Matt Arbesfeld is the cofounder of LogRocket, a funded startup designed to help front-end developers debug issues faster. Host Adam Conrad spoke with Arbesfeld about building a software startup as a software engineer. Arbesfeld explained about finding his cofounder, solving problems you’re passionate about, hiring your first team members, and the downsides of starting a company.
Show Notes
Related Links
- Episode 261: David Heinemeier Hansson on the State of Rails, Monoliths, and More
- Episode 373: Joel Spolsky on Startups: Growth, and Valuation
- Episode 350: Vivek Ravisankar on HackerRank
Transcript
Transcript brought to you by IEEE Software
This transcript was automatically generated. To suggest improvements in the text, please contact [email protected].
Adam Conrad 00:00:16 Welcome to software engineering radio. I’m your host, Adam Conrad. And today my guest is Matt Arbesfeld, a co-founder at LogRocket and a Thiel fellow. LogRocket was born out of he and his co-founder’s frustration with troubleshooting vague customer-reported issues. Prior to LogRocket. Matt was a student at MIT studying computer science today. Matt will be sharing with us his story of how he founded LogRocket, how others can find their own companies, pivot to different ideas and how to successfully launch a product. Matt, thank you so much for joining us today on software engineering radio.
Matt Arbesfeld Thanks Adam. It’s great to be talking to you today.
Adam Conrad Very exciting. I’m actually an entrepreneur myself and really excited to meet with someone who is much younger than me and who is kicking things right out the Gates with our first company. So that’s actually where I wanted it to start with, you know, as a disclaimer, what should we know about survivorship bias here? Because this is a sample size of one you haven’t started really, at least any companies we know of prior to this, because you basically came from right from MIT right into this project. So I’d love to know more sort of anything we should keep in mind here for folks listening in on sort of about your experience.
Matt Arbesfeld Yeah. I mean, this actually is my third company. Okay. Incorporate a company, so to speak. And then even within this company, this was probably the ninth idea that we came up with. So it took many, many tries to really figure out something that worked and people cared about. So definitely there’s a survivorship bias at play where, you know, a lot of what I’ve learned may not apply or is luck. But then I had a lot of failure as well. So don’t want to discourage people if it takes many tries to like get something off the ground.
Adam Conrad Absolutely. So I’m actually curious what were the other two? Because I, I, the only ones I knew of was LogRocket.
Matt Arbesfeld LogRocket is our first C corporation. So in college, my friends that I actually started two different LLCs for mobile apps that we were building. So one was a, a mobile app that let you basically have multiple phones that played audio at the same time together. And it was an app we launched. And then we had another company that did a mobile game. So basically nothing kind of in the same realm of law, docket of business software, but basically to build some mobile games on the side in college.
Adam Conrad Okay. And those, you didn’t raise any money for, it was just, you wanted to incorporate so that you could have a real name under it when you, when you publish those games exactly.
Matt Arbesfeld 00:02:34 And legal protection, you know, so that the company in case, you know, we did something wrong by accident. The company could kind of bear the brunt of that.
Adam Conrad That’s smart, even in college, knowing to lawyer up to take care of yourselves there. That’s cool. Great. So as you just mentioned, you got your start sort of at MIT, you were studying computer science. Were you taking any entrepreneurship classes? Like what really started that, that seed for you to go, you know what I think I want to, I don’t want to work at Google. I want to do my own thing.
Matt Arbesfeld It really, the first entrepreneurship experience went all the way back to third grade for me. And so I grew up, you know, my best friend and I Ben, who ended up starting a company LogRocket. We in third grade identified a problem where you have these desks in, in class where, when you lifted the desk, if you had anything on top of the desk at all, fall off and you had to lift it to get to your textbooks and your pencils. So we basically invented this contraption that would sit on your desk, a duct tape, sort of water bottle holder. You put it in your water bottle and when you lifted your desk, it wouldn’t fall off. So that was really our first foray into entrepreneurship, where we would go after school, we would charge people their lunch to basically install one of these. And that’s kind of where we, we first started entrepreneurship. You know, we’d never really took classes for entrepreneurship. Really. I focused a lot on project classes where basically the whole time, it was just open-ended. You didn’t have to sit through lectures or do homework. You would just build something. That’s really where I focus a lot of my time and found that the most rewarding I did the exam,
Adam Conrad Same thing. I also studied computer science in college, and I a hundred percent agree with you that the project based classes always felt the most rewarding because you feel like you’re building something and it feels the most real to what you’d really want to be doing. And I would imagine, especially for someone who is entrepreneurial and thinking of ideas, the project based classes sort of give you two really good signals. The first is just building stuff helps you think of other ideas. And then on top of that, you get to work with other people, which is a way in to help you find co-founders. So I’d imagined that that was something where intentionally or not, you’re sort of in these classes going, Oh, I kind of like working with this person. I wonder if they’re also interested in mobile game development. Is that something you found when you were taking those project classes?
Matt Arbesfeld 00:04:47 Definitely. Yeah. And you find what skill sets are really, you know, you pair well with and sort of where you’re like, Oh, wow, there’s this whole thing. I had a friend in college who’s really, really good at design and just found, wow, what I was paired with a really, really strong designer. We could create incredible things. So I think that really helps identify who in kind of what the right pieces were to form your, your founding team. So, and that being that person, I started long market with, I didn’t go to college, so we didn’t do those projects, but definitely learned a lot from all those experiences of building things.
Matt ArbesfeldAdam Conrad 00:05:20 You go to high school together, it looked like from the research I was doing that you both went to the same high school, maybe not the same year. So how did, how did you two meet?
