Esben (2:56)

Awesome. And what was your origin story? Because I know you and Esbet, you both were in the same accelerator as well, which is pretty rare and fun to kind of discuss that too. So curious to hear the early days. What kind of sparked this like, we need to feature management.

Edith (3:09)

Yeah, So I had. I'd been an engineering manager at content management companies like Vignette, and then I'd been a product director at TripIt, which was the number one travel app at the time. So at TripIt we had this issue where we had to not only do constant mobile releases in the app Store, we also had to do constant backend releases because what we did was parse out airline hotel emails and then update them into people's mobile itineraries. So we had an internal framework for feature flagging where we could, you know, quickly run a beta test or push it to a certain amount of our users, then turn it off if something went wrong. And I had this epiphany that I was like, why doesn't everybody have a system like, like we do. And. And by the way, TripIt was the number one travel app at the time. But I knew that bigger players like Facebook and Google had much more sophisticated systems than us. Facebook had something called Gatekeeper, Dropbox had something called Gandalf for you shall not pass. If you're nerd like me, you'll get that joke. So, but it was just something that was a little out of reach for everybody else just because they didn't have the ability to or time to build it. Or if they did have the ability, it was something that was poorly maintained just because it wasn't their core focus.

Esben (4:31)

And so one thing that I were talking a bit about here too is like, you've definitely done a fantastic job just like creating the feature management category itself. So what does it actually take to define a category? Because that is extremely hard to pull up.

Edith (4:47)

Well, I made it sound super obvious that everybody would want and need this. It was actually incredibly hard. In 2014 I realized in hindsight that, you know, I was at tripit, which was a San Francisco company and, and had embraced the cloud, had embraced agile, had embraced rapid releases. And in 2014, many companies were still trying to wrap their head around moving from a year release to a quarter release. So me spouting off, well, this will help you if you're releasing in daily. People are like, I'm not releasing daily, not even something that I want to do, quite honestly. So a lot of it was meeting people where they were. So I used to ask a qualifying question of, you know, do you currently use feature management to you currently have a system like this, how often do you currently release? How often do you currently want to? And that really made people open up about their own release practices. And I think it was really important that we came off as a trusted advisor and never lecturing people, you know, because if you start to lecture people like, what's the matter with you? Why don't you release once a day? They're just like, who is this woman from San Francisco yelling at me? But if you say, hey, what's, what's your pain point with these releases? What, what's an ideal point for you to be at? Then they would really open up about the stress of these releases. You know, if you only release once every quarter or once every six months, every release has a lot of functionality packed into it that if something goes wrong, it's extremely painful, you know, so I'd hear stories from developers about a merge party, which wasn't a party, but just trying to reconcile all these branches that were being checked in. And here then the horror stories about a release gone bad where they would literally have like a 30 hour conference bridge where they were trying to cherry pick fixes in or out of how to manage what had happened. You know, that like something that's going wrong in the field, we don't know what it is. We're trying to figure this out, we're trying to hot fix it, we're trying to figure out if we could roll back and from that, from listening to their pain, I could try to help them understand that there was this better way of doing it. If you release more often and you're encapsulating each change, it's much easier than to understand what's happened and react very quickly. Along with that, it found that our best customers were often people who had had some sort of homegrown system. So one of our early, early customers was a company called InVision. And their co founder, architect was this guy named Ben Nemil. And he built it in the past. And you might say, well, he built it, why would he be a customer? And that was why he was a customer. He's like, well, I've built this. I know about how much it took me to do a basic version. Now I want to do more, but I don't want to have my own time maintaining this like that. That's not. My purpose in life is to be a feature, a feature management platform maintainer. I want to focus on the business value. So we got a lot of people who would look at their homegrown system and say, this isn't quite cutting it or this isn't getting us where we want to be. Let's, let's find a vendor. So classic build versus buy decision.

