CHURN FM
CHURN FM is the podcast for subscription economy pros. Every Wednesday we hear how the world’s fastest growing companies are tackling churn and using retention & engagement to fuel their growth.
CHURN FM
E255 | Navigating Growth and Retention: Lessons from PostHog’s CEO, James Hawkins
Today on the show we have James Hawkins, CEO and co-founder of PostHog.
In this episode, James shares his experience navigating growth and retention at PostHog. He discusses his journey from working in outbound sales to leading a product-led growth startup and the importance of an inbound-focused approach.
We delve into how PostHog defined and focused on their ideal customer profile (ICP) of engineers, which not only shaped their product development and marketing strategies but also influenced their unique team structure, where every team member can code.
We wrapped up with James talking about the transition from self-hosted to cloud services and their strategic shift to usage-based pricing.
Mentioned Resources
Churn FM is sponsored by Vitally, the all-in-one Customer Success Platform.
[00:00:00] James Hawkins: I think a lot of businesses, it gets really easy to hyper-focus on all of your metrics. But at the end of the day, it's like you just, like if you just do the right thing for customers and have faith, that will work and don't try and, like optimize for your revenue number all of the time, ironically, you will make more money.
[00:00:00] Andrew Michael: This is Churn.FM, the podcast for subscription economy pros. Each week we hear how the world's fastest growing companies are tackling churn and using retention to fuel their growth.
[00:00:00] VO: How do you build a habit-forming product? We crossed over that magic threshold to negative churn. You need to invest in customer success. It always comes down to retention and engagement. Completely bootstrapped, profitable and growing.
[00:00:48] Andrew Michael: Strategies, tactics and ideas brought together to help your business thrive in the subscription economy. I'm your host, Andrew Michael, and here's today's episode.
[00:00:59] Andrew Michael: Hey, James, welcome to the show.
[00:01:01] James Hawkins: Thanks for having me, Andrew.
[00:01:02] Andrew Michael: It's great to have you. For the listeners, James is the CEO and co-founder of PostHog, which helps over 40,000 businesses build better products with a single platform to analyze, test, observe and deploy new features. Prior to PostHog, James was the VP of sales and channels at Arachnys. That's a very bad butchered pronunciation, but yes, Arachnys. My first question for you, James, is how did you go from selling million dollar contracts to large banks to building an open source product for engineers?
[00:01:32] James Hawkins: Sure. That's a good question. I'm basically doing the polar opposite of what I used to do. Maybe it's the touch of the grass is always greener on the other side of the fence. We used to sell our average order value race to work, race to run the sales team was around a million dollars per year, like our biggest contracts went up to about 5 million bucks a year. And I was kind of in charge of our sales team. Totally outbound, selling like, it's like 1990 stake-led sales kind of approach, selling compliance software to large banks.
[00:02:01] James Hawkins: So basically, I just went around the world doing long haul flights everywhere in a suit in hot weather, going into skyscrapers trying to get people to buy our software. It did work. Like we got... That company is not public about the revenue, but we got into like a reasonably large amount of revenue. But churn actually became a huge problem when we got larger because we're so focused on selling new business. It's basically what a roadmap, everything was oriented around. And it was one of not being more product focused throughout was my biggest learning from that job.
[00:02:32] James Hawkins: And so I eventually thought when we got larger, I was like, I just think I could see in the enterprises we're selling into people coming in with a very different approach, like charging on usage, for example, or going in through end users [inaudible]. The bar felt this, kind of clearly feels like I'm winning, strategy. And I'd been thinking about doing my own thing again for awhile. And so eventually I made a leap and went into kind of open source support.
[00:02:56] Andrew Michael: Yeah. It's amazing. Cause it is like, you say almost a complete polar opposite in the sense like selling these large contracts probably takes months to close and then moving towards, like a almost fully product led approach from what I as well gathered as well, the positioning really like geared towards engineers wanting to be like the first product potentially in into the stack and then growing with them so like they'd land and expand motion that Atlassian sort of, I think pioneered in the early days and now pretty much every SaaS business is chasing as well. So super interesting. Like what would you say is the biggest contrast between the two companies?
[00:03:32] James Hawkins: I think that's, the biggest one probably is that PostHog doesn't do any outbound sales at all. We're a completely inbound business. We have about, well we had like tens of thousands of companies are using us. We're around the 10 million run rate range. But we've done all of this and we're having a really quick growth rate by just doubling down on what's generating inbound. A lot of things generate inbound growth, also the kind of things that correlate with a higher retention, like things around being very focused on product, for example. But yeah, almost everyone at PostHog is a developer. I think over 70% of our team are engineers. Whereas, where we previously spent an awful lot of our capital, the majority of it would go into sales. It's really greatly different [to start a company].
[00:04:20] Andrew Michael: Yeah, I was reading that earlier. I loved it as well. I think maybe it will even say something like 100% of team members know how to code, if that was correct. But, yeah.