Matt Arbesfeld 00:05:28 Yeah, we went to preschool together. We were born, uh, you know, about a month apart. Our parents were best friends. So I really grew up together and went to elementary school, middle school, high school. And I have stayed in touch obviously over the years, but really, you know, Ben was the person who introduced me to technology and building things. So you have a lot to thank for just introducing me to this whole world.
Adam Conrad 00:05:50 And remind me, was Ben your co-founder if you will, of the, the desk tool that you had in third grade.
Matt Arbesfeld Oh yeah. Yep.
Adam Conrad So you guys had a history of entrepreneurship from a very early age, so I’m, I’m imagining that, you know, you do go to different colleges, you start different things when you sort of come back to it being like, Hey, I’ve been thinking about this idea. And he says, yeah, I’ve been thinking about that idea too. Is that kind of how it started? How did you guys sort of reconnect?
Matt Arbesfeld 00:06:14 Yeah, so we were actually in San Francisco for the summer, so it was a summer internship and we decided, Oh, let’s live together. You know, I was interning at a, a startup. Ben was at Google actually at the time. And, you know, just after work, we’d ideate on ideas. Some days we would just make pickles or some kind of random activity. And other days we’d say, Oh, you know, what are some problems that we’d love to solve? It would be interesting to tackle. So that was really the original Genesis of LogRocket was going home ideating. We had a giant whiteboard that we brought in and just, you know, brain storming together. Okay.
Adam Conrad 00:06:49 So before I dive into that topic, cause I’m very interested to hear those sort of early days while you’re, because even just you saying that it sounds exactly like this sort of Silicon Valley ideal scenario where you’re, you’re not in a garage, but you’re in the living room you’re sort of ideating on topics. It just sounds very acquainted to the, the idea of what a software engineer thinks about when starting a company. But before I get to that, I did think about this one quote I saw from Paul Graham one time where he talked about, if he could go back to school, he would study physics instead of computer science, because he felt like he got enough computer science education through the programming he did on his own cause he was already passionate about it, but he felt like if he had a different degree than it would allow him to sort of explore opportunities and ideas more broadly. So if you could do it again, would you still stick with the CS degree at MIT or would you have studied something else?
Matt Arbesfeld 00:07:36 I think nowadays, especially with so many ideas in just pure software engineering are covered by software engineers, that it makes a lot of sense to be exposed to a whole new area. I don’t know if physics is it, but like building spaceships or building, you know, airplanes or chemical engineering. I think there’s a lot of different areas that need software developers to, to contribute to. So I think that makes sense. I think the other side also, and tell me I’ve benefited from a lot were humanities classes, like I think a big part of starting a company is communication, writing ability to be able to craft a story and a pitch to investors. So I think like that, you know, I didn’t do as much at MIT. MIT is not really known for its, um, English classes, but especially growing up, you know, spent a lot of time just writing essays and communication was super valuable for starting a company as well.
Adam Conrad 00:08:29 The spectrum. Do you feel that folks who want to start a company or want to join a startup need a CS degree? Like, are you hiring for people that don’t have CS degrees for software positions?
Matt Arbesfeld 00:08:39 For sure. Yeah. I think, you know, you don’t necessarily have to know how assembly language works to, to write great software and many of our best offer engineers at LogRocket don’t have CS degrees or even bachelor’s degrees. So a hundred percent, I, you can be an amazing software engineer without a CS background or any real formal education for them. Right.
Adam Conrad 00:09:00 So if you don’t have friends that you’ve known for years and you don’t have college, where are people finding their ideas? Like, so you had said that you had your first idea with your best friend in third grade, where are there areas that folks could be looking for startup ideas if they want to start thinking about starting their own company?
Matt Arbesfeld 00:09:19 Original way, we came up with log DACA was an idea I had working at this startup. So I think the best place is by exposing yourself to a set of problems that are unique, you know, and that might mean, you know, say if you’re an engineer, maybe try working closer with a sales and marketing folks for a couple quarters or a couple years, or join a chemical engineering company, broadening your horizon of the types of problems you work on, I think can really help expose you to different sets of ideas. I’d say the other thing is building relationships with a diverse set of people. So I think people are spreading out a lot more from the Valley, but you know, being in Boston, get a chance to talk to a lot more people doing PhDs or people pursuing finance or are, you know, so I think kind of having communities of different types of people has really helped broaden sort of my purview into the world and sort of the types of problems that that should be solved.
Adam Conrad 00:10:15 Yeah. And you touched on a good point about exposing yourself to a lot of different people and a lot of different ideas. That sounds like a good idea for finding a co-founder. If you didn’t have somebody who was your best friend since preschool, how would you have gone about finding a founder? What it has been through something like your coworkers, how, how would you have suss that out
Matt Arbesfeld 00:10:35 Best places, likely joining a fast growing business with great, great people. You know, like one of our goals at LogRocket is that eventually when we go public someday or who knows what happens, that will be a place that people have formed friendships and they can go on to start companies have their own. And I think you’ve see that now with a lot of the great Silicon Valley companies like Airbnb and Coinbase, that people are taking the ideas they learn from those companies and starting their own, especially it’s hard with COVID, but you know, I know there’s a lot of different types of meetup groups to go to where you can form friendships there or even just reconnecting with college friends. You know, I I’ve stayed in touch with probably like five college friends, but there’s so many people that, you know, reconnecting with them, seeing what they’re doing. I think that’s a great, great source of people as well.
Adam Conrad You, for your sake, you had a close friend and you guys stayed in touch over many years. And so catching us back up to the story you are at your apartment that you’re at for the summer while you’re both at your internships and you start your idea for LogRocket. My first question for you is, okay, so you’ve got that first idea. How did you know it had legs because it’s one thing to go, Oh, I’ve got an idea. And so many times, I mean, I think of ideas and I immediately either shut them down or I post them to somebody else. And I say, what do you think about that idea? And they think of all the reasons why it’s not going to work. So how did you know that this one was good enough that it shut off your brain saying, Oh, I dunno if this is a good idea to actually we should keep going with it.
