Essential lessons for building and scaling DevTools | Dennis Pilarinos (Unblocked, Apple, Amazon, Buddybuild, Microsoft)

Brett: One of the interesting things is that going all the way back to the start of AWS, Azure, etc.

You were kind of right there as this new generation of software got started. And then you participated kind of in this next generation, when Twilio was coming up, Stripe was coming up, you started to build, your last company. Could you sort of break apart those different chapters? What was required to be successful?

Have things shifted or kind of the same tenants of building great products for that audience, distributing great products for that audience, more or less the same?

Dennis: There's kind of probably three and a half shifts in my mind, in my career, BizTalk Server was an on premise, like it was a physical box that you bought with discs and then we, you know, started to shift CDs or whatever.

there was the on premise, then I think there was cloud. So when I think of Stripe and Twilio and things of that nature, I think that's kind of in that cloud genre, right? Which is like you call an API, people can use it to perform some function. and then mobile. was obviously like kind of the next shift for really a developer ecosystem.

And the business models that actually shifted along with them, So you went from like a site license, to like a SaaS based model, to kind of free to pay. So every time you have these kind of significant technology shifts, you tend to find that there is actually a shift in the business model that goes along with it.

I think the one that we're the half that we're looking at now is the kind of the advent of this AI LLM generative AI technology.

Brett: Do you think that sort of part of the way this part of the way through this new shift that the wind conditions for what will be the next stripe or Twilio or whatnot has changed dramatically? 

Dennis: I think just first principles that really don't matter like at the at the end of the day, like these are technologies that enable you to do something you weren't able to do before. When you're calling cloud hosted APIs, all of a sudden you don't have to ship on premise software. You have this incredible REPL loop, right, where you can ship it to a cloud service and it gets automatically distributed to everyone who wants to use it.

Right, you're not sending physical disks out as one of things that things to change, right? Mobile's the idea of it being a rich interface that goes everywhere that a person is and there's obviously like Hardware functions that provide utility for that platform. But typically at the end of the day what you're able to do, these technologies allow you to do something you weren't able to do in the past. And so that's where I think you have that kind of these transformative, changes.

I think if you can apply those technologies in a way that actually provides unique utility, that's the first principle that, really matters. To what degree is technology unlock something you weren't able to do before.

Brett: When you think about building products where the primary consumer of those products or user of the product is, a developer, do do you think that that's different than building other types of products?

Dennis: For sure. Yeah. I mean, it's a little bit hard for me to say definitively because I've only ever built developer products, right? So I can only speculate what it's like to build consumer products or products for marketers and things of that nature. Um, one of the things that developers are uniquely, intrinsic is like, they've had very, very strong opinions, right?

And so like, this is a community that to this day argues between tabs and spaces. You have a very small window by which you can make a very good impression. And if you don't deliver that, that bit is effectively flipped. It's really, really hard to come back from, Hey, give this another shot.

It's gotten better or what have you. So developers have strong opinions and they're tightly held. So when you're building a product for, for that audience, it had better be impressive out of the gate. I mean, like, as you think about kind of go to market, developers don't like being sold to, they have a BS detector that's, very sensitive. And so, I think as you think about the products, it really can't be buzz wordy. It actually has to have some utility, or, or provide some, something that's, meaningful, that helps their workflow, that like solves an acute pain point.

They're mercenaries in that way. They're very utilitarian. I suspect other, customer profiles have similar things, but the minute a developer feels like they're being sold to, you're basically done.

Brett: Maybe if that doesn't go well, developers tend to struggle to update their priors, right? They have a perspective and it maybe is more fixed. How does that influence the way that you think about building products?

Because it seems like it's at odds with. Trying to get something in the hands of someone very early. See how people react, iterate, iterate, iterate, where it almost feels like you can quickly start to create a negative reputation or burn your reputation or some other dynamic.

Dennis: You know, as I think about the Unblocked journey, what have you, we've been very fortunate and the same thing in Buddybuild, actually, where the early stages, you know, month to month, year to year, you look back and you're like, embarrassed by the thing that you built. And so you kind of have to beg and plead and set expectations with the people that are using it in the early days to be like, look, we're going to be learning a lot from you. If you stop using the product, that's learning for us. Painful, for sure, but it's learning none the less. I can think of very, very few examples.

Even people who I would consider friends who after the first, you know, build of whatever we were creating, they don't come back after that. So you just have to recognize that you are effectively burning that relationship in some capacity. And it's not like a, it's not like a personal thing, it's just how it works.

And so you have to just be comfortable with the fact that you're just going to get repeatedly kicked, in the teeth, until you get it right. And those people sometimes, I don't know that I can think of any specific example of someone who used it at the beginning. I can actually think of one, he's like, there's no way

Brett: how good it was seven years later, whatever?

Dennis: Yeah, well, I could think of one because he's like, there's no way you're gonna make this thing work.

It was just for Buddybuild. He's like, it's just not gonna happen. There's any number of, technical problems or what have you. And he was one of our very first, kind of users at Buddybuild, and he used it till the day it was acquired.

Brett: You just sort of touched on this and talking about Buddybuild, what was sort of the founding story of the company?

Dennis: it was born out of the frustration of building our own app. We wanted to build an app. I'm laughing because the app, it was so silly. We want to build something slightly more interesting than hello world. And you know me well enough to know I'm a relatively impatient person.

I don't suffer fools. And so the app was called What An Idiot. It was just like, every time you saw someone doing something stupid, you could just take a photo of that stupidity and be like, what an idiot, just like post it. I don't know. We want to try it. 

Brett: But this is you, you left AWS and you said, I want to go start a company?

Dennis: In one of the things that you're in these big companies and they're like, man, these monolithic big systems is incredibly arcane. This can't be the kind of the state of the art of how the world works or whatever. I just want to, I'd been spending a lot of time in the Android ecosystem, well, let's get back up to speed on the iOS ecosystem.

And so the, the workflow I wanted to unlock is pretty straightforward. I want to make a change. I push the change and I get a build and it shows up on my device. And I'm working in a team and I was like, well, how hard can that be? At Amazon, it was super easy. Microsoft, relatively straightforward.

Brett: But you were not doing this in, I want to go explore and try to find an opportunity to start a company.

Dennis: The app was in retrospect, oh, we'll try to do something kind of consumer that, cause that was actually the interesting, cool thing is like, what apps can you build? This is really strange.

And this is so hard. I was convinced that we were doing it wrong. So I went and talked to a bunch of people and they're like, no, no, no. This is the state of the world. This is how it works. I was like, well, that's crazy. And there was no DevTool companies really meaningful DevTool companies at the time.

That was the genesis of it. It was like, we were thought we could build a company around this app. We realized that in the process of building that it was incredibly painful. And so like, well, what if we made that a delightful experience?

Brett: How did you figure out what the first version of the product was going to be or where you're going to start?

Dennis: We've always had this kind of model. We, we think about like, We call it BXT, business experience technology. Is there a business around it? And these early stage companies, you kind of have to suspend disbelief for a little bit to say like, yeah, at some point if we built something, people will like it.

We literally just think a lot about the experience. We model out At the time, I think it was sketch, but we took pieces of paper and pen. I'm like, I want to do this and I want that to happen. So on and so forth. And then you used it as a user and say, like, is this thing going to work?