[00:04:29] James Hawkins: It would make a real point of this. So one of our values is, everybody can code. So, like a good example would be, like we hired some good lotty, who's a superb kind of graphic designer with zero coding experience, but we'll make a point of making sure that she'll be able to do a pull request to put an image onto a website, for example, kind of to get into mentality of an engineer. So kind of everyone gets GitHub deeply, they get pull requests, they get the idea behind kind of commits and merge conflicts, all this kind of stuff. So they get the basic lingo. And I think it helps everyone get into our target audience's mind. And it's the productivity thing. Like I don't like the attitude of like, yo, here's a problem, can someone else clean it up? So with a broad variety of people. So it's kind of part of that too.
[00:05:11] Andrew Michael: Now, I love that as well, because also the other side of it, like as you mentioned, is it looks like it's your ICP, your focus is engineers. And making everyone in the team be able to ship code gives them that empathy and understanding of what their customers are going through a lot better. Because a lot of times in organizations, there's quite a bit of a disconnect between the different departments and customers. You maybe have a product who speaks to customers quite frequently or marketing, but then some other functions less so. This sort of gives a great way, like maybe if you're not frequently speaking with customers, you are the customer to some degree as well from that aspect. So I love that notion and super interested to see how that scales as well with the company as you grow as well.
[00:05:50] Andrew Michael: So it sounds like you're on an interesting journey as well. I read earlier that you launched the product while you were going through YSEED. It was one of the most popular products during that period as well on Hacker News. What was that first sort of like moment when you realized, okay, we're onto something here with the product and because I took a look at the product as well, it's quite a big product and like it's got a large footprint. How did you know in the beginning, like when you want something and at what, like where was the product at, at that stage? How much had you invested in building it?
[00:06:23] James Hawkins: Sure. So it was pretty quick, but the reason it was quick is because we kept failing before. So my co-founder and I spent about six months, the first six months of working together. We averaged a product a month. And so I kind of viewed, like an ideal startup in the early stages of machine that can quickly screw up a bunch of stuff. And so we're kind of on this cycle of like, okay, we try and get an MVP built in a week. I try and get users at the same. I do interview, I try and be… here basically ship some kind of super basic MVP over the course of a week. I'll be trying to get users, then we spend a couple of weeks trying to work out if they were gonna, at least, would they use the product, step one? Like, would they come back to the product, step two? Step three, would they pay for the product?
[00:06:58] James Hawkins: And we kept failing at various stages for all these ideas we're cycling through. We got really frustrated setting up a little tools to understand this user behavior every time. And eventually we were like, oh, why don't we just build kind of, like we started with just product analytics, but now multi-product very heavily. It's the primary reason people buy us is everything in one.
[00:07:16] Andrew Michael: Yeah.
[00:07:16] James Hawkins: But we started with just product analytics and we thought, well, we're both quite… I have… got a sales background, but I have also been an engineer, admittedly a bad one before. And it just felt that these tools were not built for developers. They felt built for kind of product managers to tell engineers to implement. And we were like, okay, we want online access to the data, we want to self host it, all this kind of stuff. So yeah, we were like, okay, pivot number five on to our sixth product.
[00:07:38] Andrew Michael: Yeah.
[00:07:39] James Hawkins: We're gonna build this, we're gonna give ourselves a month. And so we spent basically one month building product analytics out and making it open source. Within the first week, I got a couple of buddies to use it because we just wanted to build it into production, because of all the edge cases and stuff that can surface. Weeks like two and three, we're starting to get random people on the internet, adopting it, like we did a little bit of paid marketing. Basically, I just want to make sure that people could self-serve before we then, week four put it on Hacker News. And then it was like, instantly front page, hundreds and hundreds of signups and stuff. And then everything kind of just blew up from there. So.
[00:08:10] Andrew Michael: Yeah.
[00:08:10] James Hawkins: It was only a four week process. But we had spent six months.
[00:08:14] Andrew Michael: Six months.
[00:08:14] James Hawkins: In total, failing at other ideas that were different first.
[00:08:18] Andrew Michael: Yeah, it's super interesting. And because I think like as well, if you maybe started out with a initial premise of the idea of what PostHog is today, like you would think, okay, like there's no ways we could build all of that to get anywhere close to having an MVP that people would care about, but focusing at that time on product analytics alone is like a beachhead to get in. Like I think, as you mentioned, like the key aspects, which was at least to me, that struck me as unique is like that on-prem open source. Like I think these are things that traditionally like with product analytics and in different products or services. Like there's only really cloud options and for a lot of companies that's even restrictive, I think to a large degree, just with their data privacy and security reasons as well. So I think you opened and tapped into a market potentially that was heavily underserved at the same time.
[00:09:01] James Hawkins: Yeah. And like the irony is we actually ended up changing that later. Like we wound up, like often later on, we're like, why are some people are self-hosting for no reason. And we're like, it would just be easier if we just ran this for you in cloud. So we built a cloud product.
[00:09:12] Andrew Michael: Yeah.
[00:09:13] James Hawkins: But this time, we also built multiple, we've kind of wound up building some of the products next to product analytics, like Session Replay, for example. And then we're like, Oh, actually, now we're different enough in cloud. And our cloud growth is actually faster than our self hosted growth. And it's much less support burden for us and for our customers.
[00:09:29] Andrew Michael: Yeah.