Matt Arbesfeld 00:12:00 Yeah. You know, ultimately the only ideas that have ever worked for me are ones that I myself have experienced the problem intimately. And no I’m not so unique in this world. It must be at least a thousand. It probably a hundred thousand people just like me who are experiencing the same exact problem. You know, actually the original idea that we did in the living room at 10:00 PM, brainstorming wasn’t longer cut at all. It was a whole different idea about updating apps that are in the app store automatically. But I had built enough mobile apps to know that that was a really important problem to solve. And there were going to be people and companies that would invest in that. So, you know, anytime that I’ve tried doing an idea that I myself would not personally use or benefit from, I haven’t been successful, but when I focused on pain points that I myself know really well, it’s worked out. So obviously that’s like a survivorship bias, but that has kind of been my guiding light for how to choose ideas and make decisions there.
Adam Conrad 00:12:55 Right. But that’s an important point that you noticed the opportunity where either you could run into failure later on or that to your point of pursuing your passion really, and finding a problem that you thought was worth solving. You quickly found out pretty early that, that wasn’t going to work. And in fact, you said you pivoted nine times.
Matt Arbesfeld 00:13:13 Yep. So that, that first idea we actually, we actually got that up to about 5,000 MRR in a couple of months. So we started it kind of midway through the summer. And by the end of the summer, we had almost a hundred customers, but we didn’t feel like that idea had kind of a long-term potential because we were essentially circumventing the Apple app store review process. So there were sort of like legal and compliance issues that we felt could affect us with that particular project. Um, and we just didn’t feel like we want to invest years of our life just to have one day we wake up and Apple has shut us down. So that’s when we started pivoting. And by the time we started pivoting, we had actually already raised around a venture capital. So we had some time to kind of figure it out and go through this process of trying different ideas until eventually we tried a bunch of ideas that we ourselves didn’t really have a personal connection with. So we went all the way back to another pain point that I had. That was a founding idea behind LogRocket. And that’s what ended up working for us.
Adam Conrad 00:14:16 I want to dive into that one particular piece you said, which was we had raised money. And when we ended up pivoting leader, this is super fascinating to me. I’d love to learn. How do you convince someone to give you money on a thing that could change any day? That, to me seems remarkable, honestly, that you can convince someone that you can, you can give them money for something that you don’t know if it, it totally has legs. So how are you able to convince folks in those very early stages of fundraising?
Matt Arbesfeld 00:14:42 I think a lot of that comes back to storytelling and especially early on, investors are not necessarily saying, this is the idea that’s going to go big. They’re really in you as a person, as a team. And we knew we were going to have to make some changes to, to build a long-term business. So our investor kind of was on board with that idea that it was going to take us some iterations to get to where we need to be. And it was also a small amount of money for this particular investor matrix partners. You know, they have a half billion dollar fund. If they’re investing 0.1% or 0.01% in one company, they’re not really investing that much to, to this idea. So it’s almost like if we go buy a $15 entree and it’s really bad, you don’t feel that upset about it. You know, because it’s, it’s a small enough amount of money relative to other things. But if you buy a $150,000 house and it’s really a garbage house, you feel really bad about it. So I think there’s like an order of magnitude where below a certain point, the investors don’t get affected by losing that money. And that, that was helpful that we found investors with deep enough pockets that they weren’t upset that the original idea didn’t work. And now we’re going to have to try some different things. Yeah.
Adam Conrad 00:15:52 And you touched on a really good point, which for the folks listening at home, I’ve now collected already three awesome nuggets of advice, which I’ve heard from many other folks. And I think you’re just saying exactly the same thing, which only adds validity to it.
Matt Arbesfeld You know, the first one is solve a problem that you want to solve yourself. Having passion around the problems is a huge thing. I’ve heard that so many times, but it’s great to hear you say it because it really puts validity to it. The second thing is about finding people you really work well with, whether it’s through, you know, finding a diverse group of people where you can share ideas or spread out the duties. That’s a really good thing.
Adam Conrad But as you had said, you know, you spend a lot of time with Ben before you ever started a company.
So you had a lot of trust in each other. And then the third thing which you just mentioned was that you don’t necessarily have to have a good idea so much as you have to have a great team that people can invest in because they’re not investing money necessarily in the idea they’re investing money in the people to turn that idea into reality. So, and I’ve heard that many times as well, too, that if you have a great team, it can almost supersede a product in that sense, because if you have really good people, you can effectively build anything, right. And in lots of companies, pivot for folks listening at home, those, those are definitely things you should take away from thinking about, you know, any roadblocks you might think about and trying to start a company. But I wanted to get back to the point about the pivots. So you obviously had already raised money before you pivoted onto the actual LogRocket idea. How did you know that you needed to stop pivoting that you said, okay, this is the LogRocket we want to build. And this is something that we can actually take and run.
