, and doing some all sorts of compliance stuff as well. So
James Hawkins:Good. Yeah. I will say, no. I'm glad to hear it. I can, I also endorse post doc too?
James Hawkins:And I have double number of children I have since last time, to 2, and I will be having no more.
Jack Bridger:So yeah. And I can imagine that you, like so we're gonna dig in, but I will share one quick anecdote researching James. So, to give you a background on James as a person, James in 2020 wrote an article about hoovering and the mental overhead of monitoring when hoovering needs to be done. And I I wonder if you could share a little bit about that as like a Sure.
James Hawkins:So, on hoovering, my when I was young, I'll be like, why is my wife so pissed off that I haven't, like, I was like, I'm always happy to do the poofering. So why is she annoyed about it? I don't understand. And then she's like, well, James, the thing is, like, you're not actually thinking about it, or maintaining the house, whatever. You're just following instructions, which is much simpler, and it's not tiring.
James Hawkins:And it doesn't take the same amount of energy. Like, it's a bit like being a passenger in a car compared to the driver. And so I think this analogy is useful when you think about kind of, you know, who's decide, like, when when we're hiring, for example, we're looking for this kind of mentality of like, is this the person that's gonna realize that the hoovering needs doing? Or is this the person that will just do hoovering at fast? And there's a world of difference between the 2.
James Hawkins:And when you scale up business, if everyone is the kind of person that would realize that Hoovering needs doing, the company kind of runs in this like weird, autonomous distributed kind of way. And it makes it much less stressful, which gives you space to think about your strategy or hiring or fundraising or whatever else that you need to do, that you maybe aren't delegating.
Jack Bridger:Yeah. I I and so I think this leads us on to, like, 2 things. And the first thing that we'll we'll come back to is about hiring the right people and how you run the organizations. But I think something to kind of keep in mind for the listener as you go through that I'm thinking about is just noticing how everything that James does, he's thinking about, like, a level of abstraction above. And I think this is really uncommon, and I think it seems from the outside to be something that post doc is really good good at.
Jack Bridger:And yeah. So that's so awesome, James, by the way.
James Hawkins:We've put we've put quite a lot of work into it. Like, we have a public handbook that explains how we work, which I think has helped us to think like this. So for example, every time we have a particular problem, we try and create some kind of principle around it. And you know, the situation would be something like, a team member has asked to go on a 3 month sabbatical. And instead of just, like, we just won't solve for that team member, we'll solve the company.
James Hawkins:So we're well, in general, how do we handle people asking for some articles, like, what's our philosophy around this? And then whatever the output is, we'll put it into the handbook of the business. And the next time we'd have to think about it. And then occasionally, we'll revise stuff if one of our various philosophies turns out to be pretty stupid. Again, a book like Ray Dalio has a book on principles, which I think I I read long after we started doing this handbook, but it was quite affirming.
James Hawkins:I was like, actually, I can totally see that he has a similar mentality, for example, and is much more eloquent than me. So you should go read his book.
Jack Bridger:This episode is brought to you by Work OS. At some point, you're gonna land a big customer and they're gonna ask you for enterprise features. That's where Work OS comes in because they give you these features out of the box. Features like skin provisioning, SAML authentication, and audit logs. They have an easy to use API and they're trusted by big dev tools like Vercel as well as smaller fast growing dev tools like Nock.
Jack Bridger:So if you're looking to cross the enterprise chasm and make yourself enterprise ready, check out WorkOS. We've also done an episode with Michael, the founder of WorkOS, where he shares a lot of tips around crossing the enterprise chasm, landing your 1st enterprise deals, and making sure that you're ready for them. Thanks, Workhorse, for sponsoring the podcast and back to the show. Yeah. That's, that's interesting.
Jack Bridger:And, and then going back to, so speaking about how you, hire people that are willing to check when the hoovering needs to be doing to be done as well. When we were talking about doing this episode, one of the things that you were mentioning was that you're aggressively trying to remove 1 on 1 meetings, and I know you were already doing very few meetings as a company. Could you talk a bit about how you why you think that's important and how you do it?
James Hawkins:One of the questions we asked ourselves internally is kind of the, like, what are we winning on as a business? Like, what is the competency we need? Like, what are the competencies we need? And we're questioning, like, is it design? Like, how much does that matter to us?