[00:09:29] James Hawkins: And so we end up eventually, even though we had revenue from the self hosted product to sunsetting our paid self hosted products entirely focusing on cloud. And our growth just kept kind of accelerating. So yeah, it kind of gave us the initial traction and people to kind of talk to, which then let us figure out that we should have changed our business model kind of around a year later or so, or maybe two years later or something like that.
[00:09:50] Andrew Michael: Nice. What did those conversations go like then as well? Because I think it's not always easy to make these decisions like to sunset products and it's typically quite a tough decision to make. What did that look like at the time? Like what data were you looking at to inform these decisions and how did the team rally around it?
[00:10:07] James Hawkins: Sure. Yeah. So we had kind of, we had fairly fast growth on both the cloud and self-hosted product at the same time as each other. But it was the amount of engineering work that the self-hosted product was taking was enormous from a technical perspective. We're managing, we're basically managing other people's infrastructure via WhatsApp or email or Slack, because we couldn't have that, like set people who care about data being in on premise of the exact people that won't let you go on to their premise to like do stuff about it.
[00:10:34] James Hawkins: So it's like, therefore, now we're trying to debug Kubernetes remotely, which so we're just losing all this time and it meant that we weren't able to ship more stuff basically. And so it kind of felt like this is working for now. This will be really expensive as a business to scale. I think you could, like, go harder on that approach and build a successful company, but it was kind of a good choice to do. Both options were appealing, but the cloud version was just more us, if I'm honest. It was faster, more iterative, probably smaller ACV customers, but more of them and less friction.
[00:11:06] James Hawkins: So we didn't have to build a lot of, kind of, professional service around the technology itself. It was really painful. We got to the point where I think it was something like, cloud started accelerating so much that it wound up, that self-hosting went from 100% of our revenue to 10% because, cloud kind of took off faster. But those 10% of our revenue was 70% of our support workload. And in terms of user numbers, it was like 5% of our users or something like that. So it was really obvious with hindsight, we should have made that decision earlier. I guess loss aversion is a human flaw, maybe.
[00:11:38] Andrew Michael: Yeah, absolutely. Nice. And so you mentioned, like the company has been growing and scaling pretty fast. And I think more often than not as well, like in the early days, when we see these hyper growth moments, churn is sort of an afterthought and retention with it. Obviously the yin to the yang. How have you been approaching churn and retention then at PostHog? Has it been something that's been given attention or up until now, it's really just been about keeping up with the growth and trying to survive on that end?
[00:12:05] James Hawkins: Yeah, I think, we've been quite good, I think, about caring about it from quite early on. So even in the early days, we looked at, before we even had a pay product, we just open sourced the number of weak active users was the North Star metric for the company because it does reflect acquisition and retention in a kind of balanced way. When we started focusing on revenue, you're right, we basically just focused on new business because we're trying to figure out how to get the business model to work, whatever. But quite quickly, what started happening was, as we got larger still, we charge on usage.
[00:12:35] James Hawkins: And so kind of fluctuations in the usage of our existing base of customers is much more significant in our monthly changes in growth than the new business component that comes in from new signups. That's got more importance still as we've got multi-product on top. So now it's not only like, okay, are they using the product more intensively for product analytics, but it's like, okay, have they also adopted Session Replay or Feature Flags or our Data Warehouse or whatever it might be. So yeah, it's kind of become our primary focus at this stage, which is really cool because then doing a good job. A lot of the small things that mean you're doing a good job of retention are then driving further word of mouth growth. We grow almost entirely through word of mouth and a dash of content marketing, basically.
[00:13:19] Andrew Michael: It's amazing. Yeah, I think definitely, this is a very common pattern. Most startups as well, when they hit their hyper growth inflection points, retention normally is an afterthought, but they've somehow thought of it. And then at some point, when the existing customer base scales to a certain size, that revenue becomes significant enough to start caring about the problem and the challenge more and figuring things out. And yes, definitely, as you start focusing on these things, you start getting back compounding interest on the little wins and improvements that you make within the product. You keep people around longer, you can invest more in developing the product because your LTV grows with it, and just a lot of great compounding effects.
[00:13:57] Andrew Michael: You mentioned though, the idea of usage-based pricing as well for the product. Like what drove those, sort of intuitions to go with a usage-based pricing model? And then how have you dealt with the complexity of pricing a product where you have, multi-product usage-based platform?
[00:14:14] James Hawkins: Sure. I think the reason that, so kind of, I would say the standard at the moment, right, is the opposite. It's like you put people into an annual contract or at least like a monthly one of a set figure. I think this just screams from a user perspective. We are trying to… It's all about you, the technology company, rather than the customer. Like, I want to lock this customer in, whatever. Usage is a leap of faith. Like, hey, I just, it's me kind of showing that I believe that you will find this product useful and you'll use it more and more over time rather than less and less. And so I think if you're building your pricing model kind of defensively around locking in revenue, you're already not focusing on customers enough. So I think that's kind of, this kind of human psychology element, I think is actually quite important to retaining users.
