Dylan Etkin - Tools That Satisfy Developers & Managers, & Setting Clear Expectations as a Leader

(Intro Narrator at 00:00:01) Hello, my friends. Today, Joel is talking to Dylan, the founder and CEO at Sleuth. And they discuss the challenges of creating developer tools that satisfy both developers and managers, lessons learned on setting clear expectations as a leader, and tips for facilitating strong two-way communication with your direct reports. All of this right here, right now, on the Modern CTO podcast.

(Joel Beasley at 00:00:29) This is the Modern CTO podcast.

(Dylan at 00:00:41) You know, my daughter is—she's in college right now. She just started in the pandemic for computer science. And what you've described, I've tried to tell her a million times. I'm like, just pick something that you're really passionate about and just write code for that. It doesn't have to be any good. You know, it doesn't even have to necessarily work, but having a real goal that you want to build a thing for, you know? You learn so much along the way, you know? And she's like, oh, yeah. You know? And she wants to build these video games and stuff, and she's done a little bit of that. But I think she's also like, well, I need to learn a lot first. I'm like, oh, man. You're just going to learn a lot teaching yourself and find out where your passions lie. You know? You'll be like, I don't ever want to do the design of the sprites, and that should be somebody else. You know? But for me, it's—I don't know—a thing that you love.

(Joel Beasley at 00:01:29) Did she get to watch you go out on your own and start this business?

(Dylan at 00:01:33) Absolutely. Yeah. In fact, actually, I keep—I'm trying to get her as an intern, but, you know, it's hard. They covet their summer. So last summer, she was like, oh, I don't know. It's a pandemic. You know, I'm just starting. So I don't think so. And this summer, I feel like I've got her dangling on the hook. You know? I'm like, we work in a pretty modern way. You know? We're doing some pretty cool dev stuff here. I've got some amazing developers on the team that are going to be willing to give you time and teach you some stuff. I'm like, not everybody gets an opportunity to work at a small startup as an intern.

(Joel Beasley at 00:02:06) But you guys don't do games. You've got to call up somebody at EA Sports.

(Dylan at 00:02:09) I know. You know?

(Joel Beasley at 00:02:10) You know what I'm saying?

(Dylan at 00:02:11) I know. It's super true. I know. And she's so good at that too.

(Joel Beasley at 00:02:15) Tell me a little bit about what you guys do.

(Dylan at 00:02:17) Yeah. Well, I mean, the reason for being Sleuth is meant to be the best way to accurately track and improve a team's Accelerate metrics. And so, at a very high level, that comes down to deployment tracking and understanding when you're making changes. But in order to do things like change failure rate and mean time to recovery, we need to understand, what is a failure? And for that, we have this idea of deploy verification. And so the idea is that, you know, you hook up a bunch of these best-of-breed observability tools like Datadog or New Relic or AWS CloudWatch. You choose the five to ten metrics that matter a lot to you and your team. The things that you want to say, hey, maybe they're even things that you would alert on, but before they get to that alerting threshold. Right? I'm a dev. I make a change. If I changed RDS CPU by 10%, did I mean to make that change? Or did I make a clumsy error that I didn't realize was going to have that unintended consequence? And so for us, we want to verify each deploy against these metrics, make it really, really super easy for developers to understand the key things that they've affected and let them know in almost zero time that, hey, this is the impact that you've had so they can turn around and make a fix really quickly rather than have that go into the backlog or have that become the new normal. But when you do that, obviously, we need to understand what is normal for your database. You know? Databases swing, you know? So do you normally sit at 25% and sometimes spike up to 35%? And then, you know, what is alarmingly outside that normal? You know? So in that case, 40% is alarming and something you'd want to know about. But we also don't want to yell at you if it's at 35%.

(Joel Beasley at 00:04:08) Yeah. I remember one of the first times I used New Relic, probably five, six, seven years ago. And it wanted me to enter in all the thresholds. I was like, shouldn't you be smart enough to—shouldn't you be? I'm sure they are now. Right? Because it's been years and years and years, but that was my first instinct and reaction. And I was like, you should just know this based off of past performance. You have all the performance data.

(Dylan at 00:04:29) Yeah. It turns out it's a really hard problem. And I think the thing that is annoying is that there's probably a hundred of us all solving it at the same time. It's like, ah, if only there was that library, you know, that did the thing. But it's just a hard problem.

(Joel Beasley at 00:04:42) How did you even identify this need for Sleuth?

(Dylan at 00:04:46) That's a great question. So I'll go into the origin story a tiny little bit, but, you know, I worked at Atlassian for years and years. So back in the day, worked on Jira. Either hold that against me or don't hold that against me. I don't know. It seems like the world is pretty split on that one. And then, around 2010, Atlassian bought Bitbucket. And, you know, this was about the time when GitHub was sort of becoming really large. I took on engineering for Bitbucket. And when it was really small and there was just a few of us, it was really easy to keep up. And, you know, we were pretty early adopters of frequent deploys. You know, it was a SaaS-based product, and we were probably deploying three or four times a week. But as the team grew and grew, you know, and we reached 2 million users, it started to be really hard to keep up with what was actually changing. You know, we were doing deploys and my higher-ups were asking, hey, is this new way of working working for you? What are you changing? And is it going well? And then we were generating more and more incidents. And it just struck me that I work at the Jira company. Right? We're doing very modern agile practices. I'm working on a premier source code hosting tool and, you know, code review tool. But I can't answer what we're putting out in each deploy because we're not looking at the world from the deployment view of things. And that's really kind of the key insight was, like, man, we treat the most important event as this sort of insignificant event. I mean, really, it's the only thing that matters. It's the culmination of all of this work that developers have been putting in, and it's where quality actually hits customer. Right? And there wasn't tooling that was built to sort of go deploy-first and to provide all the tools that you need in order to make that process really smooth and to understand end-to-end how it's going. And so, you know, over time, that just sort of became more and more apparent that the industry is moving this direction. You know, teams are looking to deploy more frequently. Most teams want to get to about a deploy a day, and other teams are shooting for at least once a week. But, you know, it's where our industry is moving, and there just needs to be tooling that's built to support that. And that's what Sleuth is for.