Matt Arbesfeld 00:17:23 I remember probably the night before we came up with the log market idea, we were working on another idea we had been up for all night and probably multiple days in a row preparing for this presentation for this other idea. And we gave the presentation and this was to one of the investors. And they were like, this is just a bad idea, guys, like need to move on. After that, we said, we’re going to have a one more idea. This is the one we’re going to go all in on, you know, it’s this or bust. And we went back to the whiteboard. We thought through, first of all, what are the changes in the world that will enable a new company to be built? And then we came up with this change around front end development around more and more moving to front end and react and angular and view. We said, that’s going to be the change. What are the problems that are now in that world that we ourselves know intimately well. And that’s where we came up with the troubleshooting bogs idea, I think going back, and this is a big value in, in, in our company today is thinking from first principles, we took a step all the way back to the beginning and said like, what’s the change? What’s the problem. And then once we saw that and we said, we’re not going to change. We’re going to go deep in this idea. And we spent the next year just building it, honestly, you know, people say a launch early, but we spent a whole year just from that point, building it, you know, making the marketing website, preparing for launch and ultimately having conviction in that idea, let us go through that full year without needing to pivot or question ourselves. But that was going to be the last idea that, you know, if that one failed, we were probably going to pack it up and you know, try to apply for a job at Google. It’s amazing
Adam Conrad 00:18:59 Using that. So many companies, I mean, I even just read about Airbnb because they went public. And then Paul Graham released an article about how tenacious Airbnb was in the very early days. And they said something very similar, which was, we essentially were down to our last bit of money. They max out a bunch of credit cards and then they had to start making cereal boxes for themselves just to be able to afford to pay rent and Airbnb. They were on its literal last legs. It’s amazing to me that these companies, that they have, you know, their last legs and they able to turn it around. So what was your first signal for success now you’re on your ninth pivots, but now this is the one. So what gave you the indication that this time was for real?
Matt Arbesfeld 00:19:39 You know, I wish I could say we had everyone say, Oh, this is a great idea. And we’d go to conference because we would go to react conferences and show people the idea and people would say, Oh, that’s cool. You know, it’s interesting, you know, how about this or this product? Or, you know, can’t you just solve it this way. So I wouldn’t say there was a ton of external validation. Like it’s still for a very long time, just came internally. I think where we really started believing was my co-founder Ben implemented Stripe into our product where you could basically, people could just sign up and self-serve to buy sessions with us. And the first time a customer without ever talking to us, you know, signed up, went through the 14 day trial and put in their credit card that was like, wow, this now seems like it’s going to work. Like people are going to do this and sign up and use a product. And that was the moment where we really believed we had something, but it took a year just to get to that point. It was a year where we didn’t know if it was going to work. And sometimes we say, Oh, you know, maybe people just don’t care enough.
Adam Conrad Yeah. And I think that goes back to your passion point because the truth is even, that’s not really truly success, right? Like, just because you have one customer doesn’t mean you can afford rent or that you can pay the bills.
Matt Arbesfeld Right. But just having that faith that’s okay. We made it all the way this far to someone actually paying us money without having to basically like twist their arm to get the money from them and they want to use it.
Adam Conrad I think that’s a really good validation point and sort of a good next step. So now that we have landed on this, I actually just want to know, like for the audience, what is LogRocket like the, the ninth iteration version of lock rocket that you landed on? What, what does it do today?
Matt Arbesfeld 00:21:11 So the way I usually describe it as we’re a front end monitoring solution. So we basically capture everything users do on the front end of your application. And there’s two main ways people use it. One is, if someone has an issue on your site that you can’t reproduce, you’re able to go search for that person and basically watch a video of their experience and see technically what happened to them. And then the other way is that we’re actually proactively surface issues that people are having. So if they’re trying to click the add to cart button and nothing’s happening, we’re running machine learning algorithms to actually detect that negative experience and surface that to you. So you can be more proactive about fixing user experience issues.
Adam Conrad 00:21:51 And was that what was on launch day? I mean, obviously the product is iterated many times since then, but I’m just curious what subset of that was the first one that people said, Oh, I want to do the 14 day trial and then pay you guys money. Yeah,
Matt Arbesfeld 00:22:05 It was really, really simple. It was just a user would have a problem and you could go find their session. So it was from the beginning and actually the first three to four years, we were focused on just a single problem. And I remember I’ve repeated this again. And again, to the team of, we help developers solve bugs faster. Like that’s all we would say we had on the whiteboard. People would come to us with other types of feature requests and product managers, but we just kept focused on that single problem, which was around when a user submits a support ticket. How do you figure out what happens at user and solve their issue as quickly as well?
Adam Conrad 00:22:38 Right. And when you say session, are you actually talking about someone can pull up a video and they can actually see what the person did.
Matt Arbesfeld 00:22:44 Exactly. Yeah. So we actually basically instrument the Dom from the front end and then capture all the changes that happen in the user interface. And then we take those and we recreate what the user saw. So it’s almost like you’re watching a video and sitting over their shoulder, you know, seeing where their mouse is, what they’re clicking on. And so you’re able to go back and that’s especially helpful when you have less technical end users who can’t send videos or, or really communicate what they did. You’re able to just go pull up their session and see their experience.
Adam Conrad 00:23:12 Can you actually see the real browser? Cause I feel like I have experienced in front of an eye and my professional career have mostly been in front of developers. So a lot of this speaks to me. And one of the first things I’m thinking of is, Oh, well I need to actually know if they’re in IAE or they’re in Chrome or they’re on their phone because right. Like your point about being able to sit behind their shoulder. I mean, that is exactly what developers do for themselves, right? If someone’s trying to debug a problem and they ask the senior engineer to come over to their desk, that’s exactly what they do. They look at their screen and see what happens. So I think that’s, uh, speaking to your point of solving developer problems and helping debug bugs faster. My first thought is like, Oh, well I E especially when you started LogRocket, it was the bane of everyone’s existence on the front end. Could you get that kind of level of granularity on the sessions?
Matt Arbesfeld 00:23:56 Yeah, so we support 11 and newer. So that’s definitely a big use case mobile, you know, people go offline and you don’t know, maybe they were just offline and they had network connectivity issues to be able to actually see the browser they’re on, see all the network details, all the console, logs and errors. We also capture browser performance information like memory usage, CPU usage, and crashes. So there’s a lot of information there just like, yeah, just like if you open the dev tools, sitting with someone
Adam Conrad 00:24:26 I can imagine as a developer, you’ve got that slogan about helping developers solve bugs faster. And I can see like a video on your homepage of walking through a session. Is that how the launch actually started? Cause you, you mentioned that you’re a big part of your time was spent just creating and crafting this ideal setup so that when you launched, you we’d have everything, all your ducks in a row. So what did the actual launch look like? What was that sort of first version that you put up on, on the side?