[00:14:55] James Hawkins: One of the things I've learned, especially with open source too, the more stuff we give away, the more we grow revenue and the more money we make, which means we can give more stuff away, which means we can grow revenue even more. So there's that. There is higher complexity, for sure, with usage-based pricing. It's basically the trade-off. We felt we could give better value to customers. There's the other concept. I have the opposite point of view to one of your other guests on this podcast who was talking about how you should price to value.
[00:15:21] James Hawkins: I actually think in certain types of market, it makes more sense to price closer to cost. I think if you're pricing to value, you're opening yourself up to competition from underneath. And so our attitude has been, or our inspiration really, has been looking at things like AWS or Cloudflare, who actively work to drive down prices over time. And then you as their customer know that's happening, which means that you're not going to bother shopping around because you kind of just know it's going to be the cheapest because you can just see them emailing you price reduction.
[00:15:48] James Hawkins: So we've kind of also been like not only giving away, not giving away, but not trying to lock people in, but we also regularly will reduce our pricing as we get our own infrastructure costs down and stuff. We're not pricing exactly, to our cost. We need to make some margin to run the business and stuff, but really, really throwing out the concept of we should try and get as much money out of our customers as possible, I think is enabling us, ironically, to grow our revenue and make more profit longer term. People might use… will make more margin on some products than others, the less infrastructure heavy, for example. But I feel like I've rambled a little bit there.
[00:16:24] Andrew Michael: No, no, I'm leaving you go, because it's interesting just hearing your thought process on this as well. It definitely is a contradiction to a lot of the pricing and packaging advice you hear in terms of value-based pricing. And I think both can be true, though, at the same time that you can have a usage-based model with value-based metric according to it.
[00:16:43] James Hawkins: Oh, you could do for sure. Yeah.
[00:16:45] Andrew Michael: But I understand what you're saying as well is like you're really using from competitive landscape like you want to be super competitive on price. My only concern with a move like that is that then it becomes a zero sum game when everybody's pricing to be the cheapest and then there's this competition and then nobody, no other companies end up winning. I think it's good for customers, but potentially not for the companies themselves. How do you think about that as a notion that like somebody else comes and starts adopting the same process now as you, and then all of a sudden there's another PostHog competing.
[00:17:17] James Hawkins: Sure. The answer to this, I think, is the multi-product. So, yeah, if we're just doing product analytics and we compete on price, we're going to have an awfully slim... Like, we're going to make no money at all. Whereas the reality is we're like, cool, well, what will happen is we'll also be charged separately for all of our products. So we'll charge for Session Replay, Feature Flags, experimentation, warehouse, or whatever. And so we'll just end up making it up on cross-sell from other stuff, kind of over time. So we'd be unable to compete if we weren't multi-products on price. I think there's basically two ways. You can either be multi product or the alternative is you have some fundamental technology advantage, which means you can just go lower than everyone.
[00:17:48] Andrew Michael: Yeah.
[00:17:49] James Hawkins: There's a bit of that may be happening because we're kind of, we've come into a very crowded market using new software that like we have a database that's open source that we use on that can build our own because we're newer.
[00:17:57] Andrew Michael: Can learn from all other mistakes of the past as well.
[00:17:59] James Hawkins: Yeah. And then there are also structural advantages in the business too, like being in boundaries, you don't spend money on salespeople. So yeah, you need some, really solid other thing that means you're going to be able to afford to do that. And it probably only really makes sense in a crowded marketplace, perhaps rather than where you have a first mover kind of advantage. So yeah, there's quite a lot that goes into that.
[00:18:18] Andrew Michael: For sure. And I think like as well, having the multi-product side of things, like then you said, it comes a lot more closer to play like an AWS where you can come in and you can get all your different services from us, you pay as you use them. And then like when you're not using, you're not paying, but when it comes into churn and retention and like revenue retention itself, is this something that you focus on today?
[00:18:40] Andrew Michael: Like obviously then, as you mentioned earlier with usage patterns, I can vary quite drastically month to month. Are you seeing any sort of reliability in the metrics at this stage? Or do you think it's still a little bit too early in terms of like your revenue growth and there's a long way to go before things start to normalize?
[00:18:56] James Hawkins: No, at this point, it feels kind of consistent. I think because we have quite a, we have a very high volume of customers. We've got an awful lot of, like, we're not kind of like, okay, we have like, 50 big customers or something. We kind of have like, probably thousands paying. So we have an update, I think that it feels kind of fairly reliable kind of month over month, what's gonna happen. So we do, focus on it pretty hard. And like, you know, some of that is kind of this positioning, like we have more work to do in our positioning around like, hey, we're actually trying to decrease your pricing over time so that you don't even question looking around.
[00:19:29] James Hawkins: Or there's a lot more like, I mean, another really core principle, for example is we want our capital to go into our product, we don't want it to go into sales and marketing. So it's like, cool, if we keep doubling down on investing in the product, we think, will fundamentally will build a big, we think that will help retain better. And we have, kind of faith that if we made the product easier to use and solve the underlying problems, as opposed to, for example, getting really freaking good at customer support or something.
[00:19:54] Andrew Michael: Yeah.