James Hawkins:Whatever. And eventually, we kinda realized, again, because some decisions are like, okay, we're optimizing for design, quality and control, or we're optimizing for something else. Eventually, we realized, okay, speed, speed matters to us. We're trying to build every tool that developers need to build a successful product. So that means like a crap ton of tools, like we need to really ship an awful lot like much more than a normal company.
James Hawkins:And like today, we have, we've hired a quite we've just saw hiring quite a bit. I think we're now like 47 or 50 people or something. And we have 8 products. And like each individual product, we're usually competing with a team of a couple of 100 people who are only working on that product. So we've been really good at shipping a, like, ridiculous mass surface area per person that works here.
James Hawkins:When it comes to kind of hiring people and stuff like that, I guess we've just been we're basically structured in small teams, and each team operates like a small startup. And the principle that we kind of borrowed from Y Combinator in 2020. And we're like, okay. Like, at YC, there were like 250 companies or about 500 people roughly, and 250 products got built, which is a really good ratio. And so why is that?
James Hawkins:Well, people have a lot of autonomy, they have ownership, like the people who there are quite intrinsically motivated. They're all kind of product engineer type people. And so we kind of thought like, well, actually, I think they're getting it right in terms of the amount of stuff that's getting built
Jack Bridger:in that
James Hawkins:kind of structure.
Jack Bridger:Interesting. Yeah.
James Hawkins:So we have a team for every product that we build. The team is multidisciplinary, and the concept is they can ship without outside interference. So they know permission to go outside of their team to do anything, Like, as far as humanly possible. We give like, they will we now are a bit bigger, we track every, like, every product has its own growth review where we go through all the metrics and stuff. And we're kind of like, my mentality is we're just trying to coach these teams, we put as much energy into each product.
James Hawkins:We don't just base it on like the one that's the biggest or the most successful all around financially. And we hire quite entrepreneurial developers, like I think this is correct. But I might get a fact check. I think it's like 40% of our team have run have been founders before or something like that. But they're all technical people too.
James Hawkins:And so yeah, we have people that are very intrinsically motivated, who are good at just being like, cool, make something from nothing like, and there's loads and loads of other stuff like we're all remote. And I think an autonomous environment works better when you're all remote. And I hire people with more experience on the whole on average, rather than less experience tends to work better, I think. And these are all lessons we've kind of learned as we've scaled and hired lots of people. But yeah, ultimately, all this stuff comes down to we're trying to optimize for velocity.
James Hawkins:And so we've kind of got a structure that enables us to do that, basically.
Jack Bridger:Yeah. That's interesting. When you describe that, like, the thing it reminded me of is there's this guy, Andrew Wilkinson, who I I don't know if you heard of him, but he's like he owns this company called Tiny, and they basically are a holding company for, like, many different products. Yeah. And he meets with the individual founders and gets, like, you know, or CEOs and, gets, like, updates on, like, their projections and stuff and, like, unblocks them, helps them make key hires, all that sort of stuff.
Jack Bridger:It always sounded like that when he described it.
James Hawkins:Yeah. It's not a million miles from being an investor in some ways. Like, it I mean, it's quite far from that, but some days I'm like, maybe at scale. Like, another good example is how AWS operates. Like, you know, AWS is doing a a check.
James Hawkins:Last time I checked, it was, like, $80,000,000,000 a year in revenue or something, and they have, like, hundreds of products. But each product is a small team ultimately. I'm sure they have a lot more bureaucracy than us, but they have a very similar meeting system and so on. So it's kind of. And then the advantage is like when you're also the other thing you want when it comes to like 1 on ones and stuff, and very engineering heavy, like, again, we need to have like another cool thing is in order to be quick at shipping, we need to have a really high fraction of our team as developers.
James Hawkins:So we've tried to minimize everything else that we don't have an outbound, we don't do any outbound sales whatsoever, for example. So we're trying to maintain a really high ratio of engineers to people who can't do engineering work in order that we have as many people as possible writing code and building stuff. But again, like, AWS is again, AWS has been actually a sort of source of inspiration for the way we run, even though we're, like, totally puny compared to them.
Jack Bridger:Yeah. Actually, Franky, like, the question on, like, the engineers, like, it's kind of easy to imagine a world where, like, just having engineers goes wrong if, like, you're you're envisioning, like, something with, you know, no styling, and they don't do any marketing, and, like, no one uses it. It's, like, the kind of classic, like, indie hacker engineer thing. Like, how do you hire, like, teams which are, like, primarily engineers and still, like, build products that people really want to use and that, like, the word spreads as well?