Matt Arbesfeld 00:24:54 Right. That’s exactly what it was basically walked through. The exact problem that we had as developers tried to communicate as clearly as possible. And then I believe the first version was just an email signup where you can put in your email to sign up, to know when we launched. And then we would reach out for beta users, but we coordinated a big launch on hacker news and product hunt. And you reach out to all our old users, had all our friends ready to kind of help vote our contents. So we did a big coordinated push and it ended up, I think driving probably the first thousands, you know, emails to us. And then we kind of use those two to get off the ground.
Adam Conrad How did you know you’re ready?
Matt Arbesfeld Once we hit the one year mark, And I think he was starting to kind of wonder like, okay, maybe it’s time. We should really kind of push it out. You know, it’s, it’s dragging on a bit and we weren’t actually completely ready in terms of the products. It wasn’t quite there, but we knew it was maybe two, three months away. And this was going to be a good forcing function to, you know, we could set a date. This is when we’re going to launch by, we have to get ready. And then we set that goal and it was a lot of work, obviously, you know, we would probably code from 10:00 AM to 6:00 PM and then we’d go back. We were living together, we’d cook blue apron. And then in the night we’d work on the marketing website. So a lot of work to get there, but kind of having that date where we had to launch by was really the forcing function.
Adam Conrad 00:26:18 I wanted to ask, you know, when you, you hear about how the entrepreneur that just started over a weekend and you’ve got to launch early, and if there’s that old advice of, if you launch, when you’re comfortable, you’ve launched too late. And it sounds like you were ideating on it for quite a while and you just basically said, okay, one year, is this too much?
Matt Arbesfeld We just, it almost sounds like it was arbitrary other than the fact that it’s like, okay, it’s been a calendar year and we really need to get our ducks in a row. So you weren’t quite ready, but ready enough.
Adam Conrad So how did you know, to carefully coordinate with hacker news and product hunt? What was your sort of strategy behind that?
Matt Arbesfeld 00:26:50 At that point, didn’t really know another way to get users besides post on hacker news. I thought that’s where that’s the only place you could find products with hacker news in my minds. So that was really our only option of how do we get users. And we had done some coordinated launches on the past for hacker news and those, and had seen success. So that’s how it came about. You know, since then we’ve kind of developed new marketing channels and strategies, but I think it’s a great way to get started as a hacker news post, especially back in 20, I guess this was 2016, 2017, where hacker news was really vibrant kind of before Reddit.
Adam Conrad Yeah. I mean this year felt like it was five years so that, you know, 2016 feels like it was two decades ago so easily and a lot of things have changed. So if someone is listening today, what are those things you’ve learned? I guess it’s really two questions. One, do you actually need to have a big launch event? And if you don’t, what things would you teach others today that they could learn from your mistakes as to how to successfully launch beyond just posting a hacker news?
Matt Arbesfeld 00:27:51 You don’t necessarily need a big launch, but you might as well because it’s one of those types of things you really only do it once or twice. It works reasonably well to generate interest. So I definitely encourage people to do it on top of that though. I I’d say really think just as much as you think about your product, think about the distribution channels. And that’s something we thought about from the beginning of it’s great. If you can build a product, but if no one can find it, it’s like if a tree falls in the woods and no one sees it, did the treatment fall. So equally important to the product is the distribution. And just as we were thinking about developing this product, we were thinking about our marketing strategy and how we would bring it, bring it to market. And specifically we were focused on building out our front end blog for front end developers. So that would be my other advices depends on the market and what you’re going after. But, um, really when you’re putting together your plan for what to build, unless you have a way to reach users and acquire them, it’s not really a product
Adam Conrad And there’s benefits and drawbacks, by the way, to your, to your point about knowing your audience and finding the right distribution channels, what I’ve found as a plus for somebody like yourself is you are a developer and you’re speaking to developers. So you know them better than anybody else because you are one. So you have that advantage of being a developer tool, right? The downside, I think, which I’m sure, you know, right off the top of your head is, well, developers have put a lot of scrutiny on to things. So you know, your audience and you also know that your audience is putting a very sharp lens over what you’re building. So there’s benefits and drawbacks to every audience. But I think the resonating messages, if you are not your target audience, you have to find them and go where they go, right? Because Akron news is for hackers. It’s for software developers, it’s for people in product, right. But hacker news may not be the right place if you’re building say something for engineers, like you said. So do you have any advice on sort of finding the right people for those distribution channels?
Matt Arbesfeld 00:29:47 It, it depends on what you’re going after. I think we’ve done a content strategy here at LogRocket, because as front end developers, I would be searching react component props day in and day out. So half of your job as developers just searching for things, I would kind of think through the same for a chemical engineer. Um, actually my, my partner’s a chemical engineer, so I would just ask her, but whoever your audience is, ask what their day-to-day looks like, are they searching for things? Do they read their emails from people’s trying to sell them things? Do they pick up their phone? What does the day-to-day look like of the person you’re trying to sell to? And then how do you insert yourself in there with your product? But first and foremost, it has to solve an important problem for them. It has to have an aha moment where they say, wow, this is going to solve a big problem for me, but definitely kind of thinking about their day to day and where you can insert yourself. And in some cases there’s nowhere to insert yourself that you may not want to build for that audience. If you can’t actually get in front of them in a efficient way.
