Taking a company from startup to acquisition with Sean Byrnes of Outlier.ai

Maren Kate
Welcome to from 5 to 50, the podcast dedicated to helping startups and founders survive and thrive through the early stages. I'm your host, Maren Kate and I'm very excited to have Sean Byrnes from Outlier on and Sean, did I say your name right?

Sean Byrnes
Sean Byrnes actually, thanks.

Maren Kate
Byrnes, gosh, I'm notorious at that and I never asked before, Ugh. Okay, so first off, when did you found Outlier?

Sean Byrnes
So Outlier was started in 2015. I had previously started a company called Flurry back in 2005, so 10 years earlier, that we grew over nine years to being one of the largest analytics and ads platforms in the world. So we did Android and iOS analytics and advertising. We sold to Yahoo in 2014, I took a year off, and then I started this the following year. I can't believe it's been that long some days and some days I can't believe that it's only been four years.

Maren Kate
Wow. Okay. So what did you do for your year off?

Sean Byrnes
So my daughter at the time, it's interesting, when you think about as an entrepreneur, why would you take time off? I think a lot of us that do this, you love the creation process, you love the building. It's almost addictive in a lot of ways because of how much impact you have. My daughter at the time was 18 months old and what I realized was that I had grown up. I've been doing it so long I didn't know how to turn it off and so I'd be sitting there playing with her and I couldn't be present in the way that I felt like I wanted to be. The kind of parent that I thought my children deserved. And I just realized, after a while of trying to work on those bad habits, that they were so insidiously deep in me that I was gonna have to do something radical to reset them. And so leaving Flurry, which was a hard decision, it was right around the acquisition which made it easier, but you never want to leave something you spent so long building, to spend time with my daughter without work holding over us was hard. But it was the right decision, I managed to correct those bad habits. And now, even though I have a new company, my ability to be present with my kids is fantastic. My ability to be there when they need me. And it was worth that year but it took a while. I think that there's this weird psychological momentum that builds up when you're working. So the first few weeks after I left a Flurry I was really busy and I was like, why am I so busy? I don't have a job anymore. And I realized my body didn't know how not to work. So it was scheduling meetings, and I felt like I had to be occupied, and respond to email, and eventually I realized that I was just going through the motions because it's all I ever knew. And so those habits were very hard to rip out but I'm really glad I did. Now my relationship with my family is much better as a result.

Maren Kate
So I mean, it is kind of an addiction in some way. Everything from checking email to that busyness to just, I guess, that power and that ability impact lives. But when you finally stopped and you took that year off with like other addictions did you go through a, what's it called like, a coming down phase, like, did you go through a time that was really hard to get it out of your system or once you shifted off you were like stoked?

Sean Byrnes
Like I said, I had to be very purposeful and very intentional about it. I had to purposely tell myself not to take more meetings, not to constantly be responding to emails because I didn't have a job so what was I doing? It was really just a muscle memory that I was exercising. The interesting thing though, about the addiction, is I feel like, though not painful at the beginning, but the longer you're away from it, I heard somebody refer to it the other day as “the madness”. So eventually the madness builds and you want to create, you want to build a product and it gets to the point where it crescendos and you just want to build for the sake of building. And so I feel like if you're thinking about it like an addiction, and honestly addictions are very serious, and they’re very difficult psychological illnesses and I don't want in anyway to downplay suffering with those but if you use that as an analogy the withdrawal happens later. It happens when you're a little bit farther away. And every second time founder I know, they either find a way to control it, or they just end up building products that have no customers, or building products for no reason. Just building for the sake of building because the madness is built up so much and that actually requires self control. So when I was sitting down to start this next company, I made sure that I focus on the problem, no solutions, no products. What was the problem we were going to solve? And I spent about six months starting with about a dozen problems and whittling them down and a dozen became six and six became three. And I would go to customers and go through sales motions to see who would buy what and who can I recruit to work out what. And that meant that I had to have consciously and intentionally really pushed down the madness, push down the temptation just started building because I'm so excited about this and focused on that problem. And that was a great decision. It means that Outlier is in much better shape as a result but I had to be very intentional then too. So part of it was breaking the bad habits and part of it was trying not to jump back in too quickly and do too much too soon.