(Joel Beasley at 00:07:06) That's pretty smart because for bringing value to the customers, right, and understanding where your product's currently at, it's super important to know what's in each deploy. And, you know, I noticed myself, I installed the CircleCI app on my phone. I have a very small team. I have two developers. We build a small product. But I need to know when I go into this customer call, was that thing we talked about that was supposed to go out today, did that hit? So I go through my CircleCI logs to see if it, you know, succeeded and then passed and everything. Because I saw it working in development. I just need to know if it hit production yet. And, you know, our tooling isn't necessarily designed to be super readable in that regard. So, for example, we use Pivotal Tracker. So it's like the story's either delivered or not, you know? And so it's—I think it's pretty cool that you—you really opened my eyes. That's why I'm sort of stumbling around my words right now because you said something in such a calm way. You've probably said it a thousand times, but for me, it opened up a whole new world where you're right, we sort of treat deployment as the afterthought. All the work's done, deploy. And if you put deployment front and center, it just—it just opened my mind, man. I'm a little bit at a loss for words. Thank you for that.

(Dylan at 00:08:22) Yeah. No worries. I mean, and look, we're Circle users as well. I absolutely love that tool, but it's built for a different purpose. You know what I mean? When you're trying to figure out what changed in your last Circle deploy, they don't make your life that easy. You know? They don't go and link up the issues in Pivotal. Right? They don't sort of tell you exactly who was a part of this change. So they don't show you the pull request titles so that you can actually read the human-legible thing of, oh, yeah. This was this or that. And we'll take it a step further. If you're a team that's using feature flags, you could have launched a feature flag about a week ago, and then the question is, is it turned on or not? You know, that's part of change too. So, you really have this—in a modern stack, there's a lot of different ways that we deploy. I mean, somebody could have deployed a Terraform change, and that would have, you know, enabled some sort of feature as well. So there's a lot of different sources of change that will affect that end thing of, is this out there? Or is it? And then the question is also, is it healthy or not? You know? Did I make things better or did I make things worse?

(Joel Beasley at 00:09:27) Yeah. I saw that you had mentioned something. I was reading in your bio or description of the company that you have some sort of predictive concept? Just like I can see what will happen if I deploy or—are there some sort of? Tell me about the insights you generate.

(Dylan at 00:09:42) Yeah. Well, because we're hooked up to your source systems and because we work off of deployment events, we understand the difference between when code has hit the deployable branch and when it's actually been deployed. So, you know, because we can hook into GitHub and we get all these events that say, hey, this pull request was merged, we can show you—we know, okay, the last thing that you deployed was X. Your branch is ahead of X by a certain amount. So we can show you, hey, this is what is going to be deployed. And then we can kind of kick in and we link up basically everything. We link up, you know, your Jira or your Pivotal and, you know, to your build systems and to, you know, all different tooling to sort of build this deploy view to say, here are the issues that are going to ship in your next deploy. Here are, you know, the different things that are going to go out as part of that. So that's one thing that we do in terms of just previewing what's up and coming next. And, also, I think what becomes interesting is, most folks, as they start to scale, have a lot of different environments. So, you know, you have, obviously, your production environment. Maybe you even have multiple production environments, but then you have a bunch of pre-production environments. You know? And understanding how change is making its way through those different environments—that can be kind of complicated and interesting and useful. Right? Customers that we have that are at scale, one of the biggest things that they run into is issues around environment drift. And that keeps them from moving things forward. Right? And moving things forward at a faster pace. So, you know, if they basically have four regions of deploys and those things have drifted, that's where it can be a really painful area for them where they're like, we spend too much time making sure that these things are in sync and understanding if they are so that we can make the next change roll out and be confident that it's rolled out to every place. So, you know, previewing those changes and understanding how they're hitting the different environments is also something that can be quite powerful.

(Joel Beasley at 00:11:47) How are you getting customers? Are you, like, B2B strategy? Are you, let's give it free to the developer so they go and end up, you know, telling their managers that they need it? How are you approaching the market?

(Dylan at 00:11:58) That's a good question. I mean, I think if you'd asked us nine months ago versus ask us today, the answer is probably a little different, because we're a startup.

(Joel Beasley at 00:12:05) Share the learning. I want to know what were you doing nine months ago? What are you doing today? What did you learn along the way?