Adam Conrad 00:30:44 That’s good advice. And I think you brought up a good point, which was solve an important problem for them. I would add is a big enough problem that they’re willing to hand you a check, because as you said, sort of aha moment for you guys was that you had someone who you didn’t know who found you, who signed up for the 14 day trial and then paid money. So that’s a really strong signal that you’re onto something there. So there’s a big gap between your first checks from matrix, the launch of that first sort of version of pivot, nine of LogRocket to today where you’ve raised $30 million. So I’d love to dive in more to that sort of fundraising piece. We sort of touched on it a little bit earlier, but I’m really curious, how did you get to that next level of fundraising? So you’ve got your initial investments, you built something people actually want to pay for it. You’ve got a low version of LogRocket that, you know, you want to pursue. How’d you get to that next step?
Matt Arbesfeld 00:31:35 Yeah. It’s interesting. Once you have that first customer, then it’s all about customers. It’s all about how many customers are you solving and how much are they paying you is really the function of, for better or worse. That’s the amount of value you’re creating for, for people. So to get to the next phase, we were laser focused on hitting our revenue goals and growing our marketing and in our customer base. And so, you know, that first customer, it took us about a year to reach maybe a hundred customers somewhere like that. And at that point that’s when we started getting momentum to point to, Hey, if we start raising money, we can invest more in sales and marketing and more in engineering and keep grow this even faster. And at each point it just makes sense. You know, if we had that extra, extra dollar to spend, we’re going to create so much value with it that it makes sense to keep fundraising and keep growing what we have. But, but yeah, really after our first customer, it all became about growth.
Adam Conrad 00:32:29 You earlier said, you mentioned MRR for the folks listening. What does that mean? And how did that influence, you know, when you knew you could raise more money because a lot of people talk about, especially in subscription services, that MRR is a very key indicator for being able to either raise more money or show growth.
Matt Arbesfeld 00:32:45 Yep. So MRR is monthly recurring revenue. So that’s essentially, if you have a customer paying you nine, nine a month, your MRR is $99. And what investors are looking for is not necessarily the absolute amount of MRR you have, but the growth rate of it, because if you could take MRR from a thousand to 10,000 a month, they say, wow, if they can do that in a month, imagine what they can do in a year. And two years and three years, really the biggest focus for us was making sure each month we were adding more MRR than we did in the previous month. Right?
Adam Conrad 00:33:19 And so then once you started to see reasonable growth numbers or numbers that you felt were predictable, how did that translate into, okay, we need to raise X amount of money. How do you pick that number?
Matt Arbesfeld 00:33:29 It kind of turns into a financial problem of we have this marketing channel, be it, you know, ad spends or content. And we know it’s going to translate into this amount of dollars of MRR. So my co-founder Ben actually now runs our finance function, basically put together these huge spreadsheets of we invested all the here, we’re going to get out X MRR, you know, eight months later or whatever it is. And basically we use those spreadsheets to, to determine how much we need to raise how much runway that’s going to give us as a business, if we’re burning money and then, you know, using head count and the other costs that we have to basically figure out how much we’re burning each, each month in each quarter to know when we have to, to raise next. So a lot of it is basically turns into a financial problem and you know, the input is expenses and the output is revenue on the other side.
Adam Conrad 00:34:20 Okay. That makes sense. And the next question I have on top of that is now, you know, how much you need to raise and is it true that 18 months runway is about what you want to target? If you’re a founder and you’re looking to figure out how much do I need to raise an advance to, to support that
Matt Arbesfeld 00:34:35 Potentially, you know, I think it varies Ben and I have always targeted longer runway than that, just because it always takes longer for things to work than you think they will. And I think our first ideas took us two years to reach our first customer. So if we had just raised 18 months of runway, we wouldn’t exist right now. So I always encourage people raise a bit more than they think, if you can get away with it, raise 24 months of runway or 30 months or 36, even, especially now with so much money in the VC markets. I think you’re best off. If you can raise enough that maybe 18 months, you need another year to kind of get to the next stage, you’re better off as the entrepreneur, if you’ve raised that extra money prior.
Adam Conrad 00:35:16 And I think the really important thing, it just, it was funny because I’m looking at the notes here and it’s, you’ve raised $30 million and you have to understand there’s this like whole thing in the entrepreneurial VC world where you see the dollar amount, right? Like company raised X amount of money. And people just like, assume that’s like, Oh, well Matt’s a millionaire. Now. He just raised $30 million. And he and his co-founder split it halfway, but it’s absolutely not that way. Right? Like all that money has to last that entire time. So if you’re raising $30 million in one go for three years, you can only spend 10 million a year. And on top of that, you have to support all the people that you are employing. And as you mentioned earlier, I think you said a lot of it was going into marketing and certain sales channels. So tell me about, you know, what actually happens when you get that money, where are you spending it on and how do you ensure that you can support the system that you’re trying to build?
Matt Arbesfeld 00:36:04 Yeah, that’s a great point. It’s really to grow the team. And there’s so many expenses that go into a business. You look at rents for the real estate. If you’re an employee head count expenses, you’re getting lunches for employees. There’s so many different expenses that you CA that you figure into. And that, like you said, you have to have to last for years and years, plus all the growth that you want to achieve. So all the new people you want to hire the recruiting efforts, the recruiting resources needed to hire those people. So basically it’s, you know, again, another big spreadsheet, everything kind of just turns into spreadsheets and engineering problems at the end of the day and sort of cash management is one of those problems. And then on the other side, the best case scenario is that you don’t need to spend that money, that your customers can fuel your growth. So a lot of our focus has been also of identifying inexpensive channels, where our customers, we can take, you know, the customer revenue and turn that into, reduce the amount that we’re, we’re burning of that 30 million and use that more as kind of a backstop to achieve growth.