Maren Kate
I like that and I was listening to another interview that you did and doing a little research and you sold Flurry for a few 100 million dollars so financially you didn't have to jump back into it if you didn't want.

Sean Byrnes
Yeah, that's true. It was a good outcome. I think that there's so few good outcomes in venture capital. I feel very lucky to have made sure everybody on the team that worked on it, everybody benefited from it so I felt good about that. I think the thing about financial security is that a lot of people in Silicon Valley talk about how they earned it, right. So you know, “I earned this outcome”, I saw that word a lot in Medium posts when people talk about exits. And I will say that no part of me feels like I earned it because I got very, very lucky. Now don't get me wrong, I worked very hard and I worked hard to go to college, I worked hard to go to grad school, I worked hard to get the company started. I sacrifice a lot but really, what did I risk at the end of the day? I'm a white male. I was born in a middle class family that could afford to send me to an Ivy League school. I went to said Ivy League school which gave me a great network. I was married when I started Flurry so I didn't have to worry about health insurance or income for a few years. And the company ended up succeeding, mostly because Steve Jobs launched the app store for the iPhone and Google followed suit with the Android Market, a lot of things like that. I didn't choose my parents, right. I didn't choose the fact that society doesn't give me some sort of systemic barrier to success. I didn't choose to be starting a company in the space where these big behemoths were going to change the world overnight. So we were successful and I worked hard to make the most of that. But to say that I earned it would be a gross exaggeration over everything else because it downplays all the hard working people who did the same things that I did and didn't get as far. So when I put that in context, you think about the future, what I want to do, I enjoy building these companies and giving people a platform to innovate and to move things forward. I feel like I'm good at it and I do it as much for the team around me as I do for myself. The flipside is, I think it's also important that my children get used to watching me go to work in the morning, and watch me come home from work, and visit me, and understand what that means. I don't want my children to go through life not having to worry about money. I think that's unhealthy that your kids never have to work. I know a lot of people who have been through that and some of them are well adjusted and some of them aren't. So for me, it's less about that and more what I can do. That if I can build a platform to help deal with some of the systemic change that is necessary so that you don't have to be like me, a white male from an Ivy League school, to have a good chance of success then that's worth it. That makes all this effort worthwhile. And so with Outlier, we are a dual goal company. We're a for profit company. We want to change the world in our way. We want to build a big success but we also want to show an example of how you can build a technology company where diversity is a core value and you can actually show people a different way of proceeding than some of the bad habits and systemic challenges tech companies have had.

Maren Kate
I love that. That's wonderful. So that leads in to my next question was just in 90 seconds or less, just tell the audience kind of the founding story or the why behind starting Outlier.

Sean Byrnes
Oh, it's simple. So Flurry had half a million customers around the world. I would go visit them, talk about their business, talk about how data was used to make decisions and every single one of them would ask me, “Sean, what in the world am I supposed to look for in all this data?” and I got that everywhere I went. I realized that the state of the art in business intelligence was great but only answering the questions that we knew to ask and the next generation of tools was going to have to help us figure out what questions to ask in the first place because that was the hardest problem. So I started Outlier to rethink business intelligence from the ground up. To revisit and throw away all those assumptions we thought we knew about how they needed to work or what they needed to do so that in the next decade as business data keeps growing companies feel like the business could speak to them, that the questions that they needed to ask were brought to them and they can make better decisions as a result. That was the vision four years ago and four years later, I'm happy to report that not only has that world that we envision come to rest but it is the fastest growing area in enterprise software today. It's exciting to see how fast it's really been adopted.

Maren Kate
When you started this, did you start it with co-founders or solo?