And that was actually kind of one of the anti-patterns for Microsoft. Microsoft focused a lot on what's the technology that we're going to use? And they're just running around the campus trying to find a problem for it. So we did, what's the business, what's the experience? And then what's the technology that enables that experience?

Who cares what it is? That's still something to this day. Even when I think about like, Unblocked and people are like, oh, it's an AI dev tool. And I was like, in the same way, it's a Postgres dev tool or a Kotlin dev tool, who cares what technology we're using? It solves a very specific problem for people.

It just happens to be that we're using this technology and that will change and evolve as time goes on. So in Buddybuild it was that, and we're like, here's the experience that we want. I literally do get push, I get a screen that I can see this thing now creating a build and then, you know, we get the result and can, can push it to a device.

Brett: Did you think about building it for a very specific customer? 

Dennis: I don't think

anyone knows exactly where their customers are going to be, certainly the ones who are going to be paying out of the gate. I'd be really surprised if you knew that. I mean, it must be something that someone else is doing at that point. In this case, there wasn't really a meaningful hosted continuous integration for mobile specifically.

And so we're like, well, these are probably generic enough workflows. There's going to be differences in security requirements, logging, this, that, whatever. If we get the core of it right, then we can like expand out from there. That workflow of pushing code and it generating a build is not new

Brett: . 

When did you start to get a sense as you started to iterate and work on the company that, there was some, some really strong signs of product market fit?

Dennis: Actually the highest compliment I've ever received in my career I remember it was a gentleman who was based in Eastern Europe and he wrote in and said that like the time that Buddybuild saved him having to manage a Mac mini and all this other stuff that's the time that he gets to spend with his family and he really appreciated that. And I was like that's it if we can get people to unsolicitedly provide feedback Even if the company doesn't work at that point, I was like, I made an impact in that person's life to the point where they want to take the time to let us know.

That was fantastic. So when people start telling us that they love the product.

Brett: When you think about developing something that has that kind of love and then turning that into revenue or getting people to pay, was that very intuitive and easy? The hard part is getting somebody to actually care about it. And once they did, it was easy to turn revenue on or no, that was its own Process of tinkering to sort of figure that out?

Dennis: No, I mean, I think for developer tools, because it is a very opinionated group of people, if it integrates into their workflow and they love the product, they want to support you. They understand kind of how it works. And so for Buddybuild, and this is anything we're, we're doing with Unblocked right now, which is, pre launching pricing, we're still learning and we were still learning from a bunch of people, how they were using the product, you know, trying to find that ideal product market fit.

And so the deal was at Buddybuild, if you help us along that journey, when we launch pricing, you're grandfathered in. And the thinking is you've already paid us with something more valuable than money. It's with your time, So, like, if you take the time to use this product and provide feedback to us, feels disingenuous, then say, also give me money because we're already getting that benefit.

So the day that we launched pricing, I think we had somewhere in around 100 teams who you were using the product we had built this grandfathering free plan into it and something like 80 something of them actually, bought a plan within the first couple of days.

Brett: Even though they weren't required to. That's cool. 

When you fast forward, you ended up choosing to sell the company to Apple. Do you think the setup of the company would have allowed you to compound it for 10 years and build something very large from a revenue based perspective?

Or was, there a constraint in the business, a bottleneck that would always mean that it couldn't be a massive standalone business?

Dennis: I think you see massive standalone businesses that start with one thing and kind of shift to others, right? It's like an in road. I'm sure there's a parallel universe where that business is doing incredibly well. I think it's hard to, really meaningfully have an opinion, like one way or the other.

Uh, I think we would have continued to learn. Super excited on the team. I don't think mobile has gotten any smaller. we could have gone the distance for sure.

Brett: Are there specific things that you figured out in building the first company that now that you're building this company you're bringing over?

Dennis: There's an operational part of it. For sure. I think there's you know, personal mental health component to it, for sure. I mean, I think in some capacity, it's like, if we're in this office right now, if we shut off all the lights and had to walk through here, we'd probably bump into the table and the chairs and the cameras, whatever it is, right?

Because we don't know where they are. But if we've been in for a little bit and we shut off the lights, You probably have some sense for where it is and you're probably trying to look for a wall or what have you. And so the second time, that's the analogy that I think of. I roughly know where the furniture is, I'll probably still bonk my toe on a few things. But hopefully I don't break it. You know what I mean? Hopefully it's just a bruise or what have you. Some of the specific things, like I think managing your psychology is a big, that's a huge difference between when I did a Buddybuild and Unblocked.

Now, I wish I was still better at it. I still get, probably overly, this is a great quote, and I've heard it before, I've seen it a couple of times, but it's, you're never as bad or as good as you think you are. and it's easy to say, but really hard to internalize. And so on the days that you land that big deal, you're like, I'm invincible.

You know, the James Bond guy who jumps up at the end of it. And when, you know, the system is down for an extended period of time or demo goes poorly or a customer, churns out or what have you, you think that like, well, this is the end of the world. in Buddybuild, I lived so intensely moment to moment that I felt it to a point where it was very, very unhealthy.

Now it still hurts, but I think I have a little bit more perspective to be like, Oh, okay. It could be better. It could be worse. And it will be.

Brett: Part of what you're saying is that a big part of your identity is sort of wrapped in the company and that the ups and downs translate directly into, are you happy? Are you frustrated? Are you angry? Aren't there a lot of benefits of that?

Dennis: Yes and no. That's part of the interesting experiment of doing it twice. at some point, Unblocked will end. my journey with Unblocked will end for whatever reason. If I have the same mental headspace as I did with Buddybuild, then this experiment did not work.

I didn't modulate it the way that I, I kind of expected. The way that Buddybuild ended was great in many ways and painful in a lot of other ways, for sure. that is like one of the things I'm very interested in.

I'm very, very principled around the amount of time that I spend on a weekend working because I remember that after the acquisition had closed, getting out of the shower on a Saturday morning and thinking like, What do I do today? Because I would typically go to work, And so it was foreign to me that people wouldn't be going into the office on a Saturday.

And now I'm very strict around like on Saturdays, I try not to be on my phone or on a computer. I'm doing something else. Any number of people who told me at the time, like, you're running a marathon at a sprint pace, you won't be able to sustain it. And my internal dialogue was like, watch me.

But, you know, there's a reason the expression exists. 

Brett: But you don't, you don't worry that actually was a driver to the success of the 

Dennis: Because like now I have evidence that proves lived experience that proves that like if I get distance from it Yeah, yeah, lived experience the distance from the actual problem helps you solve it better. Those Saturdays, things that I'm like mulling over and bothering me and I just can't wrap my brain around or what have you, when I take a day away from it, you come back to the problem and it's seemingly simpler to resolve. The product ideas that we've come up with, I feel like you just start to lose the forest for the trees, when you kind of go in a different direction and do something different. When you see it, you can see what's a tree, what's a forest and it's a lot clearer to you. It's very counter to my being to be like, Let it go and it'll be easier.

it's that perseverance. I just need to like, keep hacking away at this thing and I will figure it out at some point in time. And it just gets worse and worse and worse, diminishing margin of return, which I understand, but I can't help, but do anyway.