(Dylan at 00:12:11) Okay. Yeah. No. That's fair. You know, so we're targeting two different kind of personas. Right? The Accelerate metrics and stuff. I mean, that's super important at a team level. And it tends to be either your frontline manager or the person up above that who's really interested in that information because they want to keep the code flowing. They want to keep the team moving and everything being healthy. Right? But developers don't like to be watched. They don't like to be spied on. There's a trust that exists between developer and manager. And, you know, if you're at a healthy company, maybe there's not a lot of tension with that trust, but there's a lot of companies out there, unfortunately, that aren't super healthy. And so there tends to be a little bit of mistrust, I think, around metrics and tracking. And I think the way that you get around that is that you only surface things at the team level. And then also you provide really useful tools for developers. So, you know, a developer's going to use GitHub and pull requests because it's an amazing tool that makes their life better every day. Now if GitHub can give you information based on that usage pattern, that's great. And so this is a similar thing that we do with Sleuth is we need to be providing tools that are useful for developers, like, you know, Slack-based approvals or the ability to lock deploys or previewing, you know, the changes that you're making or Slack notifications that tell you when your code is shipped. These are things that make developers' lives easier, and they're happy to use. And then, through those sorts of events, we can surface the higher-level metrics that engineering leaders are interested in so that they can make sure that their team is flowing and sort of keep improving. And so there's a little bit of this dynamic of, who do you sell to? Right? And I think nine months ago, we were thinking, maybe we sell to the developers. Right? We were building out the tools to improve much more so than we were building out the sort of metric stuff. But, you know, in building a deployment tracker, we kind of built a premier Accelerate metrics investigation tool as well. And so we did find that a lot of our conversations were happening with engineering leaders, and they were just—at scale, they were having a harder time figuring out how to take, you know, a unified approach to how to surface these metrics for different teams within their organization.

(Dylan at 00:14:37) But then they're always really interested in that. How will my developers use this, and what can we do to improve these metrics as well? So now we're targeting more that engineering leader with the accelerate metrics. But again, they need to have the promise that developers are going to be able to use the tool as well.

(Joel Beasley at 00:14:56) How do I see the metrics? Like, where do you have... Can I see them on a website sample? I know we have a lot of technology leaders that listen to the show, right? And they're having these problems every day, right? Like, how do we communicate with our executive peers on what's going on across the engineering org? Even if you only have five, ten engineers all the way up to thousands, you have this need. It happens in your everyday life.

(Joel Beasley at 00:15:22) And so you have me on the edge of my seat. I'm interested. I want to see the metrics. Do you have this on your website?

(Dylan at 00:15:29) Yeah, absolutely. So, I mean, you know, one nice thing about developing a developer tool is we get to be the best customer of Sleuth. You know, we sort of say that all the time. I'm like, look, if we're not the very best customer of Sleuth, then we're doing something very, very wrong. And so what we do is we just sleuth on Sleuth, and then we just leave that as an open demo. And the rule is, like, try not to say anything overly profane in your commit messages and your pull requests, but, you know, we don't have anything to hide. So yeah, there's a link on the main website which shows you what we mean when we show these accelerate metrics, how we surface them, how you can drill in.

(Dylan at 00:16:09) Because at the end of the day, it's awesome to see the metric. It's much more awesome to understand what has gone into making that and understand when you see something that has a high change lead time and it's anomalous to the rest of your things, you really want to be able to drill in almost instantaneously and be like, oh, that makes sense. That was that spike that we started four months ago, and we knew that it was taking a long time. And so this is not something we should worry about. But maybe, you know, you see three or four and you say, you know what? This was a new person who started. They seem to be getting stuck in review a lot. Maybe in the next retro, we need to figure out, we probably haven't been clear about what are the architectural norms that we want people to conform to. Maybe this is a problem with my team and onboarding, you know? And then you can sort of try and address that.

(Joel Beasley at 00:16:59) I like it. I like it. I'm excited about it. Is there a trial, a demo, or do I just talk to your sales team? How do you take a next step?

(Dylan at 00:17:07) I mean, we came from... The three cofounders of Sleuth all came from Atlassian. So we're big fans of the self-service model. So you can absolutely just sign straight up. You know, there's a thirty-day trial. You can use all the things in your thirty-day period. But, you know, obviously we're willing to talk to anybody. So it's easy enough to just contact us and we'll jump on a call and help with onboarding and all of that sort of stuff as well. So we're catered to the bigger folks who definitely need to talk a lot more before we jump in and actually do a thing, and then we try and just make it as smooth as butter to onboard.

(Joel Beasley at 00:17:46) I love it. I love it. I got to talk to a few people over at Atlassian before, and it seemed like they had a really, really great culture. Did that sort of come over because you were all working at Atlassian?

(Dylan at 00:17:58) I mean, absolutely. Honestly, I was really fortunate. I was the twentieth employee at Atlassian when it was a little startup. And I will give Scott and Mike, the founders of Atlassian, all the credit in the world over and over and over again. It still boggles my mind how we kept the culture that we had at 20, and that pervaded all the way through up to... I feel like they're at 5,000 or something like that now. And, um, yeah. I mean, it makes such a huge difference, and I think it has left a huge stamp on myself and my two cofounders, you know, just in terms of the openness that we sort of approach. We're very open internally with financials and where we are at with sales and, you know, what's going on, and just, you know, the way that we try and work together, the approach that we take to working. I'd say a lot of these things are lessons that we kind of learned over time at Atlassian and are trying to emulate ourselves.

(Joel Beasley at 00:18:57) That's exciting. Yeah. Culture is such an important thing. Because, I mean, you get to do your craft and you get to do your work, but it definitely matters about the environment and the people around you and the relationships you're building while you're doing that work as well. Because relationships are everything, man. I got connected to you through John over at LaunchDarkly, right? And I have come to learn you guys are good friends. At first I thought it was just, you might know each other or the PR people might know each other, but then I heard that you and John are pretty close. Tell me about that.