[00:19:55] James Hawkins: And so we have, like engineers doing all of our support until we had 100,000 users. We've got one person now who is also an engineer, handling the first line of support for certain products that have hit maturity. But we still have all our engineers deeply involved in support because they were encouraging a pull request over a triage email to happen.
[00:20:14] Andrew Michael: Nice.
[00:20:15] James Hawkins: Yeah. There's a bunch of other stuff. Like our branding goes a long way too.
[00:20:18] Andrew Michael: And on that then, support itself, is that done through GitHub then, or do you have a separate support platform?
[00:20:25] James Hawkins: Yeah. We've got a very complicated journey on this. We started with everyone in Slack. So we had thousands and thousands of people in a big Slack group, which obviously just became chaos. In the early days, it was great because you're really close to people. Like people let their guard down in Slack, they'll use memes and stuff. And you kind of really get, you want kind of volume of communication when you're trying to get any kind of signal in the early days on who people are, what they care about, whatever. Whereas if, you know, Zendesk, there's not a lot of personality, like you're not going to get anything beyond, like some technical problem.
[00:20:55] James Hawkins: You want all this noise when you're trying to figure out who your ideal customer profile is. So it's good for generating that noise early on. Later stage, yeah, we've had to move everything into... We actually built our own support kind of forum from scratch and we embedded it into all our documentation and stuff. It's really cool. And then we triaged. So we, like, will enrich the user's data. So we kind of know, like what kind of customer they are and stuff.
[00:21:16] James Hawkins: So we, and I think we're about to ship like a paid plan where people can pay to have higher priority support and stuff. So. And we're encouraging the community to help each other out too, by having this all public on our website. So sometimes users will dive in and answer each other's questions. There's a really long way to go on that still. We have quite a carefully thought out system.
[00:21:37] Andrew Michael: I liked looking at the site as well. It felt like it was more, it was like blurred the lines between an app and a website going through it as well. And I think I probably did come across some conversations as well this morning. So I can see that it connects to the dots now as well, like where that was coming from and what was it for. So very interesting, like I think from the pricing and packaging perspective, then like trying to figure that out. Obviously then, like scaling support is very interesting from the perspective you're saying, really like you're focused investing in product, make the best product… possible products so we can reduce support needed on that end as well.
[00:22:12] Andrew Michael: Where do you see things going next as well? I think from a product positioning standpoint, because I think for me as well, this is the area where, like you have a very distinct personality, very distinct like, product persona that you're going after. What was the insight that led you to double down on a specific area? How did you come to where you are today? And was this something that you, I think you mentioned that you did start with the premise and the idea of who the product was for, but how has that changed over time, maybe from your initial premise to where you are today?
[00:22:42] James Hawkins: Yeah, this has been probably one of the most important things we've done. So we started off thinking, we were working with engineers primarily. But we weren't really sure how important, for example, product managers were to us. And so we're kind of on the fence, a bit like the self-hosting versus cloud thing. It's like, in the other days, we don't really know. You're trying to keep a foot in both doors at once. And I think we're about, I don't know, probably about a year and a half in or something. We had all of this, we started working on paid products. I've had so much variety in our user base.
[00:23:11] James Hawkins: So for example, we had like a chain of three family-owned pizza restaurants in Ukraine using us. And then we also had Fortune 500s using us at the same time. It's like, okay, who are we listening to? And then we have product managers, customer success people, engineers, all sorts of people. So we went, okay, we got a nudge from our board and I think it was very valuable. It was, we need to define our ideal customer profile. And as a founder, you're like, I intensely have a strong reaction to anything that sounds like bureaucracy. I'm like, well, it's engineers, isn't it? Basically. And he's like… And we're like, no, let's take this seriously. And we have a lot of noise. It's kind of hard to know who to listen to when we build features or fix bugs or whatever.
[00:23:49] James Hawkins: And so simply what we did was as we built out our pay product, we tracked every single deal that we were trying to sell, including importantly, the ones that just dragged on and didn't really close for ages. And then we had all these hypes. So we have kind of in this big spreadsheet, like I think 30, 40 deals, whatever it is that are kind of on at any time. And then across the top, we had all these hypotheses for who would particularly like working with us. And one was like… and they want to keep their data in their infrastructure. Another one would be like, it's an engineer making the buying decision. The third one was like, they've used our open source product before moving to our paid products.
[00:24:22] James Hawkins: And so we had this big, big list, they're very specific, kind of behavioral based things, not just like they're in the aerospace industry or something kind of vague. And then we scored the customer against all those attributes. We looked at which deals kind of sold and we also looked at the anti-personnel, like which deals just don't seem to kind of close or whatever and who seems less happy. It was very obvious. It's like… Yeah, if it's an engineer leading this stuff, it gets better implemented. It sells way faster. They're way happier. We have less work as well because engineers are particularly good at setting this stuff up.
[00:24:52] James Hawkins: And so, yeah, we're like, let's just throw caution to the wind. Like our conviction just went up a lot. We're like, yeah, okay, let's just like lean into that profile. We'll make our branding less formal. We'll kind of make, now we'll put like API requests and stuff on our homepage. We went from, like kind of cutesy, slightly inoffensive branding to really being ourselves. So we have like a Godzilla hedgehog on our homepage, like crushing all of the logos of all the other products that we're competing with at once. And so yeah, we just had some fun with it, basically, because we kind of found a more informal style will probably work.