Brett: How are you able to phase shift like this?

Dennis: I generally don't stay in Vancouver on Saturday. I go to like another place. got a dog. There's like the dog doesn't care what, what work you have to do, right? The dog's like, I need to go to the bathroom now. You're like, give me five more minutes.

That only, that's only bad for you. there are just like constraints. So that's, those are systematized things that I put into place. And then I specifically asked the people that I worked with at Unblocked. Because of the same people that I work with in Buddybuild, investors, advisors, you know, a team, to be like, if you see me turning into Buddybuild, Dennis, you got to shake me and let me know, because I know it's not the actual most efficient or the best version of me that I can have.

Brett: You sort of decided that you wanted to debug this when you were at Apple post acquisition?

Dennis: I think Apple was a bit of a mourning period for sure, to see that Buddybuild had no longer existed and your identity starts to get separated from that company. And then, so I think it's probably when the early ideas started to set in, and then after Apple, when I was kind of at loose ends, this actually had far more impact on me than I would have expected or anticipated.

Brett: What was interesting about the Apple experience? 

Dennis: You get to see how another big company does it. That's pretty much it. I'm not a big company person. 

Brett: But that's interesting 'cause you spent the first 11 plus years of your career as a big company.

Dennis: Yeah. But what was I doing at those big companies? I was

Brett: new build.

Dennis: New things, right? Small projects where they're like expected not to work. I was basically, you know, and I hated this, actually, it was interesting when I was at Microsoft, I remember people going to start companies and I'd be like, Oh yeah, I'm doing a startup inside a company.

And you're like, that is the most disingenuous thing you can say. I was like, really, if you, fail, does your paycheck stop? Because mine does. If you fail, what happens to your team? They go get new jobs for the most part here, or there, or what have you,, right? Well, if I fail, the people that I've worked with also have to get new jobs, but they just spent years of their life that just like goes up in smoke.

But I was working on net new projects, so it was like, I was already wired for entrepreneurship in some capacity.

Brett: In building Unblocked, did you leave Apple and say, okay, I'm gonna go start a company? And then started a route around for areas that you were going to spend time. Or did you bump up into sort of the starting point for Unblocked?

Dennis: I definitely, you know, I left Apple thinking like, I basically wouldn't take any conversations if they related to anything technology related. I'd sat on a couple of like boards as a consequence of some angel investments, but that was it. If people wanted to talk to me about anything tech related, I was not interested in it.

Zero. I'm just kind of burned out on it. I didn't basically wouldn't talk about it, wouldn't take any meetings, wouldn't discuss it for about a year and a half. I just wanted some space from it. I mean, probably not to do something from what we were talking about earlier, which is you just need distance from things to see, what's important versus what's not.

Brett: Did you create specific systems for you? Because I would assume left your own devices, you would kind of go start working on another company.

Dennis: I did two house renovations at the same time. And so, people were like, that's crazy, you did two of the same, we were talking about this earlier, I was like, you did two renos at the same time? By comparison of being the CEO of a

Brett: So you put all the energy into that.

Dennis: A portion of it. Like the CEO job was way harder. I was like, this is child's play by comparison.

I had lots of time, the middle point there actually was, and I kind of dawned on me, one of my really good friends, I was a groomsman for him and he'd made these custom, boxes as gifts and the customization was the title, it was engraved as he thought about his groomsman. And so for me, what he made was the builder and it never really dawned on me, but I actually just love to build stuff.

So the meta point for those projects, those renovations, was that I was still building things. It happened to be in the physical world, and I loved all the little nuances and details and all that kind of stuff. But, it was actually, that, it was a certain, you very specifically remember sitting in, a conference room with a designer.

and it felt like I was back, Building software again, which is like, well, what's this going to look like? How does this we're in the same conference room, looking at a piece of paper, figuring out the experience for what the house is going to be or what have you. That's when I was like, I really miss this energy. 

Brett: And it was a pleasurable experience doing the renovations?

Dennis: I mean, to the degree that rentals are pleasurable.

Sure. Yeah. But it sounds more positive than a normal renovation.

because I'm looking at the outcome. If you halfway through, no, yeah, no.

Brett: Why does it seem that doing renovations or building houses has gotten no better in 20 or 30 years?

Dennis: That was one of the things I was thinking about doing. I was like, this is such a broken, broken system. That's a whole other conversation. I don't actually even know the answer to that. It is an incredibly underserved market from a technology perspective, but I think the go to market is actually really, really hard for it.

it feels like at some point someone's going to figure out the incantation between these two things. But you know, I didn't, I don't think, there was definitely tipping point where I was I really miss this energy and really miss the people that I was working with. I didn't really want to do that again. And then, so yeah, construction. I was that seems like a super broken problem. There's a couple of other things, but I was this problem that Unblocked is solving, this inability for developers to get the information they need. That actually just irritated me. Like I am a, I'm not a super patient person.

I walk fast, I talk fast. you know, like, let's move on with that kind of thing. I'm a mediocre developer on a good day. And so those are the two things that kind of combined to be, I hated it when I had to ask someone and wait for them to give me an answer. Or when someone had to like, ask me a question that I've answered repeatedly multiple times.

I'm like, this is crazy. Why do we do it this way? And so that was the genesis of it.

Brett: Did you just intuitively bump up against this and decided that that was going to be where you're going to spend time on or was there sort of a process of exploration that landed you?

Dennis: , It's not too different from Buddybuild was, I want to push code and a build shows up. My thing was like, I want to ask a question or be able to get the information I need and get a response with the answer. I'm tired of like bothering people. We were going to call the company bother at one point.

You don't bother me. I don't bother you. Like weird, negative connotation decided

Brett: You have interesting names. Idiot app or whatever

Dennis: What An Idiot. 

Brett: What An Idiot 

Dennis: What you say every time you see. Yeah.

Brett: In what ways does being a second time founder actually get in the way?

Dennis: Oh yeah, that's super easy for me. People, I think, give you the benefit of the experience of the first time. There's a huge amount of like luck and serendipity, you know, a decent amount of like experience and hard work and so on and so forth, there's a bunch of entropy associated with these companies being successful or not.

And so, like, attributing that to skill and giving you the benefit in the second, for the second time, statistically speaking, these things don't work, So to do two of them is like, and both of them work, is like, you know, mathematically, if you look at it, you'd be like, no, that shouldn't work.

But people somehow be like, we did it once before, so it should work the second time. And so I've been very clear, you should hold me just as accountable, if not more so, than a first time founder, right? There's certain, I'm sure there's certain things that I know how to solve, but there's certain things that maybe didn't run into that piece of furniture or whatever in that dark room analogy.

So, keep me honest.

Brett: And so what does it look like when you say people give you the benefit of the doubt?

Dennis: Any number of things, like how to run an effective fundraising process, how to do hiring all the fundamentals that you would expect.

I've learned for a bunch of things, but maybe I didn't get the full experience to know that I've, so people like, well, he did it once before he's done fundraising before. So like how hard could it be for him to do it the second time? Yes, that's probably true for 80 percent of it, but what about the 20 percent that I didn't Might not have had experienced or. The market then when I was starting Buddybuild, our seed round was 1. 1 million, the seed round for unblocked was very different.