Wes (7:54)

One thing that's maybe interesting is you were selling to developers or at least development leaders. And we had an episode previously with Chris from netlify and they also established the whole category with Jamstack and so on. Right. And their approach was a combination of doing bottom up, kind of product led development outreach, but also fueling it with a community and a vision where they kind of partnered up with some other companies to kind of help drive this change. Right. And one of the things you mentioned was the whole more frequent releases, right. Which I think was probably a big reason for why we suddenly needed feature flags, right. Or feature management was that the world was going there. And a unique insight you kind of had. And did you also build, build, build some kind of partnerships or something to, to help drive that agenda with your customers?

Edith (8:51)

Yeah. It's interesting. I was smiling because I know Chris pretty well. He's been to my barbecues. We were in the same incubator called heavybit, which was basically all companies who were doing something around DevOps developer tools, continuous deployment. And the joke was that we all kind of had the same tagline, which was move faster with less risk. So we were all trying to basically address the same issues of developers. But I'll add in very different ways, like what Elfly does is not what we do. But it was a hundred percent what you just said of the change from resistance to more frequent releases. Like, I literally remember a conversation I had with a prospect in, in 2015 when I asked him, how often do you want to release? And you said exactly as often as I do right now, which was twice a year. And I'm like, are you sure? He's like, yes. And I'm Like I'm not gonna try to, I have nothing to sell you. Like if my, if I'm trying to convince you to move to more frequent releases, like I, I'm already fighting such an uphill battle. But if we found people who were already wanting to do that, it was much, much easier. Yeah.

Wes (10:00)

Did you help drive that change or was that change just happening?

Edith (10:03)

I think we tried to help drive it, but people had to believe. So you know, I went out and gave a talk at Continuous Life Cycle in London in, I think it was 2015 or 2016 and just didn't feel much pull. But then when I went to other regions, like I gave a talk at NDC Sydney, must have been 2015 and after it, two people. And it was, it was supposed to be an industry talk, so it was vendor neutral. So I said, I'm just going to talk about how you can move faster with feature management, but I'm not going to give you a demo. This is just about how you should be using feature management overall. And then afterwards two people came up and they said, we actually want a demo. So like in a, in a rickety, you know, stand. Not even a stand, like a little lobby table, I gave them a demo and they both became longtime customers. So I'd say it was a mix of a little bit going out and trying to make a market, but, but also trying to find people that are receptive to that message.

Esben (11:07)

Yeah, I'll double click on that because I think that was the overlap too with Chris from netlify too. It's like you gotta have that pull and if you're just pushing, pushing, pushing for a new category that it's gonna be really hard to work. And like I know even with product LED growth, like I felt that same pull too. Even just in terms of like, even when the first time I heard the word I was like, that is way better than like telling people I do kind of like UX for getting people to value faster and somehow that turns into more self serve revenue. I'm like, it's product LED growth, let's make it easy. And so there, there's a lot more people that should be pulling with you. And I think another thing too that I'll just add on here is it shouldn't just be one person that is pulling in that direction. You should start to see, okay, maybe you're the first one for the first year. But then over time there's other people. Like I remember the first time it was just me and Open View talking about plg. Then it was like appcues and Pendo and like everybody else started talking about it. I was like, ah, good, thank God, not just me. So that makes it way better. Now when it comes to like how you created a lot of this category as well, what role did your free motion have in that? Like, did you always have a free motion or free model for your business?

Edith (12:23)