[00:25:22] James Hawkins: And we also could realize that it's totally fine for us to go in a bottom up way. So we don't need to worry too much about the perception of a C level person. I had a call with, the other thing was I had a call with Matt Friedman, who used to run GitHub. I asked him, how does enterprise sales work at GitHub? I'm really curious. He's like, well, basically the enterprise go to market motion is the C-level don't like, no one really likes us in procurement at enterprise because what happens is 200 engineers just force them to buy GitHub. And we're like, I can totally feel that's what's happening with our products. So we'll just do more of that instead. And instead of trying to placate 10 different people at once, it's like, we'll just totally work on engineers, be really polarizing. And–
[00:26:02] Andrew Michael: Yeah.
[00:26:02] James Hawkins: Kind of see how it goes. Yeah, conviction went up a lot through this process of creating, ideal customer profile. It also affects, like how would your pricing, like you mentioned it isn't that confusing having multi product pricing? Yes. And we're like, well, engineers solve much more difficult problems than using a freaking pricing calculator. So I would expect them to figure it out because it's better value for them is the trade off. Because we're not taking any risk on like, if that usage goes up or down, we can price it closer to cost if we can better reflect the underlying cost of providing our products to them by having, like event-based pricing rather than monthly active user, for example.
[00:26:32] Andrew Michael: Yeah. Nice. You know, it's interesting. And then the other aspect I think of it, and it's similar to the GitHub notion, is that because you're targeting engineers, you're more likely to be the first product in as well when they're building the product or service from the beginning, as opposed to coming in later from like a market or a product manager when they, a lot of times, not even the first people to be hired within the company as well. It's normally like the founders, like hacking things together and being like the tool for engineers, like you're speaking to a lot of that early stage and then you can have that land and expand motion where net revenue retention then starts to grow and expand over time with your usage based pricing as well.
[00:27:07] James Hawkins: Yeah, you're spot on. Like we engineers exist and it also applies in startups. Like engineers exist before product managers do at least a good startup, probably.
[00:27:15] Andrew Michael: Yeah.
[00:27:15] James Hawkins: The, you know, our strategy, for example, the first one is, provide all the tools. The second is, get in first because it's simpler. We kind of realized through doing the ideal customer profile that they already had this big complicated modern data stack setup. It's just going to be really hard to get them to move over and frankly I'd rather sell 10 other deals in the same time for like half the price rather than trying to get one that's like, a bit bigger and far more work.
[00:27:37] Andrew Michael: More complex, longer deal cycles as well. Very interesting. It sounds fascinating like the journey that you've been on and obviously like there's still quite a long way to go. I think you can listen to plans and where you want to take the company as well. What would you say is the thing that's keeping you up at night today?
[00:27:54] James Hawkins: Well, I've got a 10-week-old baby. Oh, from him.
[00:27:57] Andrew Michael: Congrats.
[00:27:59] James Hawkins: We are trying to... So, something's kind of, longer-term product direction. By being multi-product, it gets pretty tempting to be like, we should build freaking everything because it's working quite well. And so, we've kind of built… There are different data-oriented tooling that companies use as they get larger. So you don't need a data warehouse from your two-person startup, but maybe use it with 30 or 100 people or something, depending on what… the nature of your product. And beyond the data warehouse, there's a bunch of other stuff. And we've been shipping and shipping and shipping.
[00:28:27] James Hawkins: And we built Session Replay products analytics that we use early on. We've built a data warehouse recently and all this other stuff. It's like, okay, do we just keep going to the right on that graph of maturity of company? Well, the other is, go across. And it's like, okay, should we actually now go from the three engineers at a startup to the one marketing person and the one sales rep, what do they need? And then we just build their tools out. And you can't pull that off if you're trying to also sell into enterprise at the same time. You get big order values when you replace like five or six products, even from a small company. So that's one is, like, we're not quite sure there.
[00:28:59] James Hawkins: So again, it's just something that's kind of up for debate. And we'll probably honestly just learn it by trying it out. Like we shipped a web analytics product recently, which is like a GA alternative that's in beta. So we're watching what happens with, like, non-engineers using it, like, can the engineer implement it and then the rest of the team pick it up? Like, I think Jira was wildly successful at that. Like, they went in through engineering and then engineers just give access to everyone else. So the branding and stuff still should remain quite oriented towards engineers.
[00:29:26] Andrew Michael: Yeah.
[00:29:26] James Hawkins: But yeah, that's kind of one. And the other thing is, perhaps a constant problem in PL, product-led growth. It's hyper efficient to grow like this, because you don't, kind of, need to do very much beyond shipping code to get more customers, like more customers bring more customers to us. But if you get concerned about scaling, like how do I make sure we quadruple our growth rate or triple or whatever, it feels harder for that to feel kind of guaranteed because we're kind of relying on softer stuff like brand and word of mouth. So thinking about how we grow faster is quite intellectually challenging. It's kind of, at the moment, it seems to be oriented, yeah, we're doing more with community, we're doing a lot more content marketing, we're just shipping more products as well.