Brett: We should have done the first version. That would have been nice.

 So as you started to build the product for Unblocked, what was that, you kind of had this, clarity around the problem that you were trying to solve.

Can you talk more about how you translated that? And, it's not like you just built this thing in two months and then everything started working.

You've been iterating and iterating. Can you sort of talk through what that's been like?

Dennis: It's just repeatedly getting kicked in the teeth. It felt like I need to go to the dentist every month to get dentures put in, right? You know, finding product market fit is hard. We're probably crossing that. We are getting those emails that say people love the product.

you know, a pretty good indicator, what have you. Is it a business? We'll find out. I think so. I don't have to squint hard to, have conviction around that. But yeah, like the first two years, a year and a half, I would say it was just experimenting, exploring different

Brett: Then, yeah. What does it actually look like?

and kind of how the different, Iterations sort of connected together.

Dennis: I mean, I think the first iteration is like, it wouldn't be great if you could just ask a question and run a specific snippet of code. So select the region of code, right click, ask a question, unblock would provide you with the best people to answer the question. And then you could have a conversation in the code with that person and then it gets stored beside it effectively. So the next developer who comes into the code base can see that conversation there. Well, there's all sorts of problems with that. It sounds great out of the gate. 

Brett: And so you built that?

Dennis: We built that. We showed it to a bunch of people quickly. It kind of realized one of the

Brett: as a fully functional product or, or sort of mocks and things like

Dennis: No, like again, developers, it's like we come up with the X, the experience and technology to make that happen. We then have to build it. Beg people to use it, and then when they don't, double beg them to tell us why, because people don't want to hurt your feelings, I mean, the obvious problem with that is that it's like going to the gym, you have to, like, do it a lot before you start to see the benefit of it. So starting from scratch on a code base, And saying like getting people to learn this new habit, start generating content that could then somewhat, sometime down the line be valuable to someone, it's going to be really, really hard.

So we evolved it from there. 

Brett: So you, gave it to some people at the beginning, they're like, Oh, this is interesting at a high level and they wouldn't use it?

Dennis: No traction. Yeah. Basically. Or very, very marginal. And so we just evolved it and say like, what's the problem? The lack of basically bootstrapped information was one of the limitations. So let's evolve that. And then you just keep bundling it on.

And again, you only have a certain number of shots at this from a relationship perspective, because at some point you'll burn through your network. And so like people would do us a favor, in using the product. I can think of a whole bunch of them that had used the product in the past that, they won't reply to my email as it applies to Unblocked, but we can go for a beer, that kind of thing.

Brett: Huh.

So, so talk more about, you started with sort of this, The set of conversations that exist around code snippets. Explain the set of iterations that got you to where you are today.

Dennis: We started with like selecting region of code, asking a question. Then we're like, okay, how do we bootstrap information? There's a whole bunch of historical conversations. That exist already within the source code management system.

It's in these pull requests, right? So the pull requests are the changes that a developer wants to make, and typically there's a conversation that happens, with the development team. And we're like, well, why don't we surface those for you to discover that content as you're scrolling through the code.

It's already there, it's just opaque effectively. So we'll surface those and allow you to ask questions. 

Brett: And so that was kind of getting a getting over the cold cold start problem or like the time to value. I'm going to do something that's, takes effort today to get it, for my future self to benefit, which didn't really work.

Dennis: And people are like, oh, that's interesting. But again, kind of hit or miss in terms of the amount of content. Well, developers have conversations in a whole bunch of different places. It's not just in the source code management system, they have it in Slack, they have it in Notion, in JIRA, in Google Docs, in Linear, in all the tools, these kind of code adjacent systems.

well, okay, what if we were, like, somehow able to, like, reconcile the source code that you're looking at with conversations in Slack, or in these other data sources, effectively? And then the advent and really kind of the earliest, well, one of some of the earliest stages of LLMs, it'll be funny in a couple of years from now, we'll look at it and be like, Oh yeah, that was the super early days, but, that's an interesting way of being able to replicate how people work.

I ask a question, I get a response. I don't have to worry about this, social pressures of asking Brett a question that he might've already answered or interrupting him or what have you. and I don't have to wait for him to reply because I get it basically instantaneously. if we do it right, it's across all of the information that exists in the entire organization.

So Brett might not even necessarily know of some of the information that I'm looking for, whereas Unblocked should.

Brett: And so walk through what the current product experience, looks and feels like.

Dennis: There's two parts to it. First is as simple as What we described, right? I can, from the IDE, from a Mac app, from Slack, from a web, wherever a developer is, they can ask Unblocked a question and Unblocked provides an answer. Shows how it got that answer, provides references, which is something that's very specific to how kind of developers to work, which is like, they're inherently skeptical.

It's actually funny. Every single person who onboards to Unblocked without fail, the first question they ask is something that they're an expert in. They look to build confidence in the system. It's a hundred percent reproducible. I've never seen an exception outside of it. So yeah, we provide that answer.

We give you the, the way that, we generate that answer. So you can like click in and explore those things. And then we provide you a list of like suggested follow up questions. So if you're trying to learn the space in the same way that you and I having a conversation and be like, Hey, you should also check out this.

It helps me explore, understand the problem that I'm trying to actually, get used to get my, to get my work done.

Brett: It's trained on all the historical conversations that have happened.

Dennis: It's all the conversations that have happened across all the data sources. So it's like, it's the thing of the taking the source code and augmenting it with all the information that sits beside it. That's the kind of source of truth, and then you have to like siphon through and really understand what is the data that reflects reality.

So one of the hardest problems, one of the things that people love about the products, same thing that we did at Buddybuild, that time to first value, there is no work that you had to do to get, or very, very little, I would say to get Buddybuild to work with your iOS or Android application. You pointed it to a repo, it just magically built.

We got tons of feedback being like, this doesn't actually work. So we actually had to show you how we were doing it because people thought it was too magical. It's the same thing with Unblocked. You don't have to change your data in any way, shape, or form. We do the hard part of effectively cleaning it, organizing it, so on and so forth. Asking a team to change the way they work for some perceived benefit down the line, almost a showstopper at that point. So without any modifications to your wikis, your slack channels, whatever it is, just point us to it. We figure it out for you and we give you really, really good answers.

Brett: Is sort of the, the idea of the time to value is really important. I think particularly with developers is something that seemed like guided a lot of the product philosophy with, Buddybuild, as you were saying, is there a reason that you started with something, the first iteration of the product did have this time to value problem?

I think it was an exploration in some capacity. Like we figured we knew that that cold start problem would exist in some capacity. it sounds like they were very discrete units of exploration. There was a decent amount of overlap or like, yeah, okay, great. And now I know I need to provide some existing information.

what's been interesting about building this company kind of in lockstep as LLM's have really become a dominant potential, maybe new platform shift. I feel like that's you kind of road mobile with your last company. But there's something that feels a little bit different. I feel like when we talked, a while ago, I think you were maybe more skeptical of some of the stuff that was happening in the world of LLMs, but you almost kind of started the company right when maybe GPT 3 or GPT 2. 5 or something was created.

Dennis: In my time off, one of the things I really like to do is surf. Surfing is all about being at the right place at the right time and then executing perfectly to like catch a wave.