Sean Byrnes
So I'm the kind of person that works best at a team. I think that if I was a solo founder, I would probably waste a lot of time running in circles so I need people to challenge me. This was one of those more mature realizations you have later in life where I need a certain situation to be my best and for me it was having co-founders. One of the things that I think is funny is everybody assumes you've had success, you'll start another company, it'll be easy. You just snap your fingers and everything will happen to you. Investors will fall in your lap and co-founders will line up and everything will be easy. That is definitely not my experience. I was very lucky to have a good friend of mine, Mike Kim, who was thinking about making a move as I was getting Outlier started. I was convinced I was gonna have to be a solo founder of Outlier, based on my lack of success in finding a co-founder, and then Mike decided that this was something you want to do and he jumped on board and it's been a great ride ever since. This was the first company that Mike started so he's a first time founder and this is my second time around, but it's worked out very well on both sides. Mike’s rose to the occasion and I've learned a lot about myself to be a more effective partner here. But it's a very lonely road to be a founder, especially a founder CEO. There's a lot of things that you can't talk about with people. I definitely make sure I carve out time to talk to founders, especially first time CEOs, because often you can't talk about co-founder problems with your co-founder, you can't talk about employee problems with your employees, you can't talk about investor problems with your investors, and pretty soon you realize there's this whole class of problems that involve your investors, your employees, and your co-founder that you can't talk to anybody about. And if you bottle it up inside, it just eats you away. So I tried to be an outlet for other people to talk about and in doing so they can be an outlet for me and just make sure we realize that we're all people first and executive second and try to be a sounding board for that. But having a co-founder like Mike that I can trust and we can work through things, it's enormously powerful as an emotional structure in the startups because it's so stressful if you don't have those. It just eats you up.

Maren Kate
I agree 100%. I've experienced that both having co-founders, which is wonderful, and not, which definitely can be more difficult. So when you guys started this four years ago, did you bootstrap, self fund, did you take on venture again? How did you finance it?

Sean Byrnes
Yeah, it's a good question I have respect for both kinds of business models, for building it yourself for cash flow or venture backing. I tend to find the problems that are really at the edge more interesting. So I like new categories of software, and new types of approaches, really aggressive innovations and the problem with those is they're hard to bootstrap because you're really betting on a future. So I tend to like to raise capital as part of building these businesses. Like I started Flurry in 2005, which was three years before the App Store came out. I started Outlier in 2015, which was before the dot ai domain was even easy to buy a URL for. So I do typically raise venture funding. Interesting enough with Outlier, when I started, all I had was that problem that I mentioned to you, which is I wanted to build a system that would ask questions, that would look at your data and bring you the questions to ask of it. But I didn't really know if you could build software to do that because again, I didn't let myself think about the solution or the product. So the first six months of the company were really Mike and I, my co-founder, renting ourselves out as consultants to companies saying “give us all your data, we'll poke around, we'll tell you what we find, and let's see if there's economic value in the unknown”. Is there economic value in asking new questions? And it turns out that there was. We actually changed the course of some companies over a course of three days just by looking where they hadn't looked before so that was compelling. But then we still didn't know if you could build a product and so we started building products and we had one prototype and that was horrible. We built the second one. That was really bad, too. And somewhere along those lines, Chris Fralic from First Round Capital, who had been an investor in Flurry, he called me up and he wanted to invest and there were a bunch of investors that wanted to invest around then. I think that if you're a second time founder, one thing that is easier the first time is that you get that first meeting with an investor a lot more easily. Unfortunately, that's as far as it goes. After that first meeting you're still on your own needs to prove value but getting that first meeting is easier and that's obviously a huge step up. But the thing about Chris that was great is that he was first always a board member of Flurry who challenged the strategy, which I always appreciate people that challenge me and push me in new ways. And I remember when Chris asked, “hey, what are you doing with Outlier that’s interesting”? I told Chris, I'm like, “Chris, I don't know why I'm doing it. I haven't figured out what business we're in yet. How about we talked about this when we do”. To Chris's credit, and this is probably why he's been such a successful investor, he was very insistent that now was the time. So First Round led a seed round in us in 2015 and it was a million dollars and that was before we had anything. So a funny corollary to that was, it was just Mike and I, we just raised a million dollars, we're still working with the product looks like and so our burn rate was like $200 a month, something like that, and I remember to send out my investor update, and one of my investors was very concerned at the burn rate, and responded and wanted to talk about why the burn rate got so high. I remember responding and saying, “you know, it's just $200 it's not $200,000”. They had never seen an update with a burn rate that wasn't measured in 1000s and so they didn't know how to process it.