Adam Conrad 00:37:02 Is that a weird pivot? Cause I feel like for folks that are brand new to starting a company, you know, they’re thinking about like being in the garage, writing code until 4:00 AM in the morning, and now you’re talking about spreadsheets and accounting and all this stuff that this is not the sort of main thing that software developers are thinking about when they’re starting their own software company. So is that, is that a weird transition? Like how much of your time do you spend on the things that are not software related?
Matt Arbesfeld 00:37:28 I’d say a hundred percent now if I try to code in Beverly break something and people get the engineers on my team gets scared. If I ever try to make a code change or even suggest code changes. So I’ve removed myself from that is definitely a transition where in the early days I write code 95% of the time to now 0%. But at the end of the day, you’re still building systems and you’re solving problems. And I think if you’re starting company, it’s important that you, or at least your co-founder is passionate just about problem solving in general, because past that kind of initial product market fit, or even during finding product market fit. So much of it is not just about coding. It’s about customer discovery and marketing and idea validation and fundraising. So I think it’s almost more important to be excited about those things than it is about actually writing code. If you just want to write code, probably starting a company is not the best thing for you or at least make sure you have someone who’s interested in the actual profitability of a company on your team.
Adam Conrad 00:38:26 And that’s a really important point is balancing the positives with the, with the downside. So tell me about some of those problems that you faced along the way once, once you got LogRocket started, I mean, just because you raised a bunch of money and you got that initial validation, clearly it wasn’t just smooth sailing. So what kind of traps did you hit along the way?
Matt Arbesfeld 00:38:44 Yeah, obviously COVID um, is, uh, kind of, uh, threw everyone for a loop and we’ve really invested a lot in a great culture, a LogRocket and an in-person in-person team definitely was a transition for everyone to, to go to working alongside, you know, your dog and your, your one bedroom apartment, which is where I am now. But so I’d say, you know, COVID was definitely one of them as first-time founders. A lot of it was figuring out what we didn’t know that you learn a year or two years later. And you, you had wished like if only I had known that this was going to be a problem, I wish I’d addressed it a year earlier. So one of those things for us was like starting a sales team. I didn’t even know that sales was a thing. When we started the company and kind of waited too long to actually start a sales. I just thought developers just find products and they sign up for it and they put in their credit card. I didn’t realize I was the one sales guy just answering inner calm tickets. That was something where, okay, now I understand the value of sales and especially in larger companies, so much processes needed to procure a software, somebody stakeholders. So all those things that I didn’t know that I wish I had incorporated earlier on would have saved us so much time from the mistakes that we’ve made. So yeah, starting a sales team has won certain hiring decisions were I think so important to trust you God. And I think you alluded to this of find people that you really just enjoy working with and you sort of like, you’re excited to go in to work with them every day and being really deliberate about that as has been very important for us. So yeah, so, so many, so many mistakes along the way, but a lot of good learnings. How big is your team now? We’re running 80 people. So it’s still, still quite small.
Adam Conrad 00:40:22 You know, when you were really early, it was you and Ben, you could very much be hands on with every single hire. How do you feel now knowing that you can’t actually be there for every single interview for a team that’s almost a hundred people.
Matt Arbesfeld 00:40:34 Yeah, I feel pretty good. You know, I think that comes down to really hiring and promoting great leaders. And a lot of it comes down to the first few people we brought into the company and also, you know, recently does, we brought in having a lot of trust in their judgment. It gives you so much peace of mind to know, okay, they’re going to own their operational area. And I can have a lot of trust in what they’re doing. So I’d say that. And I’d also say really establishing clear, clear culture and values when making decisions and being consistent with the way we evaluate people. And look for greatness, I think has helped us, you know, but you don’t hit a hundred percent. There’s always misses. And you know, sometimes it doesn’t work out, but I’d say those two things have really helped us build consistency in hiring.
Adam Conrad 00:41:20 So we know what the roadblocks now were for you along this journey. When does it not a good idea to start a company? Like when should engineers not start companies because you, as you self said, you had two failed companies prior. So what were the signals that you should said, actually, you know, I don’t think this is the right company to start if at all,
Matt Arbesfeld 00:41:38 You know, honestly, knowing what I know now, there’s so many difficulties to starting a company that I think even for myself, my naive newness five years ago is what led us get to this point. Now I know how, how difficult it’s going to be. You almost have more reasons not to start a company than you do to actually start a company. Really good reasons to start a company are you enjoy, you enjoy building and you have a problem that, you know, has to be solved now. Like there’s a timely thing. If there’s not a, a change in the world that will support your idea and you don’t have personal conviction, I would, I would avoid starting a company maybe wait, and maybe kind of expose yourself to new experiences before diving in.
Adam Conrad 00:42:20 Do you think it’s binary though? Cause one thing I was, I’ve been really thinking about lately is there’s this whole group of sort of like indie hackers. Like there’s a whole site about this where, you know, you can start a small company in a you’re just constantly working nights or weekends. Like there are, uh, I think it was sidekick was one of them. It’s like a Ruby plugin that essentially was just open source. And then at some point they decided to release an enterprise version. Now that guy makes like 80 grand a month on just this sort of open-source side project. And I know Cypress, the front end framework just raised like $40 billion and that was purely an open-source project. So do you think there’s room for the companies that aren’t quite VC ready, but you know, they have enough passion and conviction that they want to keep going with it, you know, part-time?