So that's a good question. And I'll touch back on creating a movement for a second. So Martin, Martin Casito, who. Brilliant, brilliant person. He was CTO of a company that acquired, got acquired for by VMware for a lot of money and then is now a VC at Andreessen Horowitz. He's like, Edith, you're not selling a tool, you're selling a movement. And then you're selling the instruction book about how to be successful. And that stuck with me in terms of I can't just be selling a tool. I have to be selling, you know, a better way to build software and that we can help. So your question was the free. It's, it's interesting because I'd come from tripit. So tripit, those who don't know, is a freemium app where at the time it had a base of this is in 2014 about I think 10 million free users of which some percentage of them monetized into the pro version. And that was how we made money. So originally we thought that LaunchDarkly would be similar, that we would have this huge free tier and that we would monetize some proportion into pain. So when we originally started, our first customers were literally me trying to beg everybody in the accelerator to use it. I remember sitting down with Espen's co founder Christian and he gave me a very polite no. But I did get another friend from the accelerator to use it. But what happened when I tried to get outside of my friends was that people actually really balked at our self serve like so we had built this flow where you could sign up, you could trial, you could use a credit card. I still remember our very first customer that was not a friend of mine, which in and of itself felt like a huge win. He trialed the product, his tech team had looked at it, they were ready to buy. Is like, okay, have your procurement person talk to my procurement person. And at the time we were a four person company. So I was like, my procurement person is me. And I told him like, I don't, I don't want to talk to procurement. Like please just sign up on the website. And he's like, no, that's not how we buy. And I'm like. And so we kind of had a back and forth for a couple of exchanges. I'm like, it's a really nice price. It's like, I think it was like 79 at the time of month. I'm like, just please, please put a credit card down. And he's like, no, we're just not equipped to buy that way. I vividly remember it cause I was running a 50 mile race down near San Diego. So I had a lot of literal time on my feet to think. And the first part of the run was thinking about how I could persuade him to put down a credit card. And the second part was this realization that despite all my fears that we were going to become an enterprise sales company. So I had worked at enterprise sales companies before when I'd been at Vignette. I knew the motion and I was like, oh, we are getting all this pull for us to be an enterprise sales company. I can either keep resisting this or I can admit it. So I got back from my run, I talked to the team and they're like, okay. It was a small team at the time. So our daily standups, you know, all four of us just stood up and talked about what was happening in the world of software. And I called our contact at the customer and I said, yes, we could do this through procurement, but you realize it's going to be about double the price just for the hassle of doing this. And he's like, that's fine. And then I thought, darn it, I should have tripled the price, which was helping the price. So I sold it to them for at the time I really wanted to call it a five figure deal. So I think it was like $10,800. And we worked through procurement. We had to get a lawyer to help us with the red lines. We did a security review and that was our first, I'd say legitimate sale. That they. The two nice things about it were if you do an enterprise sale, you do not have to pay stripe fees. So this was actually a little side benefit for us for a little bit. Because in all my projections about money I'd built in us paying some percentage of stripe versus if you collect from customers, you don't pay that fee. The second kind of fun thing, and this was just us getting our process down was the customer mailed us a check and at the time we banked at svb, Silicon Valley bank. And our accelerator was still having some meetings at svb. So I'm like, I'll just take it in for the next meeting. I'll you know, deposit it. Easy peasy. I walked into SVB with the check and they're like oh, we're not that kind of bank. I'm like, but I have this check. And they're like we're not that kind of bank. So eventually it turned out that you had to enable mobile deposit and take a picture of that, that, that, that was our very first check ever. So that was a long winded way of saying did you always have a free tier? And the answer is yes, absolutely, we still have a free tier.

Esben (16:57)

So, so you knew. Okay, great. We, we are going to have to go down this enterprise sales train. But you've had a free motion this whole time, so why like not just kill it, get rid of it? Because so many companies, they, they don't continue keeping it for, for a very long time. They're like oh, let's just like double down on the sales LED piece. Let's, it's cannibalizing our demos. As soon as we take away that free trial or free version it's like we will get a lot more people signing up for it. And I did sign up for it in preparation for this ads like you are. I actually thought like your sign up page, even the microcopy for the reverse trial, I was like this is so good. Like you actually put thought into this. There is a lot of sales like companies that have an awful, awful prevention. I was like, I didn't sniff that out. I felt like it was well thought out. There was of course like anything, there's things that could be optimized to get people to value faster. But I was just like, you know, it's, it's not, it was what I was expecting. It's, it's really quite solid. So yeah, it maybe expands a bit more on like why did you decide to keep it?