(Dylan at 00:19:28) Yeah, absolutely. I mean, like you say, relationships are everything. You know, when I moved from Sydney at Atlassian over to San Francisco, basically he was the other engineering leader in San Francisco. And he was working on the marketplace team. And funny enough, I worked on his team for my first two months in San Francisco. They were very functional. I'm not very functional. So him and I were a little like, whatever. But, you know, then Bitbucket showed up and I was working on that team. And, um, yeah, we just... I mean, we spent a number of years in the same office sort of being the engineering leaders in that office. Obviously had similar problems, and just became very friendly there. And then when he jumped for LaunchDarkly, you know, we just kept up. And he has been amazing and such an amazing resource. And he's just a kind and generous person as well. He's been kind and generous with his experience and his time. You know, it was great for me leading up to taking the leap for a startup to just have watched him and LaunchDarkly go through that process, you know, and really getting an insight into the angst and the sort of anxiety that comes with building a company over time. You know? And he was always very open in sharing with that. And then, honestly, also one of the biggest Sleuth champions.

(Dylan at 00:20:47) So right out the gate, he kind of understood the vision. He understood why it was important. Very early on, LaunchDarkly was using Sleuth in a former incarnation. Went through a little phase where it got shut down and then brought back up to life, and then was immediately a customer again. And LaunchDarkly has been a huge, huge, I guess, champion for Sleuth usage. And, I mean, for us, there's a wonderful overlap too. We sell to the same audience. We kind of have the same high-level goal, right? Which is to help people deploy safer code faster. You know, and they're doing it through feature flagging and all of the complicated things around that. We're trying to sort of do it on the other end of that pipeline where we're saying, let's help with your deployment process and understanding the impact of your deploys. But, you know, one of the very first features that we built was an integration with LaunchDarkly. Because, like I'd said before, I can mess up the site just as easily by toggling a feature flag as I can by shipping code, right? And so if we're going to alert your developers to whether you've affected the health negatively or not, we kind of have to have that in there for that complete story.

(Joel Beasley at 00:21:59) That is so cool. You answered all my questions in that response.

(Dylan at 00:22:03) Sorry. I monologue. I have... I've been trying so hard since I've started this Sleuth journey to cut down my monologuing, you know, in sales calls, in sales demos, talking to investors. I just don't seem to be able to get past it. I just... I could be a great arch nemesis for somebody because I could just monologue, you know, and I feel like...

(Joel Beasley at 00:22:29) I love it. I love... Well, it's important to have a high capacity linguistically. To be able to speak well is important in leadership, right? It means you've got some neurons firing up there. And...

(Dylan at 00:22:42) Right.

(Joel Beasley at 00:22:42) For you to be able... I get it too, you know. That's why we have this show. I was like, let's make it an hour, and let's let... It's like a monologue competition. There's this one episode... Editing team could choose if they want to keep this comment in or not. There's this one episode where there was, I said the intro and then the guy talked for one hour straight.

(Dylan at 00:23:07) Whoa. Okay.

(Joel Beasley at 00:23:08) I know. It was hilarious.

(Dylan at 00:23:09) That's amazing.

(Joel Beasley at 00:23:10) The Slack... In the Slack, you know, we have a production channel, right, where all the audio engineers are, and they screenshot the voice spikes when it was in post because it was like Joel: none. And it was kind of fun. But, man, I love the monologue and that's what we're here for. Come on and share your... See, now I'm going to start asking you about leadership. Do you have any thoughts, favorite leadership books that you've read or any lessons that have been very impactful to you?

(Dylan at 00:23:39) I mean, I'm a big fan of Rands, you know, Michael Lopp. You know, basically every blog he writes, I always read that and go, yeah, that checks out. That seems right. You know, he just has a really thoughtful approach to leadership and inclusive. You know? And so I think that's been really influential. I mean...

(Joel Beasley at 00:24:02) How are you... Just hang out with him before?

(Dylan at 00:24:03) I haven't, actually. I did meet him once or twice. I think he talked at some Atlassian thing, and I got to say hi. I think I had looked at maybe working at Slack at some point when I think he was their engineering head or something, but I don't think I actually met him. Seems like a cool guy, though.

(Joel Beasley at 00:24:22) Yeah. No. He inspired me to upgrade my camera. So I did an interview with him, I think, I don't know, a couple months or a year ago, and he joined and he had this 4K crazy cinema camera and I was blown away. And I told my production team after that call, I was like, let's upgrade the camera and we have a much nicer camera now. But, um, yeah, he was just... You know, five minutes before, I had read Rands for a long time, right? I was very aware of it. If you're a developer and you're not, you should be, right. He writes about best practices across the board. You know, for years, and then I had seen Michael Lopp as an upcoming guest and I was like, well, that's cool. That's cool. And then right before I do my prep before I interview him, I was blown away. I was like, that's Rands.

(Dylan at 00:25:06) He's Rands. Yeah. Yeah. So that was really exciting. So I got really pumped up about it. And I just was like, here's the microphone. Say a bunch of good words. And he did.

(Dylan at 00:25:17) Yeah. No. He's good. He... I mean, he just has a lot of great topics and, um, yeah. I, you know, I think also for me, it's just I've been fortunate and worked with a lot of great leaders over time, and you kind of pick up some of the good habits and, you know, you, if you're observant, you sort of realize some of the bad ones that you shouldn't try and emulate. You know? I think for me, one of the biggest things has just over time, and this is just in life in general, right? It's like this realization that when you communicate to others, you really want to be able to communicate in a way that they're going to hear. You know? And that's not always easy. It requires understanding who it is that you're talking to, the way that you're sort of having that interaction, and being thoughtful enough to kind of, you know, say things in a way that maybe are a little less direct, but, again, are going to be heard, right? Because you want to be hearing the people that are working for you. Hopefully, you're hiring amazing people, and, you know, we've certainly been lucky and have done so. And so there's just generally going to be experts in the things that you're trying to do more so than yourself. But then you want to have that sort of free-flowing conversation. And I think, you know, you just want to make sure that you're hearing them and that you're sort of being heard and are able to have that freeform communication back and forth.