Matt Arbesfeld 00:43:03 Definitely even LogRocket. Our Genesis was, we were in school and working on it between classes or in the back of the lecture hall, you know, on my laptop coding. I think you can make a lot of progress, especially if you don’t have other distractions in life. Yeah. I think it’s really hard when you have a job or supporting a family, you know, have all sorts of social activities. Like it can be very hard to balance more than two things I can, I think can be difficult, but if you’re able to, you know, have your job and also pursue a side project, I think that’s a great way to get started. Or if you’re in school, that’s a great time to explore ideas and you know, is it really a safe, a safe place to, to try things and fail that being said, I think, especially in developer tools, the space we’re in, there’s sort of a minimum bar of effort that you have to put in to get past what a developer can do themselves. So there was, I think certain ideas you have to commit to, or at least really invest, no you’re going to spend years and years working on this problem because you need to reach a certain level before it’s going to be valuable to other people.
Adam Conrad 00:44:03 It’s a really good point. And I appreciate the advice because as someone who has a child and a day job and is doing a side project like this, it sounds like I’ve got a big uphill battle to face there it’s not purely binary, right?
Matt Arbesfeld Like not everyone has to raise a hundred million dollars and become Facebook to become a successful company. There’s plenty of people who can do this on the side and make it work. But I think to your point, it’s just about balancing those things. Like if you have a product that requires a lot of Polish, just understand that it might take you several years to launch as opposed to, you know, some of these like fanciful ideas where they could launch it in a weekend. Well, that clearly has a very different bar for what the customer is looking for. Then, like you said, something in developer tool space.
Adam Conrad 00:44:42 So the last thing I kind of wanted to pivot on around this topic sort of, of like the part-time work is also around the topic of remote. You talked about this earlier, that COVID is obviously been a challenge for not just your team, obviously for everybody, it’s been a massive challenge and it’s almost impossible to avoid any podcasts without talking about COVID right now, how is your company pivoting on this? And we were talking about pivots. How are you pivoting in terms of being a remote organization? Is that something you want to continue with? Are you going to invest in a hybrid option? You mentioned that having an in-person culture was super important and super valuable for the folks that you have in your teams. How are you thinking about this going forward?
Matt Arbesfeld 00:45:19 We’re still seeing how things play out. That being said, I think, you know, a big part of why I’ve always loved working at startups is the group of people you’re working with and you just don’t get the same vibe or value from it when you’re just staring at zoom all day and talking to them. So I’m a big believer in at least having the option. Like I consider work, like are the places I enjoy going and talking to people and having lunch and you know, hanging out or I’m a big believer in at least having that as an option. But really since the early days we believed in, sometimes you just need a place to focus. And I think home is a great, or you know, your home office is a great place to be able to just like sit down and solve a problem, no distractions, but I am a big believer in having a place to go. That’s like a community and something you enjoy waking up and commuting in for.
Adam Conrad 00:46:08 So if a founder is starting today, what advice would you give them about the options they have? Certainly remote has become more prevalent at companies than it was a year ago or even six months ago. If you were starting today, how would you think about that equation?
Matt Arbesfeld 00:46:21 Yeah, 100%. I would build your team in person still. I think especially the early team. It just fun to have a small team that you enjoy going into the office with and talking and hanging out after work. That’s where so many of the best ideas come from when you’re a small group of people. I think once you get past a certain point, then you can start to say like maybe the talent when we hire a distributed workforce will be worth the kind of loss of that culture. And that kind of comes down to the specific business and types of people you need and where the talent is. And also where you yourself are located. I think having like the first 10 people be able to be in an office together and hang out and do fun things is at least was super valuable for Austin. I think it’d be hard to recreate that over zoom,
Adam Conrad 00:47:10 All the competition out there, which is never ending in the software space. How do you find those first 10 people? Because I would imagine you’re competing against salaries like Google. You’re also competing against missions of other small startups. So there’s a lot of competition out there. And on top of that, you have to find really great people because as you said earlier, it’s those first few people that really set the tone
Matt Arbesfeld 00:47:30 For the rest of the culture. I mean, that sounds like it’s really arduous because you’ve got so many other factors playing into getting those first few people. How did you make that work? We were really careful about not hiring until we had traction with our idea and some amount of conviction. I think it, it helped that we’re building a developer tool. So developers we’d reach out to developers say, Hey, we’re hiring. You know, we have some initial customer traction and they would say, this is a product I would use and love. And that really helped us initially. So I think a lot of it comes down to the founding team. You should hopefully be able to get to some amount of traction and then you can use that traction momentum to then hire more folks. If the idea is really good and people believe in it and they can have like a personal relationship to it. I think that really helps for hiring people past the founding team. The first 10, who I think are the most critical hires.
Adam Conrad Obviously every hire is critical, but those particularly are really important, right? So for engineers that want to start their own software company, any last words of advice for them, don’t be discouraged by the naysayers.
Matt Arbesfeld So much advice out there. And so many people who will give their 2 cents on your idea and it can be very easy to be swayed by their other people. But if you have an idea that you really believe in and you know, is applicable, like I would say, just try your best to tune out the noise or at least put it through a filter and believe in yourself, I’d say is the most important thing. Because if we had listened to the naysayers, we wouldn’t be where we are today. And that’s my number one piece of advice.
Adam Conrad What a great way to end. Where can we find out more about you in LogRocket? Yeah, you can check us out at dot com. You can also go to our blog blog.library.com, which has an awesome front end tutorials. Then we also have a, actually a podcast for front end developers as well that you can check out that we just launched HotRocket.
Adam Conrad Oh, that’s a good play on the company name. I like that.
Matt Arbesfeld It is. It is. Yeah. We missed out on bLogRocket. So hot rocket to be log really does capture a lot of things in the software space.
Adam Conrad That’s really clever. Awesome.
Matt Arbesfeld Thanks Adam.
Adam Conrad Thanks again, Matt.
[End of Audio]
SE Radio theme: “Broken Reality” by Kevin MacLeod (incompetech.com — Licensed under Creative Commons: By Attribution 3.0)
What’s a runway? I don’t think that’s mentioned?