Brett: You really did all the exited founder stuff, the surfing and travel and,

Dennis: I don't know the travel I couldn't, no, there was no traveling cause it's COVID 

pandemic, right? So I'm like, yeah, I get to do whatever I want. 

Brett: Guess I'm particularly curious as you've been starting a new company as it's sort of really been picking up. I assume you've been constantly thinking about. Irrespective of even what you were trying to do with Unblocked, there's sort of this thing going on over here and I would imagine you're kind of ping ponging between trying to understand what's happening with the enabling technology.

Could it be applied to the different sort of developer experiences that I'm trying to create that feels different than what you were trying to do when you built your last company, where it was like a clear problem that developer has have, but we're going to solve this problem with, with this new product.

Dennis: I can see why you think they're different, but they're actually very similar. It's like, it really does start with what is the pain point? what is the thing that's frustrating or terrible about this developer workflow, this developer experience? And then what's the technology that we need to use to like solve that, pain point, the stuff that we were building prior to the LM.

And I'll yeah, totally acknowledge the fact that I was like inherently skeptical, around how good the technology could be. But in my career, I don't know that I've seen anything, it's unclear to me what the blast radius of all of this, this latest generative AI technology will be. I think we only know over probably the course of a decade or so.

The rate at which it improves though, I've never seen. And so I was inherently skeptical to begin with, and I think there's still a lot of FUD. I think there's still a lot of noise. But to the degree that you use any technology to solve a real problem, that tends to have staying power. that solves this problem in a very unique way, in a way that wasn't, you weren't able to do before.

Brett: And so does that mean that you end up and the team ends up constantly spending time trying to figure out what's. possible now today that might be wasn't possible six months ago or whatever?

Dennis: Yeah. I think most people who create products are embarrassed by the thing that they built a year ago. Now I'm embarrassed by what I built three months ago. That's the way I think about it. We had a customer who had reached out to us several months ago, and had to go through a bunch of internal, issues before they could actually use Unblocked.

And they reached out this week. I'm glad you waited because it's going to be mind blowingly better than what you would have expected before.

Brett: I mean, this has always been the case, but it feels particularly acute that if you think about the space of, like, generative AI, plus the developer tools, it is incredibly crowded. There's a lot of people saying similar things. Does that impact the way that you think about, Product marketing, product development, what you're doing, or are you kind of the type of founder that's just, let's focus on the customer, build something that's going to solve a big problem for them and if there's 30 other things that sound similar to us, I don't really care. It doesn't matter.

Dennis: Yeah, I think the tide's in and when it goes out, we'll see who what's actually remains, right? I think there's a lot of noise and distraction, but this is typical, this is typically the case with any new technology. You know, I think if one of the keys to success is actually building something that people love and it solves a real problem for them, then the tide can come in or out or what have you, and you'll be fine.

Brett: Do you find when you're talking to customers that they're like, Oh, I've heard this before. I tried this other thing that sucked. And so now I'm assuming that what you built sucks.

Dennis: That's a perfect insight. Yeah, absolutely. You could decompose a lot of our customer conversations that too, Oh, we've never heard of this thing before, or we tried it and, or we tried to do it ourselves and it was terrible. In the first, you, you have to deal with a certain amount of skepticism in both.

It's just a different starting point. I think it's because it is the thing that we do and the technology evolves and we obsess over it. One of the things that people find surprising is how delightful it is. I was talking to a VP of engineering of a company that's using it.

And he's like, you guys just, the level of craftsmanship and the way that you build the software, the actual experience, you treat developers like people, as opposed to a lot of rough edges or what have you. I don't know, certainly that's one of the things we did at Buddybuild too.

we cared a lot about what the end user experience was, in the product, the way you reach out with the support, so on and so forth. And so I think that's a differentiator that a lot of these companies just don't care about. I use other tools, I use other services, and when it takes a day for someone to get back to me on intercom, I've basically flipped a bit, I've moved on.

Brett: And do you think it's that most people think about all of those sort of little rough edges in isolation and so that any singular rough edge maybe doesn't matter, but that they're the winning strategy is if you're obsessed about all of them, that they compound into sort of this ultimate experience that, you wouldn't go and tell somebody that you have to go and try Unblocked because all the rough edges are filed away.

But there's something about it that you kind of like know when you experience it.

Dennis: I think like it's the culmination of those edges, right? Death by a thousand paper cuts. But sometimes they're just there's a lot of noise and it's like, oh, I built this new VS code extension that does code completion for you. there's endless number of like limitations the way you install it the way where any any number of problems that would have you go to reach out so like the product is mediocre and then you have an issue and you try to reach out to support and like no one gets back to you or they get back to you in a week or whatever it is.

You know going back to those Amazon leadership principles being customer obsessed matters a lot. And so especially for a developer audience who is very, very opinionated. If you make a mistake if there's a small window in which you can apologize and try to recoup that trust or respect, you really have to take advantage of it.

And I think there's a lot of noise, but I just don't think a lot of people think about that stuff.

Brett: In terms of inculcating that in the company, do you just behave that way? And everyone models that behavior after you, and so sort of you're unconsciously aligned, or is it something you're constantly pushing as a small team?

Dennis: I think a lot of a company's culture is a consequence of it's founding team or what have you. I think the founding team at Unblocked and, you know, same with Buddybuild was customer obsessed. It was funny when we looked through our kind of our, all our support channels at Buddybuild, the two, the people who answered the most questions, over time were my co founder and I, he beat me.

Which just drives me nuts to this day. I would get up at four or five o'clock in the morning and do support, for Buddybuild And it's the same thing I do with Unblocked every piece of feedback that comes in, I see it. I respond to most of them at this point. I think that will be less likely the case as time progresses, but I will see every single one of them for sure.

Brett: How do you then make sense of that amount of feedback coming in at any given point in time on big and small things? How do you decide how you're pulling all that together to decide what to do or work on next?

Dennis: I think you need to have an internal intuition or roadmap for what you think is important and then you use that as calibration. Something might be tied and you see a bunch of feedback. This is to unlock this customer or to solve these pain points, there's different classes of problems that you're going to solve, need to have an opinion and then that data helps shape that opinion.

Brett: What does an opinion look like? 

Dennis: I'll give you a very classic example. at Unblocked, we built GitHub as our first source code provider that you could connect to. Then the question became, was it GitLab or Bitbucket that we should do next?

So I had an opinion on that. And then we saw a bunch of people who wanted to use a product that were using one versus the other. And so it helps substantiate in this case, the right opinion. I can give you tons of examples where I had an opinion the data came in. And we had to go the other direction, if you just constantly follow what everyone's asking, I mean, that's just kind of product management 101. You need to be thematically consistent with what you think the business priorities are, build a product relative to those things, and then figure out which adjustments you need to make relative to that customer feedback.

Brett: Have you been picky about the type the shape or style of customer that you want as your first 50, 100, 500, or 

Dennis: I mean, it was the same thing with Buddybuild. At Buddybuild, we would take anyone who would have us. At the end of Buddybuild, I was like, oh, you're this role? Not interested. Like, I just know that's not going to be a good fit. It is the process of constantly refining through, and I don't know, I would love to know if there was a shortcut. But typically, it's just you have to sit there and spend the time. See what questions are being asked. See what users look like and figure it out. Unblocked is not a great tool for four people who've been working on a product for a year, have co locate in the same space.