James Hawkins:I think it's only self selection. Like, we no. We have no product design security and things that are kinda shocking probably. We have no product design by default. So we don't scope stuff out in Figma or anything before, like, literally engineer builds it however they want.
James Hawkins:And we have no product management by default either. It changes a little bit as we get bigger. We don't have a the person that decides what to build is a developer, a postdoc even above, like, we have one product manager about to hire a second. They're exclusively in a coaching role. Like, they do very deep data analysis and stuff and go deeper on kind of product questions you might have, or frankly, we think product manager would actually do a better job than a developer would.
James Hawkins:But they don't dictate the roadmap or decide things or build product requirement docs or anything. So like no proper default, no really, or at least in the conventional definition of it, and no product design by default either. So yeah, you're like, in theory, that looks like you're set up for a rack in terms of UX and stuff. But the reality has just been that there are lots of developers who are product engineers now, and that's what they consider themselves. Yep.
James Hawkins:I think it's 1. I think our product strategy is a little bit simpler than if you're building like, we're not building new like, every product that we ship into our platform, we will only ship products that already have product market fit externally, on day 1. The innovation becomes kind of once the product is out in an MVP, or has once it's got to product market fit, and we're starting to make money off it, and our attention is good, and so on. The, that kind of means it's just a little bit simpler. Like, if I just told you, like, hey, you know, come up with some UX or product analytics, you would be able to do it like half baked kind of reasonable job pretty easily.
James Hawkins:We're not copying stuff, but it's just like everyone's so committed with the concepts of the tools we build. Everyone's used to it. Yeah. It's just like, not as hard. Whereas I think if, for example, it was like, we're coming up with a new, you know, if we're like the first ever people to conceive of the idea of like a vessel to contain water or something, like, that's a massive challenge.
Jack Bridger:You know,
James Hawkins:so when my hands are pretty good, like, and
Jack Bridger:that's just
James Hawkins:the best I can come up with. Like But
Jack Bridger:it just yeah. But it just falls through. And I just don't Yeah.
James Hawkins:Some dude's like, no. No. No. We should, like, dig this thing out of the ground and, like, bake it and then use that for to store things in or whatever. Like so I think, our product strategy enables a lower it creates less work from a design, like, product design perspective too.
James Hawkins:So that probably makes it a little bit easier. And then we also have, like, a now we've got a big user base. It means that we get into an iterative cycle with users very quickly. Like, we ship something users, we often ship it in like a private preview for a few customers. And then like, I'll put it in like a feature preview where people opt in, then we roll it out gradually.
James Hawkins:And we use our own tooling and stuff to do all this stuff. So it's very quick as well to go into users' hands and start iterating with them. And, you know, our engineers will talk to our customers, they'll solve their support problems directly. Like we got to 100,000 users before we had any dedicated support people. And like the first guy we hired is, like, 18, and it's his 1st job, or, like, 2nd job or, like, 1st proper job, I guess.
James Hawkins:And then I have to a 170,000 plus users. So, yeah, we've kind of, we're pretty good at getting stuff into users hands quickly and getting our engineers to just work with them directly. And then we're hiring for people that kind of developers that you kind of see talking to users, and you have a history of doing that kind of work. But it is a thing now. It's a trend we're actually trying to encourage, like, our tooling is designed to help other companies operate more like this.
James Hawkins:Because it just gets rid of tons of bureaucracy and specialization and like handoffs between teams and alignment and meetings and bullshit. And these are all things that get in the way of engineers, like, and get in the way of developers building successful products, which is what we wanna achieve.
Jack Bridger:Yeah. That's actually interesting that you, I saw in your site, you have, like, your ICP as like a public thing, which is also, like, kind of actually seems obvious when you see that, like, you would, of course, like, you would want to make that public. Yeah. But I guess no one does. But, like, did you always know that that was your ICP or, like because you are the ICP of I think you basically describe a post hoc as, like, the ICP of post hoc.
James Hawkins:Yep. Yeah. We absolutely are. We're no. ICP's been we're in a big journey with, like, the ideal customer profile that we're shooting for.
James Hawkins:Like, we've been all over the place trying to get this right. I used to talk all the time after our our board meetings. So it was kind of always like, who are you building for? Well, how are they reacting to it? And I was like, why are we getting asked this question, like, over and over and over again?