Edith (18:05)

Oh gosh, couple of reasons. So like I said, my, my background was I'd been at tripit and literally my job was to try to get people to sign up for our paid motion. That's what we did in the early days. Spent a good amount of time trying to tinker with getting it better. So we, we've kept it one because we like it and I, we've kept it two because we sell to developers and developers often are somewhat allergic to salespeople unless they need them. So a really familiar pattern to us is a developer will sign up with an account that we know is a made up name, Kick the tires get happy enough and then want to involve a salesperson. So I think there was this false dichotomy, I'd say in the 2016-19 era of your either product led or sales led. And you have to pick one. You know, it's either like you have to talk to a salesperson or there are no salespeople. When you think about an enterprise, it's not really how they want to buy at all. Like, that's kind of saying that unless you decide to rent an apartment, you can't even enter the door like that. You're going to get to look at some stale screenshots of what the apartment interior looks like in front that decide to sign a lease when usually if it's a enterprise buyer, they do want a developer to evaluate it and then they really want to talk about the actual terms after they have some base level of like, okay, we understand how the basics work. Can you have an SE come in and really explain how it will work with our architecture? Can you have our security expert talk to your team and get some confidence? Can we talk about what this actually means in terms of guarantee, guaranteeing levels of support or what our actual price might be given? You know, we might be a very large enterprise and want to talk about minimums, overages, all that sort of stuff that you sometimes want an actual person. So that, that's, that's why we still have the free layer is we see it actually as part of our sale. And to cut it off, I think would cut off a large amount of people that do want to experience it without talking to a salesperson. But then you want to talk to a salesperson. And thank you, by the way, Wes, that's very nice for you to say that our flow is good.

Esben (20:12)

Definitely there, there's a lot of thought on that. And I was like, oh, this reverse trial example of on the signup page, I think I was like, oh yeah, screenshot. That, that's, that's good. I guess one of the next things I would be curious is you recently came back as CEO. So two part kind of question. One, like, what gave you the. I guess like, hey, now's the right time to kind of step away as CEO. And then what was the kind of main reason you decided, hey, like now's the time to come back? Like, I know right now, I could take a guess. But like AI disruption, it's massive. Like every. What we're seeing a lot of companies like their founders come back because it is as far as like you imagine your business is A car, it's like it's a two hands on the wheel kind of time to be alive. It's exciting, but you also have to be able to maneuver and adapt very quickly. And that is hard with just hired CEO for sure. So love to hear your take on that.

Edith (21:04)

Yeah. So I started the company In 2014, around 20, 22, 23, we crossed 100 million of ARR, which I thought a huge achievement at the time. We were one of the fastest companies to do that. Caveat with at the time because now it seems like some people cross a hundred million faster than it takes me across the street.

Esben (21:24)

I know it's like 12 months or less.

Edith (21:27)

So back then, and it feels weird to say back then about something. Three years ago, I felt we were growing at this nice steady pace along with the market. Like every year more people were moving to the cloud. Every year more people were saying, hey, I need more frequent releases. But it was a very. It's a predictable motion. So we hired an external CEO and what changed was AI is just so exciting. It's so fun. Like I said, I was making a joke, but it was somewhat true about just the rate of change now is incredible. The way that used to be a long discussion trying to convince somebody of maybe you should move from six month releases to a month is suddenly the. The customers are doing themselves. The customers are saying there's a tsunami of code coming in and I need help. Like, it used to be that the, the barrier was always hands like keyboard. Like if you wanted to produce more code, you needed more literal hands like keyboard. That's not the barrier anymore. Now the issue is that you could produce code a hundred times faster than you could three years ago, but the rest of the tooling is not a hundred times faster. And that's really where I think launchdarkly could play. A key part is all the things I talked about at the beginning of the podcast about helping our customers launch measure and control. We still do that. So, like if you're pushing 10 times as many changes as you used to, we could give you a framework where you can put them to different populations, understand how they're performing, and roll them back, you know, very quickly. So we've evolved our message to being the runtime control for the AI era. And that's really resonating with our customers. And I think it's a really exciting place for us to be. The other thing that really excites me is just how it's fun. Like, so we're using AI pretty heavily internally and like literally. So I started as an engineering manager and I remember spending hours in bug scrubs, you know, where you're triaging what bugs go into what release. One of our engineering managers just pointed Devin at a backlog and said, make me the tickets and make me the PRs done like. So that just like streamlined so much of work and made it fun again. So instead of like sitting around talking about what bugs we're gonna put into what release and T shirt, sizing them and trying to agonize about who's gonna do what, it just gets done. So that's just fun.