We knew that out of the gate, that's obvious, there isn't a huge hysteresis in terms of content, they haven't developed a lot of systems and so on and so forth, how do you progressively move up the line there? Is it 4 people? Is it 20 people?

Is it 50 people? And so you need to have some assumptions, try to find people who substantiate that, and it's a multivariant problem. And they're geographically distributed and they only use Microsoft Teams. Well, we don't have team support at this point in time. You're done. So like, oh, 20 people doesn't work?

Well, no, no, it's because they had that one other. It's just a lot of iterations and a lot of conversations and a lot of kicking, kicking in the teeth. I was talking to a founder the other day. he has companies doing fairly well. And he's, I think, ready to exit.

Honestly, I think I just people telling me no. Like now that I've solved it, I just don't find it challenging or interesting. And I was like, Oh, that's a very unique, uh, unique 

Brett: I feel like you'll be okay with more people liking it.

Dennis: Yeah.

Brett: Shifting gears slightly, a lot of the folks that are, I don't know if it's entirely, but certainly a lot of the folks came from Buddybuild that you're building Unblocked with. So obviously you've got something right in those hires. What were you looking for back then, or if you were to kind of zoom out and think about, at least in the context of the way that you like to build companies, What's differential about, like, those human beings relative to other wonderful people that you've worked with?

Dennis: I guess we have a shared experience. It's actually an interesting dynamic because as we're growing the team now, we have this pre baked culture and perspective on things. And so we have to be mindful of new employees joining the team and, you know, Now growing it and making sure that like sanity checking that some of the stuff that we're doing that have been normalized, do they need to be improved or not?

typically that's a lot more of organic process, but there's, 10 of us effectively right out of the gate that started and like have that shared that background and that culture. So as we bring people in, I'm very, very mindful and explicit in observing what are predispositions that we have that we need to reevaluate or reassess. What makes them special.

Brett: When you assembled them at Buddybuild for the first time, before you had created a company or anything, you obviously got something right because you want to do it again with them. And so, like, when you dissect kind of what's going on for those people, what's unique about them that made them work in this setup?

Dennis: We share the same set of beliefs, right? We're like, we like to operate the same way. This will be something that's very important to us and we prioritize it. Not to the detriment of like our personal relationships or mental health or anything like that to the degree possible.

we share the same values. We are a team of people who are, have we're very clear what our culture is. We have a strong bias for action. We are customer obsessed. We disagree and commit a lot of those Amazon leadership principles echo, slight variants of them, within these companies as well.

You know, there's a handful of people at Buddybuild, who didn't stay with the company while it still existed or what have you, and it's typically a variant of some of those shared values,

Brett: And did you explicitly look for them?

Dennis: I think so, but I mean, arguably we had assessed those when we interviewed those people who didn't work out as well. I think the characteristics of the people that have impacted me the most are the ones who like are empathetic and curious. I just love that. And humble. my favorite people are the ones who, like, have accomplished a lot of things kind of have nothing to prove.

They're very genuinely curious people and, like, really empathize with whatever thing that they're thinking about at that point. That's kind of, like, my magic trifecta. And so I've seen that, I like this quote, which is like the empty vessels make the most noise, right? So if you take a glass, make it half full and then run air over it, the more water there, the more substance there is, the less noise it makes, right?

Empty ones make the most noise. so I look for people who don't make a lot of noise. I think that's something that like, I definitely aspire to. And there's a handful of leaders that I've worked with who were like, Very, very good listeners. even the way that you, this is, you remind me of one of them actually, the way that you, every time we've always had conversations has always been, I can tell that you're listening to what I'm saying.

Brett: Why is it so hard to hire accurately?

Dennis: Yeah. I mean, it's incredibly imprecise, even in the way that we're talking about right now, the things that I'm saying and what your understanding of them, we have a whole bunch of shared context, or if I say something, you're like, Dennis is, is joking, or he doesn't mean that or what have you.

But you're going to spend 2 or 3 hours with someone, 5 hours, 10 hours with them in an interview and be able to come up with a determination, this is something that we're going to spend 50, 60, 70 hours a week together on, but I've only known you for 10 and we're going to go through this journey.

The probability of that actually working out is actually pretty miraculous when it does. 

Brett: But it's also amazing that state of the art hasn't come that far.

Dennis: I think there's certain things that you can only, it's only through lived experience. Sharing the pain, when someone churns out or the wind or the joy when something you win a deal or what have you. How do you replicate that with technology? that's something like how people react to It's more obviously like it's obviously more interesting when people react to the negative experiences or situations or what have you.

Brett: When you think about where the people came from, the first 5, 10, 15 people that, that were really successful in the context of Buddybuild and now potentially in Unblock, do they come from all sorts of small companies, big companies, startups, there's just no rhyme or reason.

Dennis: Arguably I was a big company person when we started Buddybuilds, right? I worked at a big company, Microsoft and Amazon. What does this guy know about starting and running an operating business or what have you? I think that's just intellectually lazy.

 I don't know where I would find it now. I copied and pasted it. You had these great interview questions, which I thought were the most, you know, those behavioral type. Yeah, there was behavioral type questions that you can see all over the place.

You know, I've done interview training at these other companies, but the thoroughness and thoughtfulness of the questions that you asked specifically, we would use those as a tool to figure out How do we basically speed up the first 30 to 50 hours of working with someone? 

Brett: I think everyone's very competency focused, can someone program in a certain level of proficiency. It's easier to assess competence than it is, will the setup here be the right fit for someone?

I think it's easy to say, well, there are good people or bad people or incompetent people or whatever. But I think. Most of the time, the setup is just wrong for that person.

Dennis: Our interview process for both companies has always been two parts to it. One: they have some level of technical competence that we expect them to have if they're you know an engineer hire as an example. But even if they're phenomenally technically competent if they're not going to fit in the company what difference does it matter?

if they're just like, Oh, I don't care about, you know, customer feedback or, you know, I stopped sharply working at four 30 and like, you'll see me the next day at eight 30 in the morning or whatever, those are, you can be phenomenal and still not going to work.

Brett: Do you think there's a difference in building a great product and being able to turn that into a great business, or are they basically one and the same? 

Dennis: I've seen a lot of very, very compelling technologies that don't end up being interesting or compelling businesses, even if they do solve problems. I think it's really hard to have a meaningful business with a mediocre product, but I can think of a handful of companies that do that too. It'd be interesting if you went back and talked to the employees who were early at Microsoft and early at Amazon, how much, historical retrofitting has been done to go and do, and I see a couple, there's actually a couple of interesting YouTube videos that talked about folks who were very early at Microsoft now, and there is a lot of, Retroactive rationalization of how things kind of played out based on certainly their perspective.

Brett: But in the, in the context of you building your own companies, your kind of current way of thinking about it is if we build a product that delivers sufficient value, the business will follow. 

Dennis: I guess I, yeah, I have an opinion here, which is if you don't have a compelling product, there's no really point in building the business. We're at this phase now where we have what seems to be reasonable product market fit and it's worth running, I think of these as basically a hypothesis with a series of experiments that you run.