(Joel Beasley at 00:26:41) Yeah. I learned that a lot with my family. If, you know, especially my wife, right? There are times when I want to give feedback, but I'm like, if I say anything right now, this is just not the time. And then I'll just wait till later and it'll come up and then she'll be open and receptive to it. And, you know, there's when you want to give the insight and then there's when the person's willing to receive it. And those are... They never line up at the moment of issuance. At the moment of tension, they're almost never aligned, right?

(Dylan at 00:27:14) Yeah. I feel like my teenagers have also been some of my best, you know, they've helped coach me through this. You know, because you just say a thing and then they're like, you know, and you're like, oh, man. Okay. Yeah. The timing was terrible on that. Or maybe I was just... You know, I didn't say it in a way that could absorb it.

(Joel Beasley at 00:27:34) Yeah. Watching the kids grow up is, and I've said this a couple times, but it's like watching AI algorithms boot up and come online and start to gather data. It's fascinating.

(Dylan at 00:27:45) Yeah. No. For sure. And I mean, yeah, when they hit the teenage level, I mean, they're basically adults, you know, with the occasional, you know, they dive into this childish reaction thing and you go, how is that the case? You know? I did not expect, um, you know, that sort of behavior. But for the most part, you're just interacting with adults now. It's kind of great, actually.

(Joel Beasley at 00:28:07) Yeah. I'm excited, man. You know, we're getting ready to go on this ten-week trip around the United States, 7,000 miles. And I've got the four-year-old and a 2.5-year-old and then my wife. So it will be an adventure unto itself, right?

(Dylan at 00:28:25) Yeah. Oh, gosh. I remember those. That's exciting. Two years old is young. That's... They're going to challenge you on that trip a little bit, I think.

(Joel Beasley at 00:28:36) Well, they challenge me every day in the house. So...

(Dylan at 00:28:40) Oh, man. Yeah. You know, this pandemic, I tell you, the folks... I mean, obviously, you've just interacted with a lot of people, but the folks with the young kids, I'm like, you guys deserve an extra medal of freedom from the president or something like that. Just doing your day job and having the kid just run in and out of frame and hang on the side of you and stuff. I'm like, yeah. My teenagers, you know, I'm like, I'll be out of the room at six. It's all good, you know? Easy.

(Joel Beasley at 00:29:06) Oh, yeah. Yep. And my wife, she's on the sales team here. And my daughter will make guest appearances. Because some days, the kids just don't want to go to school.

(Joel Beasley at 00:29:15) She'll just break down and she just doesn't want to go to her daycare school. And so sometimes she makes guest appearances on sales calls.

(Dylan at 00:29:24) Fair. People are pretty responsive. She's good at the close, right?

(Joel Beasley at 00:29:31) Yes. I will start teaching her closing lines. These are selling lines. These are closing lines. Here's how you go back and forth.

(Joel Beasley at 00:29:36) How much have you had to learn about sales? Like, you know, my background, seventeen years software engineering, then started really getting into, like, it's a very different mindset entrepreneurial from just building large scale applications for people, like four or five people that you know, like an investor world. And then actually operating a sales org because that's like an entirely different beast. How did you learn those skills to sell?

(Dylan at 00:30:01) Yeah. I mean, gosh. I'd like to think that I have. You know, I'll give credit to one of my co-founders. You know, I think when you're starting a startup, finding the right co-founders is like kind of everything, you know? Like, yeah, it's great to have a good idea and it's great to find something that has market. You know, but maybe if you've got a clever enough team, you're going to make all of that go. But, like, you know, finding these people that are basically, you know, almost as close as like your significant other, you really got to love working with these people, know that you work well with them. And so, you know, I was lucky that my co-founders, like, I've known them for, you know, almost fifteen years. And we had worked together at Atlassian and we just have very varying skill sets.

(Dylan at 00:30:45) So my COO is kind of running sales, and, like, he's done the most to just really learn how to sell. And, you know, we're fortunate in the sense that we have really fairly extensive networks. And so we have, like, great advisors. Like John actually is, like, you know, one of our sort of engineering-oriented advisors. We have a great marketing advisor, and we have a great sales advisor as well who has helped us with sales training and just, like, understanding the sort of the art of the sale.

(Dylan at 00:31:13) And so I tend to play the SE role, the sales engineer role in a lot of our calls. But yeah. Like, I mean, it's one of those things I think you mostly learn by doing. And then you just, like, read voraciously and, like, you know, do some sales training. And again, like, I guess our advisors honestly have been the most helpful. We did some, like, war gaming with, like, our sales advisor. And I was like, I'm going to do great. Let me just demo this thing up. You know? Like, here we go. And he just tore me apart, you know? And I was like, everything he said, I'm like, oh, yeah. No. I can totally see how that was. Yeah. I could have probably done that differently.

(Dylan at 00:31:50) So, you know, those sorts of things. And then just repetition. You know? You start to, like, look at the faces in the sales calls, and you can see when they're getting, like, the blank stare or where you're getting the, like, oh, oh my gosh. Like, you know, like, yeah, that's it. And you're like, I'm going to do that again. Like, those words worked really well, you know?