[00:30:06] Andrew Michael: Yeah.
[00:30:06] James Hawkins: It's a little bit stressful, but I think, again, I think the answer to that I'm coming to is like basically just don't rely on venture capital because then you can just, then it doesn't really matter if you grow at like three X or two X or four X or whatever, as long as you just run efficiently and you're running kind of profitably or default alive. So that's been the other big topic that I've been thinking about the last few weeks.
[00:30:27] Andrew Michael: Yeah, definitely like the venture capital aspect. I saw that you also raised to series B. I think it was in today's market. I think things are changing rapidly. And I think, like, the venture market is also going to change rapidly as well over the next few years. I think just with, like the tools that we have available now with generative AI, like there's going to be a lot more startups popping up, a lot more people being able to build with a lot less resources needed to build those. So I think, like having the freedom and flexibility to grow on your own rates, I think is powerful. And then also I think like, what I love though with the open source components is that yes, you may have like a large product base. You almost have an extended team working and incentivized, like to pull and help build your product as well with you.
[00:31:10] Andrew Michael: So I think you've got this motion now that almost feels like it just keeps on getting stronger and moving faster as the company grows with you. So I think like there's the elements of all our content that you're doing, but I think the open source component is like a big driver and a big, like defensive mechanism that you… for that will stop others from coming in. If you can get to that sort of size and scale where you're building these tools going up and down the stack for engineers, and they may be, eventually like across the board because of like, ethos and the incentive and the focus on developers and on pricing. And it definitely sounds like you're onto winning formulas. Well, yeah.
[00:31:46] James Hawkins: So I think the thing that's kind of maybe struck me is like, I think we're onto a long-term winning formula, whereas it'd be really hard in our nature of how we're growing at the moment to like, we're not going to get from 10 to 100 million in like a year or something nuts, which other companies do achieve, like exceptionally good companies will achieve that kind of result sometimes.
[00:32:01] Andrew Michael: Yeah.
[00:32:02] James Hawkins: But we're just not well suited to it. And we should work with what is working, which is like, we should just lean into like, we should always ship more products so that our growth doesn't ever flatten out, whatever. We should always be hyper efficient so we can ship more products, so we can go into more and more markets instead of then spending, like all our capital on outbound and sales and stuff. And then just trust that we can build good retention because users are using us across a bunch of tools. We've done the right thing in terms of pricing and stuff. So just really strongly holding on to that belief and like leaning into it more and more as we get bigger rather than getting kind of boring and like value extracting, which is where the temptation lies, I think, especially if you're fundraising.
[00:32:36] Andrew Michael: Yeah.
[00:32:37] James Hawkins: Because you're like, I just want my revenue to be as big as possible. So it looks impressive so I can fundraise more. And so we're really trying to get out of that. So we've gone from like a very hype-y startup when we started with just open source, lots of adoption, lots of invested demand, we kind of flipped in 2021. Three things happened at once basically. My, with a lot of paid demand, we had a lot of investor demand to do another round. And my kid got cancer or our first kid. And so I looked after her and I was like, okay, she's not doing much better.
[00:33:03] James Hawkins: As we can only do one, I can't do these two things at once. I can't fundraise and figure out the business model and do all the sales. So just focused on revenue and sales and it just wound up the business being really healthy in the long term. So I think we just got lucky. It's the wrong word to use, but like it gave us a jumper thing on the market and just crapping out later on.
[00:33:25] Andrew Michael: Yeah. Well, man, I'm sorry to hear that, but I'm glad to hear that she's doing much better as well. I think that's also like, a… building a startup is hard, but then as well, like trying to balance things with family and then especially hearing that, I think that's like, incredibly difficult to manage and then work through as well. Like, how was that period for you as well? Like, has it something that you've recovered from as well mentally, because it cannot be easy at all.
[00:33:49] James Hawkins: Yeah, I think, again, like it's just like everything's relative, like, you're like, well, this kind of sounds like it sucks because it does suck compared to what the average is, which is to not get cancer. But once you've got cancer, you're like, man, there's like a lot of worse cancers than this cancer to have. So it's kind of relative. And I was like, well, I can work from home. I work in all remote company.
[00:34:07] Andrew Michael: Yeah.
[00:34:08] James Hawkins: Like I didn't build it all, remote for this particular reason, anticipating this would happen. But it meant that I could still work.
[00:34:12] Andrew Michael: Yeah.
[00:34:13] James Hawkins: And so yeah, work was just like, I'm like, well, I've got two things to do, like look after her and work hard. And that's pretty much all I've got. Like you can't even, it's a bit like when COVID was kind of a thing where you can't really go outside or do anything if you've got someone who's that sick. And so it's like, well, like, works kind of the thing you actually do when you–
[00:34:31] Andrew Michael: Have nothing else that you can do.
[00:34:32] James Hawkins: That's kind of nothing else at that point. So yeah, it kind of, I was just very lucky that I was able to keep working. I think it stopped me from going stir crazy.