And so the first experiment that I like to run is can we build something that people love? And if you do, then the next experiment is can we commercialize it in a material way? If I put a dollar into this machine, am I going to get 2 dollars out minus 6 dollars out? What does that machine kind of look like? But you need to have like that, that first part of it, that, that product in there first. If we didn't find that, we would have shut the business down. You know what I mean? Would be like, Hey, we're turning capital or acquisition or whatever the thing might necessitate.

But now we're interested in running that next experiment, let's see how, commercially viable this thing is. One of the biggest learnings I had at Buddybuild, I love behavioral economics. It's like one of my favorite things. I remember one specific customer saying under no circumstance when you start charging pricing, will we actually pay for this thing?

We have this internal culture of building it ourselves, not relying on third parties and so forth. And they were the second customer who were grandfathered into a free plan to like buy our most expensive tier at that time. And so I was like, I know nothing about this.

Brett: So what did you make of that?

Dennis: people are completely, we'll say one thing and do another thing.

And when it comes to pricing, It is it s own art and science at the same time.

Brett: What are some of the interesting things that come to mind for you in terms of pricing, maybe specifically in the context of, developer oriented products?

Dennis: I was very fortunate, Stewart Butterfield was one of the advisors at Buddybuild. And I distinctly remember being in a very noisy restaurant, which we both didn't enjoy. And we were talking about pricing. And he was talking about adjusting Slack's pricing.

they'd done it, I don't know, a year earlier. And I was fretting over our pricing and what we should do. And he's he's like, just double it. And I was like, what are you so much heartburn? Oh, you're kidding. I was like, yeah. Grandfather, the old people, and just double it, move on. That was like the, the end of it.

and he was right. We doubled it. It had no impact on on the business in terms of, customers coming in or what have you, other than we doubled our revenues on that day, moving forward. It was, it was a great lesson for me. You really, as a, founder, I think one of the blind spots that you have is you don't appreciate how much value you're creating for people and you incredibly under, undervalue it and charge people accordingly.

It'll be an interesting experiment for, for Unblock too.

Brett: When you talk about the second experiment of commercializing these early signs of product market fit, can you spend more time kind of picking that apart? You talked about this journey in to, you know, sort of iterating on the product. When you think about pricing and packaging, go to what is this next experiment actually look in the context of, the company?

Dennis: So true product market fit is the intersection of when you can identify the person you're going to sell it to, you know, exactly what you're going to say to them and you have the product to actually go and deliver against that value. . 

Brett: It's the repeatability of it? 

Dennis: Yeah, so it's like, you're like, you line up 100 people, I'm looking for the following ones and I can tell you which ones they are. And if you don't match that criteria, don't bother stepping across the line. You know, I'm not interested in chatting with you. And so, and the way that they vet that, one of the people who works on the team, he's calls money, liquid love, They love the product so much that they will, they show you that love by paying with liquidity. So what are the next, how do I think about that next phase? It's like, how do we get razor sharp? with every iteration that we've done of this, I'm getting increasingly clear when I sit down with someone, if they're going to be a good customer or not.

We have this, I'll give you an example of that one heuristic. Most actually every team that we've interacted with that has Confluence loves Unblocked. Confluence just happens to be the thing that is a signal for us that they've matured as an engineering organization. They have a lot of data that's scattered in a whole bunch of different places and that's not as organized and they feel like Unblock can help them discover a lot of the the knowledge that exists within their organization.

So like that is a signal, which is like, if those a hundred people lined up first question, who uses confluence? Step on up, right? So we just have to get more of those kind of heuristics, getting tighter on the messaging, getting tighter on like who that person is. and then building the machine around it from just like a economic perspective.

how much does it cost to find that person? How much will they pay you? How long does it take around for, you know, does it make sense?

Brett: Are there any things that you've sort of figured out in, positioning and finding the right language to articulate the value prop?

Dennis: It's actually funny. I love going back to the way back time machine and looking at landing pages for companies that are quote, unquote, wildly successful now, because you see the evolution of the marketing. Actually one of my favorite people, I hope she listens to this podcast is Ariel Jackson.

I told her to her face. I told Bill this as well, which is like, I've never loved and hated working with someone so much at the same time. She's so, so good at her job and it drives me absolutely bonkers. Thank you, Ariel. I think the marketing message changes as you know, as a product evolves, obviously, but as the market around it evolves, right? So I don't think of, Unblocked as a developer AI tool. I just think it was a tool that helps developers and happens to use AI. I think other people might, for any number of reasons, want to position Unblocked in a different way and say, and ride the AI wave, it's just not my, not the way that I currently think about it, but maybe that changes in a year or two.

So I think that positioning and marketing will definitely change over time. 

Brett: have you found that even when you keep the product is the fixed variable and even potentially the, shape of the customer that you were talking about, the, the same variable that changing the positioning dramatically changes how people will engage their interest level?

Thank you.

Dennis: I don't know that I have enough experience to speak to that, definitively, certainly not with Unblocked. With Buddybuild, yeah. I mean, like the way we described Buddybuild at the very beginning was very different from the way that we extended it at the end. By the end, it was like, it was a message that was catered to cut through the noise and find that exact customer.

We knew who it was, we knew what works and we had the product that actually did it. I think for a lot of early stage companies, finding that messaging and positioning to start off with, and then constantly rehoning it and revisiting it. Well, for me, I should say that's probably the, one of the most challenging things to continue to do.

It is a very, very time consuming thing. And the feedback cycle on that is it's, you don't get it on a daily or weekly basis.

Brett: It's interesting because when we, we spent a tremendous amount of time kind of thinking about this topic of product market fit, particularly in, some sort of business context. I feel like it is, underappreciated both the subtlety of all the different dimensions that, separate different types of customers, but a lot of it is the positioning and the language that you're using.

Even when you hold the product constant, most of the people we have the opportunity to work with tend to be product and technology oriented. And so I think they always think that the number one thing is I have to go fix the product. I have to evolve the product. But they often don't think about the messaging and packaging and language I'm using around this in that sort of same iterative fashion.

Dennis: I don't know if it's da Vinci or one of the famous Italian classic artists or whatever, right? I think the analogy that comes to mind is that the art exists under the marble and it's just carving away the stuff to reveal the art that was already there. So I think of like the product in kind of the same way in that you're going to constantly chip away and shape it and so on and so forth.

But that is just one of the pillars that you have to kind of go and solve. Messaging and positioning is also a related exercise that you have to kind of go and do. I have a notes file where I first, I can look back to the very first messaging and I was like, that's horrific. It's not even close to what the thing does. And the product, even if you adjust it for the static, this thing will continue to take shape and so will the messaging and it has to evolve at the same time. Yeah, I think it's a fool's errand to think that if I just solve this thing, then the rest of it kind of takes care of it, or I have to like adjust this and leave that static.

Brett: It almost feels like a key unlocking that if all the pins aren't lined up. It's really not going to work. And it's the act of sort of lining all of those thing up, those things up that gets a company into, like, really strong product market fit.

Dennis: The thing that makes it frustrating with that key analogy is that those pins continue to move, you're adjusting, you're adjusting, and finally you get into this like sweet spot where the thing kind of drops through, and then you realize, oh, I have to do it again.