James Hawkins:And we have this guy called Ali Regani, who's brilliant. He used to lead the continuity fund at YC. I'm like, I just know we're going to be that that this question is gonna come up and up and up and up. I think we're basically we in the early days, we're building for engineers then kinda like startup like, engineers who worked in startups, and then we were kind of like, oh, maybe actually we're focused on self hosting. So it's companies that they also need to work in places that care about keeping data in their infrastructure, because all this, like, we're struggling with our in cloud early on.
James Hawkins:And then eventually, what happened was is our product quality went up. We realized, oh, actually, no. It is literally like, is developers building new products, whether it's in a big company or a small company. We're not folk like, we have a self hosted open source product, but our focus isn't that. It's like, yeah, cloud, like, cloud adoption.
James Hawkins:And we have, like, SOC 2 and HIPAA and stuff. But we have this, like, long journey over we basically raised our series, I think at the end of 2020 ish, probably we started focusing on this, because we've had so much noise from the repo because we had adoption everywhere. We had, like, a chain of, like, a chain of 3 pizza restaurants in Ukraine using us, and then we have, like, international banks using us and stuff. And it was just like, you don't have loads of YC startups using us. And it's sort of we had so much noise.
James Hawkins:The challenge was like, okay. Who are we listening to? And then it was also like, do we care about product managers? Do we care about marketing people or growth engineers, or do we just care about engineers? The answer wound up eventually just being like, don't we just care about engineers?
James Hawkins:These other users will use the product, but they're secondary. So now we're not gonna optimize, like, our pricing model or our marketing or sales towards these other user types. We kind of our go to market is basically developers insist on our product, and other ones have to implement it. And because we work with new digital products, the engineers usually pick the tooling, because they're the only people there. Like in a startup, a good startup usually is all technical people.
James Hawkins:So like developers are doing every role, like the developers, the salespeople, they're the support people, they're the marketing people. And so we're like, actually, yeah, we should just focus on developers, do a good enough job there, and get enough products being adopted. That it's so annoying to move off a postdoc anyway, that like, the other users will tolerate us. That's what we're like, we need to be compatible with them. But we don't have to kind of market towards them or anything.
James Hawkins:It's kind of what we've learned so far. Maybe that will change at more scale, but I'm not sure.
Jack Bridger:Yeah. That's funny. It's kinda funny how, like, the conventional view is like or you hear it all the time. It's like developers don't have buying power. And then your your strategy is entirely based on, like actually, they have all of the buying power at the good start ups, and it's I I guess your belief is that it's gonna change as well in that all the best companies are
James Hawkins:Yeah. I think it's a direction of travel. Like, we've been trying to understand in so, like, we interviewed, like, a bunch of, like, series, like, CD, startups at least, trying to understand, like, who decides what to build. Like, or, like, does the most senior product person report into the CTO ultimately, and I were trying and like, most of the majority, like, we had like a slim majority, I think it was like 5 out of 8 or something. We're like, don't Yeah, it's that way around.
James Hawkins:And then like, of the remaining 3, I think 2 of them were like, we're trying to make this change in the in favor of engineering. Interesting. So I think and those are companies that aren't users. It was, like, random ones. And so I think Yeah.
James Hawkins:It was a small sample, but we've, like, been trying to understand this. And now it really feels like a strong trend. It's exciting. We're getting engineers kind of like when we post job ads. Now we have like this whole our following is like product engineers.
James Hawkins:And so like ton and like we hold to hire product engineers, like engineers who decide what to build and hunt them to apply for jobs because they know that's kinda how we work, where they'll be able to, pick what they work on. Yeah. So it's like we're in some sort of weird pyramid hype scheme or something.
Jack Bridger:That's, yeah. That's, one one of the things as well you mentioned there is, like, the kind of, like, the open source project and how and for people that I don't know if we touched on this, but you started out as, like, this open source, analytics or product analytics with I don't know if that's how you describe it, but, like and that got a ton of adoption before you had any of a cloud offering, and then you launched cloud offering, and it did really well. And now you're just focused on the cloud, like, the cloud service. But do you think that if you do you think posthog could have been built the way it is now without those Ukrainian pizza chains along the way? No.
James Hawkins:Yeah. I mean, like, it'd be much harder. Basically, had I been able to, like, fundraise tens of 1,000,000 of dollars and convince and then just, like, hire a whole team and then just, like, build by ourselves and just ship, like, 10 products in 1 or like, we have 8 products now. So build 8 products from scratch, without really much validation. Like, perhaps we could have done this, but, like, it's a particularly unappealing style of startup to build from scratch to get multi products because you have to build everything.