[00:34:41] Andrew Michael: Yeah, no, for sure. I could see that as well on the flip side of it's just sitting there thinking and waiting and cause you're pretty helpless as well for the most part, what you can do like touch wood, like I have a little son as well, but like even when they are, they do get sick, you feel like there's nothing you can really do for them besides like be there to give them love. But yeah, I couldn't imagine anything worse is also happy to hear things open on that side. The thing I'm interested in then before we wrap up today is what's one thing that you know today about churn and retention that you wish you knew when you got started with your career?
[00:35:14] James Hawkins: It's pretty basic, but the importance of it, like I think all the other companies have worked out, none of them grew as successful as what we're doing right now. And it basically is because I think a lot of businesses, it gets really easy to hyper-focus on all of your metrics. But at the end of the day, it's like, you just, like if you just do the right thing for customers and have faith, that will work and don't try and, like optimize for your revenue number all the time, ironically, you will make more money. Like probably, like ask Elon Musk, although it looks like he got screwed recently.
[00:35:42] James Hawkins: But it's like, you know, set some ridiculous mission or whatever and like, do some cool thing. You'll end up getting paid kind of proportionate to the kind of problems you're solving for people. I think, like I bootstrapped a company early on. And we're all focused on new business or whatever. And it's like, no, I should have looked after the existing base. But we just never did. And the second company, the same thing kind of happened more on market. And this is PostHog is the one company I've worked in where I've managed to learn from that.
[00:36:05] James Hawkins: So it's taken me like 10-ish years or whatever of working to kind of clock that, like actually like, look after customers properly, worry more about the existing customers than new customers. And you'll end up with more new customers anyway, and it'll be a lot easier and less stressful.
[00:36:19] Andrew Michael: Amazing. Well, yeah, it's been an absolute pleasure hearing about the story today and learning from the growth that you've seen at PostHog. Is any sort of final listeners that you want to leave there in, final listeners… Is there any final thoughts that you want to leave the listeners with before we wrap up today?
[00:36:35] James Hawkins: Yeah, I think the last one's just like being human. I think it goes a long way. Like, I think it's easy to look at this as a dry subject. Yeah. It's like, oh, thou shalt do X, Y, Z to improve thine retention or something. Is thine a word?
[00:36:47] Andrew Michael: I don’t know.
[00:36:47] James Hawkins: Yeah. Like, for example, like we've seen this in our branding and stuff. We're just ourselves online. You can see, if you go on our careers pages, for example, you'll see everyone that you would be working with, the actual issues they're working on, their goals for the quarter, how much money you'll get paid and everything. And so just being very transparent with like, we're both human beings at this company.
[00:37:06] James Hawkins: Like when we do our customer success team, like one of them has recently put in, literally he's put a pop-up with his face on it into the product if you are looking at our billing page a lot. And it's like, Hi, I'm Cameron, and he doesn't use any capital letters or anything. So it's kind of clear. This is not a super automated, like boring professional message. And it gets way higher. It'll get like a way higher response rate and a better kind of reception from users.
[00:37:28] James Hawkins: So kind of trying to make clear that you are, people behind the business and being very transparent and building trust, I think there's a long way. So people actually try and work with you as opposed to if you come across as some faceless corporation, no one's going to engage. They're just going to like, pull their credit card out of the product.
[00:37:41] Andrew Michael: Yeah.
[00:37:41] James Hawkins: So yeah, just, I think trying to be authentic and human even though you're a corporation or whatever, I think makes a very large difference to be able to actually engage with customers properly. And I think it's one thing we've done quite well that not many companies do.
[00:37:54] Andrew Michael: Yeah. It's surprising to me that not many companies do this because like they, all of a sudden everybody just forgets it. There's humans at the end of it. And like, when you think about positioning and marketing, like, are we selling to marketers and enterprise companies and let's write this boring copy and forget that on the weekends they have lives and outside of work, they do other things, like the only thing they care about is his data and stuff.
[00:38:14] Andrew Michael: And yeah, it was really refreshing, like going through the site and just sort of seeing the positioning, the messaging, like the way you talk about the product and the team. So I think you're definitely doing some amazing work. I'm super excited, like, to go through it all today. So thanks so much, James. It's been an absolute pleasure having you today and wish you best of luck going forward. Thanks for joining.
[00:38:32] James Hawkins: Thanks for having me, Andrew.
[00:38:33] Andrew Michael: For the listeners, we'll make sure to have everything in the show notes. Yeah, it's an absolute pleasure. Have a great one.
[00:38:46] Andrew Michael: And that's a wrap for the show today with me, Andrew Michael. I really hope you enjoyed it and you were able to pull out something valuable for your business. To keep up to date with Churn.FM and be notified about new episodes, blog posts and more, subscribe to our mailing list by visiting Churn.FM. Also don't forget to subscribe to our show on iTunes, Google Play or wherever you listen to your podcasts. If you have any feedback, good or bad, I would love to hear from you. And you can provide your blunt, direct feedback by sending it to Andrew@Churn.FM. Lastly, but most importantly, if you enjoyed this episode, please share it and leave a review as it really helps get the word out and grow the community. Thanks again for listening. See you again next week.