Brett: , When you bring a product to a customer, you leave it with them, you see how they use it and their behavior is a good indication of the value that they're finding. When you're iterating and working through different ways to position the product.

What type of resonance are you looking for that kind of makes you feel like, oh, this direction may be more interesting than that direction.

Dennis: You know, getting on effectively a form of a sales call, when you're doing demos or what have you, if you're not taking those opportunities to iterate on how you talk about it, prescribe, describe the product and like seeing do people lean in, reading kind of the room, seeing the body language, seeing the response, that's a huge missed opportunity.

In the same way that you iterate on the product, that's your opportunity to iterate on the messaging. Because like how you talk about it, and seeing how a person reacts based on their profile, what have you, helps you narrow in or triangulate what you would use in any of like the written marketing material or spoken market material, whatever it might be.

Brett: Do you run demos in any specific way? 

Dennis: Well, I want to understand a little bit more about you and kind of how you work or what have you. Understand if like, this is a problem that you face. is this something, a problem that you can actually relate to? Again, I'm not trying to sell the product. I'm just trying to understand if this is like a real pain point for people.

And then typically people, I'm very kind of transparent. These are typically the objections that people have. How do we think about security? What are we with pricing? All these types of things. So I don't know, I've never really sat in other demos. 

Brett: No one's ever demoed anything to you.

Dennis: They have, but it's typically bullshit.

You know what I mean? Like, if I were to be blunt about it, it's staged and it doesn't feel authentic. Whereas like, no, this is actually the product on our actual code base in our actual selection. Oh, this is how we use the thing. And so like, if you were a member of the team, this is what you would see in this environment or what have you.

I'm genuinely there to learn. I'm actually curious, to see if it, works for them or what have you. My goal is not for me to say, oh, we've got another customer. I actually kind of prefer to be like, we didn't land this customer and here are the reasons why. They're not the right fit. We didn't talk about the right way. They don't use Confluence, whatever those things are. I'm more interested in the no's. I probably characterize, or I probably run the demo in a way that helps me substantiate that. Remember, like, in the same way that you would iterate on a product, I'm iterating on how I describe it, who the person is that I'm chatting with.

Brett: Were you always comfortable with people saying no and kicking you in the teeth? 

Dennis: No!

I don't like it to this day, are you kidding me? It's like, one of the worst feelings, 

Brett: how do you maintain intellectual honesty? What I've noticed is that most humans, if I have my widget and I'm bringing it to you, I want you to like my widget. If you don't like my widget, it hurts my feelings.

And because I own my widget, I have an endowment bias. So I think it's irrationally good, generally speaking. And so what I've noticed is when you're spending time with customers, you can overly contort yourself to try to hear whatever you want. But it's antithetical to sort of getting something into really strong product market fit.

Do you just recognize it's unpleasant and then that's fine? Or do you ever think about kind of that topic of intellectual honesty as it relates to interfacing with customers?

Dennis: Yeah I mean, I don't think you can be fully intellectually honest If you're talking to the person, how we're sitting, the tone of our voice, where we're wearing, all these things have influence. It's the only way you can be truly intellectually honest if you have absolutely no connection with that person, they pick the thing up of their own volition and they start using it or not using it.

Otherwise you are influencing the experiment, right?

Brett: Do you think at all about sorting those or a way that you try to make sense of all of that at any given point in time or the balance between all those different opinions, your own intuition and the direction that you want to go, sort of all that type of stuff?

Dennis: Maybe another Amazon thing. I think the people closest to the problems are the ones best positioned to solve them. I think like I truly, truly, that's like one of the things that I was really, really impressed by at Amazon.

Jeff would have a perspective, but he's like, who's the person actually solving it? You know, give them the data, but let them make effectively the operational decision. I'm being flippant. I don't, listen a lot, but the things I optimize for and the things that I sort, sort on, it's people who've been there, done that. If I have conviction that they've actually experienced this thing before, what kind of idiot doesn't take the advice of someone seriously, internalizes it for someone who's been there, done that, and it has like an interest in you being successful. That is like such a wasted opportunity, that would be my magical superpower.

If I could like shake people's hands, we've talked about this in the past, shake people's hands and benefit from all the life lessons without having to go through the experience, I would do that over fly or be invisible any day. What do I sort by? Yeah. That's one of the things that I, the investors that I work with, as an example, I like people who have operational experience, who've, done something similar, that can say like, Hey, Dennis, I know every situation is different, but when I had something similar, here's what I did and what worked and what didn't work.

Brett: Particularly because the team is small and you're very close to the customer. you have an opinion on everything. But you will, lean on the opinion of the person that's working closest to whatever the problem you're working on.

Dennis: At any point, at any lifecycle, the company is someone who's like, I think it would be a really sad day, something has gone terribly wrong and kind of my management leadership or whatever that they can't come in and be, I think you're wrong, and here are the reasons why.

In a respectful, open way, in the same way I would approach them if they, if I thought they were wrong. But I don't ever want to be in a place where, well, because of my job or the CEO or whatever it is. the best, like, example of this that codified this, we were Buddybuild, we hired a VP of engineering, I was walking through San Francisco and I called him up to talk to him about something.

And he says, hey boss, how's it going? And I stopped him right there and I was like, I'm not your boss. I am, but I never want you to think about me that way. Like we work together. You have a slightly different job than I do, but like you have to work together to do this. So the opinions, I mean all the time, like, uh, we, I'll have an opinion on some product thing or some go to market thing or what have you, and typically if I've thought a lot about it, it should be right. We hired this woman who started in December and she's been hammering away at a bunch of like the the kind of community management marketing stuff.

I've been working quite close together on a bunch of like just refining it over and over and over. As I focused on something else for a little bit and came back, my predisposition is like, well, what does this person know? Like she's hasn't been thinking about this problem for as long as I have so on and so forth.

But she demonstrated very clearly that she's evolved it and made it a lot better. So I was like, ah, that's an opinion that like, I can now, I respect and I learned or what have you. So yeah.

Brett: Maybe on a related note, a fun place to end is you started to talk about this one when we began the conversation, but you've worked with all sorts of really interesting, smart, talented people. Is there one, or are there a few people that has imprinted themself on you in a more substantial way that like, there's a lot of residue That's just kind of stuck with you over a long period of time.

And if so, like, is there an interesting story or anecdote that kind of brings it to life?

Dennis: I got to spend some time with Stuart when we were building Buddybuild, and seeing how thoughtful and deliberate he was in building Slack as kind of a bystander there. And like some of the insights, I think that's been remarkable.

There's a craftsmanship to building software and I appreciate the fact that there's an art and a science to it and, you know, he'd be one of the people that I think does a really good job with it. He feels the pain. When the software is bad, either mine or someone else's that I use, it's a visceral thing for me.

It's not like an academic thing. It's profane laced outbursts in the middle of the office, being like, how could you possibly build something that you think that's acceptable? Like, this is the thing. 

Brett: Lloyd has this idea from who started Looker that, great software is an act of empathy.

Dennis: You want to get more of those messages that say, thank you for your software, I get to spend time with my family. Like, that is insane.

Brett: Well, thanks for joining us. This is great.

Dennis: Thanks for having me.