Maren Kate
Oh, that’s funny. Also I’d would love to see what your what's it called, you know, like months left. It'd be like 50,000 or something.

Sean Byrnes
Exactly, exactly. So we were lucky in that way. I mean, having First Round Capital and Chris on our team from day one has been a big part of our success. I knew we would raise venture funding. I thought it'd be a bit longer that we got before that but it turned out to be a good thing and to Chris's credit, because I think, as most founders, the madness kicks in on me too and one of the effects of the product building madness is that you tend to discount how hard things will be and I definitely did not realize how hard it would be to build what we wanted to build. Now in retrospect, as I described Outlier, as an automated business analysis platform that plugs in your data wherever it is and can analyze it automatically using a dozen forms of machine learning, it sounds complicated. It sounds like it should take a long time to build. And you know, naive overconfident me was like, “ah, we'll have an MVP in six months and then we'll just go to market and it'll be great” and two years later, when we finally had a product that really did work the way we wanted it to, I was glad that we had money early. We were lucky that Homebrew was the same way. Satya from Homebrew reached out, was very compelled by how well he and Hunter got our vision and they convinced me. We weren't looking for a second seed round at the time but he convinced me it was a good idea and we're lucky to have them on our team as well. And so we've been very lucky that what we were working on captured the imagination of some great investors that really wanted to get on board with us and the support they gave us made up for my over optimism on how quickly we thought we could get to market. So overall, I got very lucky in that way as well.

Maren Kate
I think one thing also true to mention is something you said, how you and Mike went around and hired yourself out as consultants to figure out the early product. I think that's another really powerful thing to do, who may be people that need to earn money, they need to maybe self fund or Bootstrap. I've done that a few times myself. You start with the problem and you just start solving it yourself and then after a certain time you're gonna be able to take some of that money invest in the project you're working in and you're gonna get closer to it. So I've always liked when people do that. I think it saves some stress down the line. And then where are you guys today? What's your current size?

Sean Byrnes
So we're closing in quickly on 35 people here, which is exciting for me. We have primary offices here in Oakland. We have a secondary office we opened up recently in Virginia because it's just hard to grow in the Bay Area if you're growing as fast as we are. We have some remote people as well but, yeah, the team is growing fast. It's exciting to see how fast what we're doing has really captured the imagination of the industry. So our customers are going to be some of the biggest fortune 500 companies in the world. We also do business with the U.S. government. And so this idea of automated business analysis has really come of age in the last 18 to 24 months as companies realize they have so much data and there just aren't enough data scientists or analysts in the world from them to hire them to look through everything to make sure they know where all those emerging problems and emerging opportunities are hiding. And Outlier is a turnkey solution that can be up and running in a few days on your data lake or your data systems and start providing that to you. So the value proposition, everything, and all of it comes back from that problem. We stayed so focused on the problem and didn't fall in love with solutions until we felt like something really was resonating with customers that we gave ourselves the flexibility to find something that really was magical, that has a foundation in the future, right? Because you want to, I think a lot of why for me that starting off, manually consulting was so important was that there's so many founders that spent so long building companies that never really had the potential to get as far as they needed them to, that they had some sort of limit on how big they would get. And maybe that was because the market was too small or the market was too competitive. Like if you're going to play this game and go through the stresses, it has to be worth it. You have to be building something that can potentially be worth all that effort you put in and it requires a lot of thoughtfulness to get there and focusing on the problem is one way that I use to make sure that I'm always focused something that will be worth the effort you put in because even though I do this because I love it, every time I'm here at work, I'm not with my kids, right? Every time I have to go on a business trip my kids are old enough to understand that I'm not there. And I do it but it has to be worth it. I'm not just gonna do that for any reason. It has to be something we're building towards in the future where we feel like the impact is real. So that focusing on the problem helped. I will say, ironically enough, and I don't think founders realize this, they get so enamored by things like artificial intelligence and we use a lot of machine learning because I'm so old that I went to grad school for artificial intelligence back before we called it data science, and they don't really realize that every great artificial intelligence or machine learning based product starts with a human doing a job. And they do it, and they prove value, and then we automate it with machine learning. It doesn't happen the other way. A self driving car drives itself because it watched people drive cars. So it's not that doing it yourself is low tech, or primitive, or whatever else. It's a fundamental building block of even the most advanced products and businesses in the world. I'm not embarrassed by it, I'm proud of it. I think it's showing that you really understand what it takes to build a business.