James Hawkins:And that's, you know, and like the fundamental problem with startups is like, they don't care. And it doesn't the 2 feel like conflicting. Like, it's pretty hard to like, if your goal is to figure out as quick as possible if anyone cares. And then your goal is also to build like multiple each of the $1,000,000,000 competitor to compete with them. You're gonna have a it's like feels like a huge gamble.
James Hawkins:So I think it would have been way harder. So just so we had a little bit of traction and a bit of a following, which I think made it easier. Again, a good quote actually I had from Ali, who I mentioned was, as you get, like, once once you have a bunch of users, you start being able to see around corners, that you just have much more
Jack Bridger:insight into,
James Hawkins:it's so hard to get any insight in the early days. Like, in the early days, it's like you're I'm trying to I'm, like, harassing people on the team to use our stuff. Yeah. One at a time. Everyone's ignoring me.
James Hawkins:Maybe I get like a meeting every now and then.
Jack Bridger:Where I'm at at the moment.
James Hawkins:Whereas now the page is like, Matt, if you post something, like, it could just wind up on, like, it'll wind up on social media and stuff without doing anything. And a whole lot of people react to it and all sorts. So, it's made it easier. I couldn't have figured out the correct strategy. So the answer is actually no, like, I could not figure out the correct strategy with the information I had at the time.
James Hawkins:Like I had to like do something. And the problem we paid was close enough to the growth strategy that we're able to get from that to a strategy that's better.
Jack Bridger:Yeah. That's that makes sense. And what what are, like, the kind of examples of, like, sitting around the corner?
James Hawkins:An example would be we have tons of come like, lots of users were asking to stream data from postdocs to a data warehouse. And we're like, oh, okay. Like, we kinda understand. Like, they told us. The users were like, this is because we want to join our product data with our revenue data.
James Hawkins:Like, we wanna analyze our highest paying customers, for example, or customers who aren't paying anything or whatever. And then there's a bunch of other data we also wanna join with to get more insight. And, like, cool. Okay. So our users have told us that.
James Hawkins:So you've got exports to data warehouses. And they're like, oh, like, tons of people are using this. Like, loads and loads and loads, like, 100 and 100 and 100 like, now thousands of companies are doing that with us. And so we're like, oh, well, like, let's just build a warehouse. Because, like, everyone's streaming data to a warehouse.
James Hawkins:And, like, we've actually seen some companies that blew up, like, they totally rocketed in their growth on us. The first bill is nothing. Then the next bill was $400. And then the next bill was $9,000 And it went all the way up to like $50,000 a month, someone was spending with us is one example. And then they moved to a data warehouse and the modern data stack.
James Hawkins:And we're like, oh, like, we just should really build this because we're not going to churn people who blow up in usage. I mean, exact people we'd want to retain are the exact ones we'll lose. And we think this whole mundane stack is super messy and complicated and dumb. And we'll just build the whole thing in one place. And we won't build it for like Shell or JPMorgan.
James Hawkins:We'll build it for like these startups that are growing quickly that are quite small teams. So they might have like 30 people or 100 people or whatever. But again, we wouldn't have even realized that one day, that was kind of like a thing, because we're not data engineering, data analyst people. But our users kind of pulled us there. And then we did something about it, like we just shipped it.
James Hawkins:And now we have our own data warehouse, for customers to use. And we've got early customers on it and so on. So it's just like one example, but, like, we never realized that, because we just had a lack of, like, industry knowledge, I guess, until we hadn't used this.
Jack Bridger:Yeah. That that makes sense. That's very yeah. So you're, like, kind of just being pulled by them as to what they want you to do and then thinking about Yeah.
James Hawkins:You can kind of just gauge, like, the market much better. I know it's the same with, like, you know, another example now is, like, we just have a public road map. So we just list, like, this is stuff we think about building. And I can give us a really early, like, because it's like a little group of users to talk to early on, and we get like a rough sense of what people actually want. And it often contradicts what you'd expect to feel.
Jack Bridger:Like, one
James Hawkins:of our developers wants to build, we have an item, we haven't got feature flag support for Java, like we haven't got SDK for it. And, like, I think our engineer who runs feature flags is like, we just don't like working with Java. But the thing is, like, 270 updates or something. So it's just like, we probably should just build it. So I think, yeah, we can get a signal, we can kind of use the community and website and stuff as like a signaling mechanism to give us a little bit more insight into what to do to say.