Wes (24:00)

And work should be fun. I think that's definitely a big part for me too in how I decide what I want to work with. But I think one thing, we saw another big kind of CEO coming back some years ago, Euan from Intercom. And when he got back and the whole AI, he even got back before the AI boom and shifted the focus of the company a bit. But he shifted it even more when AI happened and he basically did a full transformation of the company. Now they're more trying to get known as fin with an agent. Do you think LaunchDarkly needs to go through a similar kind of transition or how do you look at that? Because you can say with tools like Claude code cursor and so on, they're definitely changing the industry of IDEs and so on.

Edith (24:46)

Right.

Wes (24:46)

So LaunchDarkly needs to do a similar kind of full 360 on the company.

Edith (24:51)

That's a great question. You know, I think we are constantly. So part of our own DNA is learn and grow. So that, that was one of our core values is learn and grow. You know, because in a world where you can release, like I said, not quarterly but monthly, but hourly, I always wanted the team to iterate very quickly. So now with AI, we're also looking at a lot of our own internal processes about how can we use AI effectively. And it will state, you know, we have large enterprises who trust us, so we have SOC compliance, gdpr and we need to be able to use all these tools in a safe and compliant way. We're not a four person startup with a dog who could just YOLO everything. We need to do it in a way that gives our customers confidence. But it is just fun. So we have a mascot called Toggle. Uh, it's a, it's an astronaut. And our design team made a gem in Gemini, which is basically a set of poses of the mascot. So that if I want to make a, you know, the mascot on A bike riding around our office in Lake Merritt takes me two minutes. And what used to be me describing the idea to a designer, then building it, me saying that's not what I want, and maybe it taking it like a half day back and forth, like, just do it. And I post. And the joke now is I've become the best one at using it because I could just do what I want. And I am, I'll admit, a terrible, terrible, terrible designer from scratch, but a very good prompter. So that was to answer the question about rethinking what we're doing. Yes, everything. Like, I'm always thinking about what is possible now that we have these tools. What could we be moving faster? What could we be doing safer? What could we be doing more effectively?

Wes (26:52)

So maybe a follow up question to that. So we also had. Wes had an interview with Jason Fried from Basecamp, and one of the things he says is the founder's most important job is to introduce Risk. So do you think that's why you needed to come back? Because you are the only one who can introduce Risk to the company because you have kind of built it from the ground up. How do you think about that?

Edith (27:14)

So I think my job as a founder is to make sure the team is having fun. And I don't mean fun in a, hey, let's go play ping pong for two hours. But in a, it's fun as an engineer, as a product manager, as a designer, to feel like you're producing something. And that ethos also applies, you know, our sales and marketing. You know, it's fun to go visit a customer and have them say, juan starkly changed my life for the better. You know that I don't. I got my nights and weekends back. Things are much smoother. So. So my job as a founder is to remind the team of how we change people's lives and. And also to encourage them to have fun in their own jobs. You know that these new tools, new way of working can reduce so much toil. Uh, you know, so I gave an example of the, you know, nobody really enjoys as a bug scrub, but if you can say, hey, with these new AI tools, we could triage much faster, our tickets will go much faster. We can focus on what's actually fun. That is fun. So that's what I see as my job. Encourage people to learn and grow and encourage them to enjoy their work.

Wes (28:21)

I think that's an awesome way to look at the role of a founder. I think also what's maybe another dimension to that? And let's bring back netlify. Matt this week has been launching new features by building apps every day. And I don't think you would see that from many CEOs of unicorn companies, right, going out building an app. But we are seeing that right now with some founders and I think that's also part of it for sure to be a startup within the startup or a startup within the business. Right. Like, basically this AI era is coming and you need to think a bit more startupy and like build again. Like find the joy of building new stuff.

Edith (29:01)

Right.