(Joel Beasley at 00:32:10) Yes. You sound coachable. That's, like, super important, right?

(Dylan at 00:32:14) Right. But I think that's a—I would hope any leaders that I've worked with in the past have, that's been a huge attribute is coachability, right? You just want to know, like, are you confident in the things that you should be confident in? And are you coachable in the things that, like, you also should be? You know? I recognize coming from an engineering background that I don't know anything about sales, you know?

(Joel Beasley at 00:32:38) I mean, now you know some. Yeah. That's good, though, that you were coachable and now you've learned, and now the business is growing, right? You guys are gaining traction in the marketplace. Tell me about that.

(Dylan at 00:32:48) Yeah. I mean, it's exciting. You know, it's good fun. I, yeah. I mean, it's, it's interesting to see, like, you know, where we grow. And, I think, you know, that's another thing that you learn is, like, it's a long process to get in at, like, you know, one of these very, very large names. You know, and they just have, like, so many, you know, things internally that are very unique to them. You know, but then, like, you kind of find, like, your sweet spot. You know, maybe it's like, you know, like, companies of, like, 500 or, you know, like, 300 or that sort of thing. And, I mean, as the customer base grows, it's just a wonderful avenue for learning with the product. You know?

(Dylan at 00:33:29) Like, I think that's one of the things that I've found most interesting about the growth phase is that you start with this kind of vision where you're building the thing that you think would be great. You know? And then at some point, you realize, like, you're kind of just building, like, things that, like, you know, some people did find certain parts of it great. And now you're just, like, focusing a lot on those. You know, where you're just getting very real world feedback about how to, like, make that, like, even more potent. And so I think that can be, like, a really exciting part of growth. You know? It's like almost making the product honest, right?

(Joel Beasley at 00:34:02) Do you find a specific—and by the way, I love your responses to this because, you know, my company is small, but we are going through these things, right? And so, like, yep. As you're talking, I'm like, yep. I remember we did that, and I'm waiting to get to the thing that I'm like, I didn't know that. That's why I'm excited, but I'm curious. We found, um, you mentioned, like, the sweet spot of the market, and that was actually something because when we started deploying, you know, marketing and ads, we were kind of like all over the board. We had a couple guesses of different personas that were loosely drawn that might be interested in what we're doing or have the problem, right?

(Joel Beasley at 00:34:36) And we started hitting them. And then what we found out is—and we sell leadership training. It's like leadership training for technologists. And what we found is that people don't—like, if they have under a hundred people, it's like a luxury something they might do because they really like leadership training. But once they start hitting, like, a hundred, a hundred twenty-five people, their employees start asking for it. And so we're like, oh, sweet. So we just narrowed all of our marketing down to, like, you know, 100, like, starting at, like, a hundred to three hundred or a hundred to five hundred. And then we noticed, like, by the time they get to, like, four or five hundred, they definitely have, like, some employee at the company who's got either fully dedicated or, like, at least 25% of their role is buying, selling, like, leadership development type tools.

(Joel Beasley at 00:35:22) So we were, like, let's get in there before that person gets in there, right? Like, let's be the product they know before they have someone come in and bring the other products they know, and then we'll grow that way. So we found that like a hundred to like four hundred-ish is the areas we target for marketing. And it's not to say that, like, other areas wouldn't work or that there's not other variations of our product. We, like, have this random, like, Fortune 50 client because they said we can use it for this, but I didn't have enough resources to, like, actually pursue that as a line of business. Because you can, you know, you're constrained to what you're able to do. So I'm like, thank you so much for buying. I'm so glad this is useful to you in that regard, but I can't drop everything because we have a little fire over here and we just need to stoke that fire and turn it into, like, crazy fire. Like we need to make a hundred times bigger.

(Dylan at 00:36:09) Yeah. No. I mean, it's—I think what you just said is really interesting and it's something that we run into too, which is, I mean, firstly, you know, I have always said this right from the start is, like, if you're a team like under ten people, you can kind of look each other in the eye and be like, did you deploy that thing? Yeah. I deployed that thing, right? You're like, cool. Done, right? Like, so there's not a lot of, like, confusion.

(Dylan at 00:36:33) We do have some folks that are sort of small like that, but they tend to be, like, super forward-looking, right? Like, they're, like, they're interested in being, like, best practice from the start because they have their eye on growth. And they're like, I want to put in a solid base so that we can accelerate this thing up to, like, you know, two hundred developers or a thousand developers or that sort of thing. But the larger the dev team gets, the, the more, like, it just becomes the wild west, right? And you run into these companies that have, like, you know, two thousand developers, and they're like, I don't know what's happening, right? Like, we've got a lot of change, and I don't know. Like, some teams are good at it. Some teams are bad at it. Like, we really have no idea. But, it starts to become an issue, like, I think, like, once you get past that two pizza team size. You know, so, like, that sort of, like, thirty to forty developers is when it really starts to, like, show up for us. And those are really easy teams for us to get in with because, like, they don't necessarily have SOC 2. They don't have, like, firewalled off their, like, source code. They don't have, like, you know, like, a team of lawyers that are, like, wait a second. Like, check this thing, check that thing. You know, so they can adopt and they can try, like, really, really easily. And then, honestly, like, these teams of, like, two thousand or whatever, they kind of needed this tool even more so.