James Hawkins:Yeah. It's like a lot easier going this right. The problems get from being very existential to being quite much narrower. Like, it goes from, like, am I trying to build, like, a bicycle or a marshmallow factory or a shed, all the way down to now? It's like, which SDK do we prioritize, like, computation Yeah.
James Hawkins:Out of Java and, like, Rust or something? So, yeah, the problems are, like, so much narrower, and you have so it's 2 things. It's, like, narrow more narrowly defined problems, and also user to actually run them by. So it gets exponentially easier, as you get bigger, I think.
Jack Bridger:Yeah. It feels like and and that kind of, like, ties in a little bit to how you run post dog as well. You were mentioning just off the call about, how you built the session replay, which is the feature that I I'm I installed today. Yeah. How could you talk about that a little bit?
James Hawkins:Sure. So, yeah, back in the day, post log is, you know, just product analytics and open source and self hostable. So we have, like, a paid self hosted product and whatever. And, we did a hackathon, and one of our engineers decided that he would build session recording. So the ability to watch peep, like little video clips of your users using your product to your website.
James Hawkins:I thought, man, I think the other tools in that industry are kinda lame. They kinda feel a bit cheap to me somehow. I'm not sure. Like, it's a bit invasive. Like, I'm not sure if developers will like it or take it seriously.
James Hawkins:It feels like a marketing thing to do. But the way we work is if an engineer wants to build x, and I want to build y, we will build x. Because also you kind of get people's best work was sort of my belief going into anyway. I'm like, well, if he really enjoys it, like, before we do a really good job, I'm like, you know, it's what will come from it. And so he shipped it, and it completely transformed our entire business.
James Hawkins:So he just bought it. And he built it because like a few users were asking for it. It just felt like one of those features that you should obviously ignore, because because it was like a different product. It's the one we're working on. Yeah.
James Hawkins:And, anyway, he built it immediately, got a lot of usage. And now a few years later, every company on earth that does prototype analytics is basically copies done this, but it's so useful having it together. And he yeah. And that that that shipping that product made us realize, oh, hang on. We can actually like, people really like us because we're multiproducts.
James Hawkins:Then we started charging for it separately, and then we realized, oh, this is great because now we actually got more usage when we started charging for it. It went from being like a feature that was free was the perception of it. So like, oh, no, it's a distinct standalone session recording tool that can compete with like Hotjar or FullStory, because there's pricing attached to it. It feels legit. And then we got more usage and much harder feedback from people, which meant the product quality went up because we could like invest in it properly because we're making money.
James Hawkins:And we had good feedback that was more demanding. And so the like usage went up, the quality went up, and the revenue went up by charging for it too. So kind of treating as a first class product. And then we replicated this for we start looking at like all the other tools. And we think there's like at least 30, and probably more like 50 or 60 plus that we could build with like this.
James Hawkins:And then now we have 8 products, which want to build the next 5. And like none of this would happen. And then this meant that we could compete in cloud because we already differentiated by being wider. So it enabled an awful lot, and it just comes from, like, listening to developers who have experience.
Jack Bridger:Yeah. And how like, this guy who came up with this, who who wanted to build it, like, how how do you think about, like, is he is he, like, now still running the session replay part? Or is he, like, what else have you got? What what else do you wanna build?
James Hawkins:It it comes from all over the place, though, like, these ideas. So, he actually left because he was killed. But he, would just repeat if he had this kind of thing happen. So to give you another example, there's a guy called Marius, who, until very recently ran our product analytics team. He now runs a different team.
James Hawkins:And you you think, okay, the person runs product analytics, like, what are they gonna be working on? And you kind of would assume, like, more filters or, like, more visualization types or something, bugs, scalability. But, no, Maris is like, I'm gonna build Hog QL. I'm like, what is that? And he's like, well, I'm gonna build this, like, middleware layer that, like, translate the standardizes all of our querying between our front end and our database.
James Hawkins:And it will, like, rewrite SQL in a, like, nice format for, which is the database we use. And everyone was like, here are like, 90 reasons why this is a dumb idea. And it won't work. It'll be terrible. It's ridiculous.