Wes (29:01)

And, and it sounds like you look at it in that same perspective with having fun about, we have a new technology here, let's see what we can do with it. Which I agree with. It's not like the Web 3.0 as we had in the past, right. This is actually a technology that a lot of people can benefit from and it's enjoyable to work with. So that's really cool.

Edith (29:24)

The example I think about a lot is, as I said before, I'm a terrible artist. I once, I think drew a chicken with four legs, but with a lot of the new tools out there, I could just express my thoughts and have the AI make me look good. So I was always an artist in my head. I just couldn't do it. So I think about a lot of the new AI tools are helping people be creative in a way that they always were. They just didn't have a way to actually do it.

Esben (29:54)

I love that. And to wrap up here as well, one of the reoccurring trends throughout this chat and I always like to pick out something unique that is so you. So the biggest thing here is having fun in your work, having fun with your team. So for everybody who's listening, who's like, hey, you know, this is a little different. Like, you know, we always talk about like build moats, you're going to attack the market and like all that, like war analogies and stuff like that you hear again and again. It actually gives you a, a kind of like a serious vibe. Like, oh, you're in business, you like hope to kill in a good way, like business wise, but not actually killing people. That'd be terrible. But for your perspective, what would be your advice to another founder if you're. They're in this room today just listening to you and what would be your advice? Like, here's how to have more fun. This is like the tldr. Here's what I got there.

Edith (30:49)

That's a great question. I, I think the reason why I have Fun is because I love software. You know, I started the company because I had been an engineer, I'd been a product manager, I managed support, I'd been in marketing and managed salespeople. And when I started LaunchDarkly, I wanted to make the world of software better, you know, and it feels still amazing to me when I do a customer panel and people say, Edith, you've made my developers lives better. So my advice for every founder is to make sure that you're doing something that you actually care about. Espen and I have, we were in the same accelerator and the founders who didn't do so well were the ones where they just didn't have any genuine empathy for their customers or their problem. You know, they're like, hey, let's do this idea. But I don't really care about it. It's, it's, it's super easy then to get burnt out if you don't actually care. But if you're actually passionate about what you're doing, you'll always find rewards. And those rewards are monetary, but also just satisfaction that you're, you're making the world better.

Esben (31:52)

My dad always reminded me when I was watching a car and I wasn't getting paid, he's like, a good job is its own reward. And yeah, I, I think there's a lot of truth to that too. Sometimes if you're really into it, you do like what you do. Sometimes, yes, there is the monetary side. Other times it's like, hey, like when I write a book, I don't actually care necessarily. Like this sells many different kinds of copies and stuff like that. It's like, hey, I actually enjoyed the process of it. I learned a lot throughout this, even if it, it doesn't help anybody else. But of course the goal is to help a lot of other people, so absolutely love that. Now, where can people find out more about you? Like in the show notes, we'll have launchdarkly, we'll link to it as well. But where do you hang out online the most where people can stalk you, ask you questions, and let you know what their takeaway was from this?

Edith (32:37)

You know, I don't have a great answer that I should get back to you on that one, actually. Sorry. I think like anybody else, I'm often on, I'm often on LinkedIn.

Esben (32:45)

Sure. Radiant default. Like dead cool. Well, thank you so much for, for coming on, Edith. This has been a blast. And you definitely reminded me as well to just have more fun at the end of the day and really lean into that because yeah, it's contagious too. And as the founder, you set that culture as well. Like other people follow your example. If you're not having fun, it's like they're not having fun. Yeah, it's super cool that you leave with that. So thank you, thank you, thank you. And to wrap things up, thank you everybody for listening to this version of the product led podcast. Make sure to rate review this on wherever you listen to podcasts, whether it's Apple, Google, you name it, Spotify. I'm going to read every single one of those reviews and that's how I know how to improve this. Also, if you want to stay in contact with Bean and learn what is going on in the world of PLG and every single week get the best actionable deep dives on product led growth. Make sure to head on over to productled.com forward/newsletter I am personally writing each of these deep dives every single week and you're going to get a ton of it. So make sure to head on over there to product led.com forward slash newsletter.