Maren Kate
Yeah, I think that the whole idea of doing unscalable things is doing things at least once yourself especially in Silicon Valley. I live in New York now but I was in Silicon Valley for six and a half years and one of the reasons I chose to move to New York is I wanted more of the balance. New York has a very different mindset. There's just a lot of different types of people. I found myself getting more influenced by whatever was shiny in San Francisco at the time and going off course versus sticking to the tried and true. So are there any growth numbers you can share? Anything that can give the audience an idea of your trajectory, kind of, how you've done in the last four years?

Sean Byrnes
We've been compounding growth like crazy. I think that we are on every measure almost four times as big as we were this time last year. We're definitely at the growth stage in the business. You know, in the early stages of the business you're trying to survive, right? You're trying to prove that you have enough to live to see tomorrow and you develop, and you purposely have a scarcity mindset where you're trying to focus on being scrappy, on experimenting, on finding product market fit, and then if you're very lucky, and you find it, and you stick to this growth strategy where the growth idea is can I make sure our growth rate is always increasing, we're always growing faster at the next step of the business. Because, you know, the thing that kills most startups, I don't know how much people talk about this, we talk about companies that fail at the beginning because they never find product market fit, and that's obviously the most common cause of death, as you just never find the right formula, you never find that value proposition. But if you do find it, the second most common cause of death, which is what will kill you at that stage, is that you start to grow more slowly the bigger you get. That the things that you do begin to lead to a declining rate of growth. Sometimes it's because you saturate your market. Sometimes it's because you're doing things that are unscalable that you can never find a way to scale. Whatever it is, it ends up killing you because you run into what I call “the plateau of death”, where your growth rate slows down enough that eventually your growth starts to look like a plateau. So the goal of a company like Outlier is can we always focus on making sure we're growing faster the bigger we get? And I don't think it happens by accident. I think there are a lot of companies that stumble on to opportunities that are so big that they grow that way. I'll tell you Flurry was that way. When the mobile app ecosystem exploded, you know, forget about market share, the whole industry exploded so fast that you could double every six months without really doing very much. But in most businesses you have to be very thoughtful about that. How do you increase the leverage of your team. Because you can't hire people fast enough so how do you make sure that your team can be twice as productive, four times as productive, 10 times as productive over time? How can you make sure your revenue is growing that fast? You have to focus on things like customer retention and customer satisfaction, customer growth. There's a reason why we didn't do much outbound marketing or outbound sales at Outlier for a long time is we were trying to make sure that the product we built generated fanatically enthusiastic customers who love what we did and were passionate about it and the engagement on the platform was very high. And as a result, we built one of the first BI, business intelligence, products that measures engagement like a consumer app on a daily active users basis where the daily active user rates are enormously high. So by building that foundation, I knew when we started to do things like marketing and sales, we weren't just dumping users into a leaky bucket. We were building on a foundation of satisfaction and account growth. And as a result, a lot of our customers are growing 50%, 100% in the first six months of working with us because we had that formula of satisfaction down so well. So now things are going well. I think that it's not, you know, partly it's us, partly it's the winds of history. This whole movement around automated business analysis or what Gartner calls augmented analytics is bigger than us. It's across a whole industry. We're lucky enough to have not just, you know, making our own noise but all of the industry analysts like Gartner and Forrester talking about this. All the big companies are trying to make a shift in digital transformation.

Maren Kate
You catch the wave at the right time for sure.

Sean Byrnes
Startups are like that, right, where you need to catch wave, align yourself with the winds of history.

Maren Kate
Timing is everything. I have a totally side question that just hit me. So I've done a decent amount of these interviews. One thing I've noticed is when I talk to companies that are completely bootstrapped or self funded, mostly like bootstrapped, versus venture backed companies, the bootstrapped ones are really fine with sharing revenue number sharing whatever and the venture backed ones almost never do. And my last company was venture backed and we also kept it really close to chest. Why do you think that is? I just realized, I have no idea. Is it a legal thing or is it where we're afraid our competitors will know how big the market is?