James Hawkins:And his idea kind of got really shredded internally. And in reaction to this, he basically just stayed up until 3 in the morning for 3 days and built it and came back with it. And then
Jack Bridger:it's irritated. Everyone was like,
James Hawkins:oh, actually, this is working. It's probably operates. It took another 6 months to get into production or something like that. But now it's completely transformed how we build. Like now everything sits on top of this thing.
James Hawkins:We can ship new visualization types much. We can ship, like, we can, like, embed visualizations really easily across the product set and so on. And now his whole bag is, like, he's in charge of building. He wants to build at the moment, he's literally building a new programming language for us for a bunch of simply crazy sounding reasons. And with that, he will then build 2 more products, because we wanna, like, build a fuller featured CDP and a messaging tool.
James Hawkins:Like I said, this this programming language is is designed to make it easier to ship these products. So, again, this is the only the kind of thought that, like, an engineer will have. It will not come from a product manager ever. It has to become very technical for these ideas, but it's gonna happen repeatedly. So, yeah, you just have to sort back people a little bit, to do this.
James Hawkins:And it does take courage. Like, it doesn't fit into it. It's a you know, you have your agenda. You have what you've told venture capitalists you're gonna do with the business. And then engineers will just call the random stuff.
James Hawkins:That may not sound like what an investor would wanna hear, but we've just stopped caring at this point. And we our engineers are very close. They are similar people to our customers, and they're working directly with our customers, and they're working directly with our codes. They're probably the best place to decide what to build pretty frequently.
Jack Bridger:Yeah. And now does it feel like does it feel like, oh, that's not the right thing to do, like, to do? Or is it like you're so used to, like, you see that perspective now? Or is it, like, fully, like, you're like, I don't think that's a good idea. Oh, it's only I'm just gonna trust you.
James Hawkins:Yeah. It's only a couple of years to to, like, build the trust to do it. The cut, like, Maris push the sort of concept of Maris talks about is, in stand up com in comedy, there's the concept of yes. And so by default, an organization will be no, but like someone suggests something, and you will sit there and you will say, here are 20 reasons why this is done. Like, I'm gonna play devil's advocate, whatever, and everything gets torn apart.
James Hawkins:And then you do that repeatedly. And people just stop suggesting ideas. Because there's no point it takes energy to suggest ideas. And I feel like to put their stuff on the line. And so that's why organizations get boring and crappy when they're big and stop building new things.
James Hawkins:And so we were really like, okay, no, this is actually very important. Like, this is a good back. This has been a good example of like, no, I'm going to push for it. Like, And as the culture said, by default, you need to encourage new ideas to be built. And regardless, you don't need consensus to ship them.
James Hawkins:So it does kind of mean, but I think it's a really important cultural point. There's some stuff on like, I think Pixar has this concept of like ideas of franchise, which is pretty similar, where you already have to, like, be encouraging, especially as you get bigger and you have lots of users and stuff to you have to go out of your way to encourage ideas rather than everything going a bit sensible and boring, and then you're hitting some local max, with what you're working on.
Jack Bridger:Yeah. And then you have to have, like, innovation days and sort of Yeah. Exactly. Yeah. The yeah.
Jack Bridger:That's a that was actually, like, really mind blowing because I noticed, like, I I I feel like I can be like that sometimes with, like, shut down stuff rather than, like like, I'll have my own idea of, like, okay. We wanna do this, this, and this, and then
James Hawkins:Yep.
Jack Bridger:K. That was that that was a learning for me right there. Yeah. And, one of the things on that strand of, like, fun, it sounds like it's kind of almost like a frivolous thing, but it's I think it's actually important. And you've wrote written about this in an article about, like, taking fun seriously and how Elon Musk, says the most, the most entertaining outcome is the most likely.
James Hawkins:Yeah. If you if you're expecting like, if you're having fun like, basically, if you're okay. So say a story. My cofounder and I have this exec coach because we're, like, living stereotypes of founders. And he's super, super great and competent, whatever.
James Hawkins:We went on his coach. I think he's coached 23 or 24. I don't think he's, like, public. I'm not really sure if he's, like, public about it. He's coached 20 someone named him, but he has, like, 23 or 24 founders who have worked who've built unicorn companies.
James Hawkins:And some of them are, like, big. They're much bigger than, like, 1 or $2,000,000,000 valuation. They're, like, an order of magnitude more. We went out for dinner with him after we've been working with him for, like, 6 months or something. We're like, what's the difference between someone that builds a $50,000,000,000 business compared to, like, a 1 or $2,000,000,000 company?