(Dylan at 00:37:50) But then because of their size and because of that variation inside, it's probably like your Fortune 50 customer where you're just like, you're a lot of work. You know? And you're like, I'm not sure that, like, doing all that work just, like, right now makes the most sense. You know? And they have these, like, weird edge case things where they're like, you know, oh, we actually generate these artifacts and, you know, deploy them to, you know, Australia and whatever. And, like, it just has to be done that way. It's been done that way for ten years. And you're like, oh, okay. Why? And they're like, I don't know. Nobody knows why. Like, it's just how it is, you know? And they're like, but can you model it? And you're like, uh, I don't know.

(Joel Beasley at 00:38:35) Right after I went through that integration. So, like, we got the sale or, like, we thought we got the sale. Like, we're, like, we got it. And then it was, like, six plus months of back and forth of, like, to get into it. And then I realized after, I said, you know what? This makes so much sense now, why people charge so much more money to enterprises, why it costs more. Because I always would think about that. I'd always see them on the site. Like, why do we always have to call for a quote on the enterprise, right? Or why is—if they do show the enterprise pricing, it's significantly more and, like, for single sign-on. So I always thought—and then I realized after I went through it, I was like, I just spent essentially the entire amount of money I made just to walk through this process over six months. And so I would need to charge them double for me to, like, make money.

(Dylan at 00:39:20) That's exactly right. Yeah. I mean, it's just—I mean, I think it's a universal issue, right? Like and, this is another place where, like, you know, Codethem all was amazing for us because, you know, they're a good five years ahead of us. And, you know, like, they can just say, like, look. Those people right there, like, they're going to chew up all of your time. You know? And I think the thing that is, like, the weirdest, weirdest thing is, and I don't know if you've run into this, where you start to wonder, like, is this a fireable customer?

(Joel Beasley at 00:39:52) Oh, yeah. We've had to fire customers before. Yeah.

(Dylan at 00:39:55) Yeah. And that's, like, the weirdest—especially when it's, like, a very—like, it's a brand that, like, anybody on this podcast would totally know, right? And you're just like in your brain, you're like, no. That would be an amazing customer to have. We should super have them. And then, you know, you get like two or three meetings into it and you're thinking, I think it might not be worth having this customer. Like, I don't think right now is the time. You know? Like, they're a fireable customer, you know, for some reason or another. And, like, it's just my brain still hasn't gotten around how that works, you know? Because, like, the naive side of me is like, but you want that logo. Like, you should do anything in the world for that logo, you know? And then my more rational sort of CEO side is like, there's a lot of efficiency. I don't know.

(Joel Beasley at 00:40:43) For us, it was, we got really good at identifying them to stop them from even becoming a customer, right? We're like, alright, these are the attributes. And we looked back, you know, annually we review, or I think we actually do it quarterly. But we look at all of our customers. We figure out who our best customers are and what similar attributes they have and how can we just get more of those, right?

(Dylan at 00:41:02) Mm-hmm.

(Joel Beasley at 00:41:02) And in that, we noticed that the customers who pay the least are the most difficult. And so, like, alright. Well, there you go. And so let's—and then we figured out a couple other things like, just personality styles when going through our sales process. And so we essentially have these red flags that happen, right? And we just will tell them upfront, you know, that we don't think it's a good fit, you know, and that's just what we have to do. And the way we got the certainty to do that, the way we got the confidence to do that is because we did take the not great ones for a while. And every single time it was so much stress on the team and it was so much anxiety and so much time away from the customers who are awesome that, you know, it was hard for me to tell a salesperson who's, like, new and trying to build their ramp that, like, when this occurs, don't take it. And every once in a while, I let them take it. Because you kind of have to. You get that as a parent, right?

(Dylan at 00:41:57) Yeah.

(Joel Beasley at 00:41:57) Every once in a while, you kind of just kind of, like, let them go mess up.

(Dylan at 00:42:00) And I think there's, like, you know, there is always that little, like, hope where you're just like, maybe this one will be different. You know? But yeah. I mean, we're in a similar spot where I think we've—and I mean, you've seen the same thing as a startup. Like, you kind of mature your different functions as you need to, you know, like, just-in-time maturation of, like, all the different functions. And so, you know, our sales function has been maturing over the last, like, six to nine months, and we've gotten a lot better at—you know? And the same with our marketing function. And so we've gotten a lot better at, like, targeting the folks that do work for us, you know, and then, like, understanding, like, how to sell to those folks. And so similarly, like, our pipeline is much more full of the people that kind of fit for us, and that will be, a less of a hard, you know, sort of, like, process. But then, you know, I guess, fortunately, we still have some of these other ones that sort of show in via, like, outbound or, like, you know, just sort of, like, connections and that sort of stuff.

(Dylan at 00:42:55) Yeah. I still struggle a little with the whole idea of firing some of these. I'm like, maybe. Maybe. Just maybe.

(Joel Beasley at 00:43:03) Yeah. It's, um, everyone's got to—you just go on your journey, right? And you might learn something new along the way, and it's just, you know, being open and walking the process. You know, I've—so many people, like when I'm out doing public talks and whatnot, they'll ask me for, like, secrets or, like, what's the thing? And I'm like, there's really no replacement for just raw experience. Like, there are some things, you know, Elon Musk says starting a company is like eating glass and staring into the abyss. And I was like, that sounds about right. Yeah.

(Dylan at 00:43:34) It's so true. I remember when I was catching up with Edith Harbaugh, his journey with LaunchDarkly went very quickly for me because it wasn't my primary thing. And I remember meeting with him and him being like, product market fit is the hardest thing ever. It makes you want to basically die every day and whatever.