Sean Byrnes
Oh, no, I think that it's because venture backed businesses realize they're being judged at all moments by the investors in the next round. So if you share the information, without the context that goes with it all, the only possible thing that you'll do is alienate the next set of investors. I'll give you an example. So for us, I'm an enormously capital efficient guy. I don't jump on this mega round thing. I mean, the series A for Flurry back in 2006 or seven was 4 million dollars. I thought that was a lot of money. Now, that's like a small seed round, but I haven't changed, right. So our series A here at Outlier was a little over 6 million dollars that we closed at the beginning of last year, right, and that is, I think, a small fraction of what most series A's are. Most series A's seem to gravitate more towards 12 to 15 million dollars. And so we're growing fast, we're doing well but if investor heard all our core metrics, your ARR, your net retention, without the context of how much we've done with so little, how do you compare that with somebody who raised 30 million dollars in their Series A? I would sure as hell hope that a company that raised $30 million in their series A would have made it farther than we made on our $6 million series A. I guarantee they didn't make it as far as efficiently, or growing as fast as we are. But the reality is that investors, for better or worse, use these rules of thumb that become a challenge. And so as a venture backed company, you realize at some point that if you don't tell your own story, and somebody else is telling it for you, there's gonna be missing chapters that downplay it. If you're a self funded company, it doesn't matter. There's nobody judging you. There's no evaluation. So you can be proud of what you're doing in a way that it's harder…I would like to believe that as you're seeing a shift change in venture where there's more operators becoming investors, that's changing. That there's less rule of thumb investing and more critical thinking going on. But the reality is, there's a lot of investors out there where it's not about critical thinking, it's about pattern matching.

Maren Kate
Totally. “Do you look like Mark Zuckerberg? Are you at this revenue with XYZ?” Yeah, I remember it. That makes a lot of sense, actually, and I think that's one thing to realize. I don't know if it's neither good nor bad but just when someone's dealing with investors period, realize that you need to craft your own narrative. Like when I first raised money and up until maybe even a few years ago I just kind of always assumed that like if someone was in a suit, and especially if they were in finance, that they were infinitely smarter than I was. So I would go to these meetings, I'd be like, well, I'll explain everything and then they're so smart they'll get it. Like they went to Stanford they didn't drop out of their state school so they’ve got this. Then recently, I'm like, oh, you know, everyone has their core, everyone has a genius around something, usually. It's not always what you think it's going to be. Maybe they're just phenomenal at networking, or maybe they're really good at convincing LPs to give them money, or whatever, but that doesn't necessarily mean that they're going to get or understand your business. It doesn't mean that there's not going to be biases, whether they realize their bias or not, that happens. So crafting a narrative is an important part.

Sean Byrnes
It's absolutely the case. And it's not like everything about our company is a secret either. But the reality is that you also, as a venture backed company, you want to have people have an incentive to meet you and hear from you and talk to you, like any sort of sales process. Frankly, one of my biggest filters for investors is whether or not they'll come to Oakland to meet with me. There's a lot of investors who work in the Bay Area who've never been to Oakland.

Maren Kate
You’re like, com to to Stand Hill Road. Drink our free water.

Sean Byrnes
And it sounds really silly but part of working with investors is trying to tell the difference between real interest and just people wanting to educate themselves in the market or people who are looking at a competitive investment and want to compare you against them. And so having ways that you can essentially test that is an important part of just the process of getting to know investors. Because if you were to publish everything about your business on the website, how do you begin an investor relationship? How do you know what is biasing those investors? You probably will still have investors come to you. There'll been people who like it. But how do you know about the ones that you missed? How do you see about the access problems about why people are not talking to you? So it goes both ways. It's the same reason. And I always lament the fact that venture investors don't put their investment process on their website as well. It's very hard as a first time founder, if you're raising money from an investor, to know what to expect and so why shouldn't you put the process on your website. And a lot of that is for the same reason. The investors don't want to give an new founder a reason not to reach out and talk to them because their processes are largely flexible in many ways. And so there's lack of transparency on both sides. It's just how it is today in the market and hopefully it changes at some point but the reality is that's the way that it works.