James Hawkins:And he's like, well, basically, the ones that build the biggest companies like working in them, and, like, they don't look normal. And so I'm like, yeah, I can kinda see why. Like, it just takes a long time to build a big business. So if you hate your job, you're not gonna get there. And I could so easily see how, like, oh, great.
James Hawkins:We get, like, a 100,000,000 revenue. Let me cash out or whatever. And then the whole company kind of falls apart because we've, like, just dragged ourselves there and now everything's, like, churning. I think it happens a lot when we go public, or certainly when they sell. Even when it's like a good exit kind of thing, it's like repeats this.
James Hawkins:This cycle happens every time almost. But then there are exceptions. And so we've been like, well, we're not here to build, like, a 1 or 2,000,000,000 business. And, like, it's not that far from where we are. And, these, like, little point solutions we're competing with are worth that much.
James Hawkins:Like, let's try and go, like, bigger. And, you know, if we sold the company tomorrow, I'd just be building a startup the day after from the first bit super crappy. Yeah. So, yeah, I think, like, fun sort of means that, like, people stick out more. It means you get better work from people because they're enjoying what they're doing really simplistically.
James Hawkins:So it's like a matter of pride starts motivating people like we have this with our website. Like our website was fun to work on.
Jack Bridger:And you can tell when you very fun to go.
James Hawkins:But the team that work on it are incredibly motivated, because it is just kind of like that. And it just means the standard you can just push really far. So, like, it's kinda easy to push people when they're having fun. Or is there anything about time? Like, you can't push people to deliver a world class work, I think.
James Hawkins:Not, like, on a long term basis. And then also, like, simply not churning really strong people is a very basic way to keep high shipping speed up. So, like, listen to them, trust them, give them autonomy, and they'll stick around. And, like, probably they're better than you anyway quite often. So, like, yeah, this stuff, I think, really matters, and it seemed better if you can get it across to your users too and get kind of marketing benefit from it as a secondary thing.
James Hawkins:But yeah, just give me a little bit interesting space. So much nicer to work for here long term. That's kind of the core reason why, like our business, strategy is like kind of based around this. Like, we didn't want to do the self hosted thing long term because it just felt like it was gonna be huge drag. Like, I don't really want to fly around the world and see saying sorry to customers that there's some problem with our infrastructure.
James Hawkins:There's wrong issues stating everything or whatever. So, but we had to, like, do it a little bit to kind of realize that's not what we wanted to do, that much. The cloud product and having a very quick self serve motion was much more appealing. So, yeah, I think I think your I think your fund is not necessary if you want to build a big business. I think if you wanna build a huge business, it probably is.
Jack Bridger:That's a that's a good takeaway. And that's probably that's probably good for living a long time and just having a good life as well, I guess.
James Hawkins:Yeah. Maybe. Unless you get to drive or something. That's what you find.
Jack Bridger:Not too much fun. Yeah. Everything in moderation. Yeah. Everything in moderation.
Jack Bridger:Yeah. Make a make a fun SaaS website. That's that's the kind of fun you should aim for. Yeah. Fine.
James Hawkins:As long as it's, like, a B2B Fintech or something. Yeah. This is a bit like hard learning can be fun, isn't it? It sounds pretty lame now talking about it out loud.
Jack Bridger:No. It's great. It's great. James, I think that's a that's that's a wrap on the on the time. I know you've got a it's it's late in the UK, so, it's probably bedtime.
Jack Bridger:But, thank you so much for Certainly not. Certainly not. Okay. I have more stuff to
James Hawkins:this day. I've got plenty of work. I work, like, I stop in the afternoon, and I work late at night, like a weird day because my kids are asleep, and it's much better than rushing and trying to wrap up, like, 5 or 6 or something. Well, working anyway. Yep.
Jack Bridger:And, so and before you go, if people have you got anything that, like, you wanna, talk about post hoc or, like
James Hawkins:Just my post hoc.
Jack Bridger:My post hoc.
James Hawkins:No. Nothing more to add.
Jack Bridger:Okay. Great. Well, thanks again, James. This was really, really fun. Yeah.
Jack Bridger:Thank you, and, hope everyone goes and installs post hoc.
James Hawkins:Thanks, Jeremy.
Jack Bridger:Thanks, James.
Listen to Scaling DevTools using one of many popular podcasting apps or directories.