(Dylan at 00:43:56) And then we met again like four months later, and he was like, yeah, it's rough. But in my brain, I was just watching success go up this ramp. And I remember him saying something—I think he might have even said this on your podcast before—where he was like, if this thing didn't work out, I don't know what I'd do next because going through product market fit is really, really rough. And I remember thinking like, come on, it wasn't that rough for you guys.

(Dylan at 00:44:22) But now I have a whole other sense of empathy. Right? Having gone through it myself, I'm like, yeah, that's an angsty period. It's rough. Even when you look at the curve and you zoom out and you're like, things are going great, you're just like, oh god. Being in the curve, it's a different kind of feeling.

(Joel Beasley at 00:44:43) Yeah. It's tough. So you shared a few communication lessons. You mentioned that there is a couple behaviors that you want to avoid. Do you want to talk any more about that or share any more good leadership insight? Because a lot of the people here that are listening, they're tech leaders. Some of them are first-time leaders, and they're looking up to you, Dylan. They're like, teach us how to be a great leader. Share us some insight.

(Dylan at 00:45:10) Yeah. I suppose the things that—there's some north stars for me. I always feel like the idea is that I'm hiring amazing people, so I need to trust them. I need to empower those people. Give ownership. I forgot what—there's some book that talked about giving away your Legos or giving away your toys. I'm a giant fan of giving away toys. And then just being really cognizant of the things that you need as an individual to feel comfortable about doing that. Right? So like, what kind of reporting do you need at what sort of cadence? And how does that interaction work? How do you trust and then have those moments of verification? And I think that's kind of the only way you're going to scale. But it does take a lot to give away some of these toys, especially in a high-growth startup.

(Dylan at 00:46:06) Like, one day you are doing the marketing function and then you're giving away the marketing function. But again, I guess if you're hiring the right people, it feels amazing. Right? You're just like, oh my god, I was doing such a terrible job of that. And now there's a professional doing it. This is wonderful. And as long as I'm getting the information that I kind of need to feel comfortable, suddenly I feel like I have a fifth arm or something. Or wait, how many arms? Two arms? Third arm? Fifth arm.

(Joel Beasley at 00:46:36) You're an alien, Dylan. Yeah.

(Dylan at 00:46:39) There's a lot of functions. You've got to give away a lot of toys. You get extra arms all the time. You're just adding them to your person.

(Joel Beasley at 00:46:47) I get what you mean, man, because I've handed it off unsuccessfully and handed it off successfully. And when you get that magic person—as I call them magic people—when you get that magic person, I'm always on edge when I make the hire, like for how they're going to perform in the first couple days because I have just found through hiring, adding and removing fifty, sixty-plus people to teams. And in the past couple years, there's two or three types of people. There's the type of person that's just going to crush it. They're going to be the fifth arm, as you said, which I love.

(Dylan at 00:47:21) That's the title of your book, The Fifth Arm.

(Joel Beasley at 00:47:21) Yeah. They're going to be the fifth arm. Then there's the person who is going to turn around, be hired, and ask, alright, what do I do now?

(Dylan at 00:47:32) What? That's the worst.

(Joel Beasley at 00:47:32) Oh, that's the worst.

(Dylan at 00:47:34) I know, right? And you're like, uh-oh. Yeah. That is the worst.

(Dylan at 00:47:40) I mean, I guess the other thing around leadership as well is just set expectations. You know what I mean? And be clear about it. One thing I absolutely despise is when somebody has expectations, they aren't clear about those expectations, they also haven't made space for the individual to be successful in and amongst those expectations, and then they're upset. And I'm like, well, whose fault is that? Is that the individual's fault, or is that yours? I feel like it's very clearly yours. Set your expectations and then be cognizant that you need to give the individual space in order to achieve those things. Don't load them up with 1,500 things and then be mad that they didn't get to the one that you thought was important. Be clear.

(Joel Beasley at 00:48:28) Yes.

(Dylan at 00:48:29) So, you know, that guy that you hired that needs to know what to do, as long as you told them, this is a job where I'm going to tell you what I need you to do, and here's the list in priority order, maybe they're an okay hire. But at the same time, it's lovely when you get the person who's like, I've got this. Stand back. It's cool.

(Joel Beasley at 00:48:48) Yeah. The third person is the person who says I've got this, but they don't have it.

(Dylan at 00:48:52) Oh, and then they don't. Oh, that's—well, that's the verification part, right? Where you're just like, hey, that's cool. You got this. Right on. Just make sure I know this, this, and this at this sort of cadence. And then you're like, I need to remind you about this, this, and this because we've hit that cadence, and I am not seeing any of this.

(Joel Beasley at 00:49:14) Thank you so much. That's the thing. I'm like, you're supposed to be better than me. That's the prerequisite.

(Dylan at 00:49:18) Yep.

(Joel Beasley at 00:49:18) Dylan, how do we learn more? I want to learn more about Sleuth. What do we do?

(Dylan at 00:49:25) You just come to our website and you ask for a demo, and we'll get back to you in a couple of hours, and we'll absolutely jump on a call with you. And I'm sure we can make anybody successful with Sleuth and help your teams get better at Accelerate.

(Joel Beasley at 00:49:39) Thank you so much for listening. And if you found this episode useful, please share it with a friend or colleague who you think would get value from it. And if you have topics that you'd like to hear discussed on the podcast, either add me on LinkedIn or send me an email, [email protected]. Every time I get an email or LinkedIn message, it absolutely makes my day and inspires me to keep going.