Maren Kate
That's fascinating. We always encourage companies to, you know, like step two or three in the interview process when they're hiring is to actually reach out and explain ‘this is what the process looks like. This is our timeline’. That really helps with the candidate experience and that would actually work really well the same exact way with investors. Explain your process, explain your timeline, explain where you are in your fund. Okay, I have just a few more questions to be mindful of time. This has been super fascinating. And then I'll wrap up with just the three I always ask. What do you think are the most overlooked business applications for AI in your opinion? Any that you think about are glaring and someone maybe not in the space wouldn't think of?

Sean Byrnes
Yeah, I think that to date, most applications or artificial intelligence have been to build better tools and software, in general historically, has always been a tool. It's always been a hammer or saw that a person used to produce a result. And whether or not that's a CRM system where I just type in data or a CRM system that reminds me when to call someone, it's still a tool. I think that the real promise of what machine learning can do is automating things that we used to need people to do. People are by far nature's best production of a general purpose adaptable information tool, right? We can learn new tasks, we can react in uncertainty, we can make decisions with little information. But it’s traditionally been the only tool we had for information businesses. Now what machine learning is giving us at the scale that we can operate it out today is another alternative. And today, you see that happening in advertising, where 80% of online ads are bought by robots that are doing very sophisticated learning about who's gonna click on what ads. Most fraud systems are run by artificial intelligence, although they're informing a person, a fraud analyst, as to what to look at and what to look for. So there's enormous applications where you’re starting to see this move away from humans being the best fit to machine learning being the best fit but we're only at the early stages. There's so many business applications out there where a machine learning system can probably be better. I think you see examples today where recently there was a study that a machine learning system was more effective than human radiologists at spotting cancer and it's because the machinery algorithms can find things that a human can't find with the eyes that we have. So a general purpose tool, like a human brain, is not going to be the best at finding tumors, especially if there's very subtle information, and that's gonna be true of many different disciplines, in business and beyond. So I think we're really just scratching the surface of what can be done and it's exciting for me. I think it'll open up entirely new opportunities out there. I don't think you'll lose a lot of jobs in the short term, but I think it will be like the Industrial Revolution. There'll be jobs that transform, jobs that disappear, and new jobs that are created. I think that truck driving is probably not going to last forever but there's a whole new category of fleet management, right. Who's gonna actually manage those fleets, decide where those trucks need to go? How can we optimize for how they're delivering across the board? So you'll see an evolution and the change in work and we're just in the early stages of that. So the AI companies that are going to be the most exciting are the ones that aren't just building tools but they're really building systems that can do things that people can't do.

Maren Kate
And do it better. I like that. Okay, final three questions. What is your favorite book or podcasts in the last year?

Sean Byrnes
I don't read a lot of books about business because I find most business books are survivorship bias summarizing one person's experience without enough data for me to believe it's reproducible, so that won't help you.

Maren Kate
It could be another book. It could be fiction. I read a ton of fiction.

Sean Byrnes
Most of the reading I do is to give me a stress relief. There's a fascinating book I read recently called A Small Fiction, which is essential a conglomeration of an author who writes short stories in tweets. So short stories that are 140 to 280 characters long and it is enormously stress relieving because each story is so small but they actually are all complete short stories, they’re fascinating. It's a great way to get out of my own head and really entertain myself by reading a dozen of those and be entertaining myself across the board. I also spend a lot of time doing art. I find that, as a founder, the hardest part is that you are subject to forces you don't control and sometimes those are the people around you, sometimes it’s the market and art is interesting because you can be the governor of an entire world. And that feeling of being able to control everything you put into it is enormously stress relieving that way because it makes you feel in control for a brief moment. Even though most of the time you feel like you are driving out of control truck down the highway. So those are the things that work best for me. I don't have any specific podcasts, or I take that back, there is one podcast I find enormously fascinating which is, there's a podcast called Song Exploder, and they go to musicians and musicians break down how they wrote the songs that we know and love on the radio or online.

Maren Kate
Okay, I've heard of that. I’ll have to look it up.

Sean Byrnes
It's fascinating because you hear the process and the reason it's so useful for me is you hear about the modest beginnings of the songs and you hear about the struggle the artist goes through to turn that initial rhythm or that initial melody into a song and how much it takes and how hard it is. And it's a microcosm for startups. I mean, people always see the finished product and they're like, “Wow, look at that thing you built”. They don't see the modest beginnings. They don't see the ups and the downs and the crises. And I mean, Flurry was a huge success but we were insolvent three times. I only got paid for like half of the years the company was in business. So it was a struggle every step of the way and it's a good reminder. I love that podcast because it reminds me that the struggle is real and everybody's going through it. It's not just me. I'm not the only one who finds this hard.

Maren Kate
Yeah, you're right about the survivorship bias when you think about business books, too. Because, as someone that's went through a mildly public failure at my last startup, I got so many emails afterwards that people are like, “oh, you know, I went through something really similar but at the last moment something was saved”, or the story was just never really covered. And I never knew that at as a first time founder. I just assumed that everybody who succeeded, it was relatively linear. I think people are starting to talk about failure more often. But it is really important for everyone to know that no matter what creation process you're doing there's gonna be a lot more failure and a lot more falling down than there is success and time after time the more times you get up, the better chances you have. So at Outlier, what business tool could you guys not live without? What do you use a lot and love outside of your own tool, obviously?

Sean Byrnes
We are a remote friendly office. Our very first employee after Mike and I, employee number one, was based in Virginia so we have been remote friendly from day one. I thought it was an important competitive advantage for the future. And so what we can't live without is the Google Suite. We use Google Hangouts for video conferencing. We use Google Documents for everything. Everything we do is online so that we can collaborate. Because I'm a firm believer that if your first reaction to a problem is to jump on a whiteboard to talk about it, you're not remote friendly. And so we rely on a lot of those collaboration tools in the G Suite to make remote working work for us. So that's absolutely 1,000% what we're built on.

Maren Kate
Oh, yeah, I call it the Google Stack we do the same thing. Okay, so my last question, for founders who are listening who are somewhere between the five by their 50 person stage, what is the best piece of advice you've either received at that stage, or that you can give?

Sean Byrnes
So as you're growing the biggest and most important thing is to maintain your focus because focus and lack thereof is what will make or break the company. So when your small focus is given to you because there's only so much you can do but as you grow their’s potential to try and do more and it is a very leering siren that will lead you to fail as a company. So you need to become the focus of the company as you grow, as you have more people, and more money, and more time to try to make sure you focus on whatever it is that makes your competitive advantage. I wrote a blog post years ago that was titled Nobody Got Acquired for Paying Their Bills on Time. And so many founders get lost in the gears of business building, in the gears of the company, in the gears of everything happens. They lose fact of what made them special, what was that spark that got them started and what is that competitive advantage that if you find it, you use it as a wedge. You put 110% of all you're focus behind that wedge and you ignore everything else because that's what will make or break what your success is. So as you're growing, if you lose that edge, you lose that focus, you can't get it back. So fight for it. Fight for it like there’s everything on the line because you need it to be there.

Maren Kate
That's great. I love that. I just literally wrote that down and the name of the blog post so I can check that out as well. Okay, lastly so how can people reach you? You're on Twitter.

Sean Byrnes
I am on Twitter. Just @sbyrnes on Twitter. My email is just sean@outlier.ai. I'm there as well. I'm always available. I try to help out as best I can. I can't help as many founders as I used to because Outlier is growing and my kids are growing too so my free time it's fairly limited but if I can ever help I try to do my best to do that. There's so many people that helped me over the years who asked for nothing in return and were there when I needed them. The best I can do is hopefully be there for somebody else.

Maren Kate
Awesome. Well, I really appreciate you taking the time to chat with us today. I'm sure this will be super helpful to a lot of folks. It’s already been really helpful to me. Thank you so much, Sean.

Sean Byrnes
Thanks for having me and thanks for doing the podcast. This is great.