Michael Pryor - Baremetrics

This week I talk with Michael Pryor of Trello. We talk about learning to code from computer magazines, working in a grocery store freezer, co-founding Fog Creek 17 years ago, coming up with the Trello concept, getting acquired for a huge sum of money, what it’s like working at a public company and so many other things. Hope you enjoy it!

Josh Pigford: All right. Hey Michael. Thanks for hopping on the call.

Michael: Hi Josh. Thanks for having me.

Josh: You’re in New York, is that right?

Michael: I am. I’m looking at the Statue of Liberty right now.

Josh: All right. Is that from in front of your office you can see the Statue of Liberty?

Michael: Yup.

Josh: Well that’s good. That’s a good view. Are you from New York?

Michael: This is the interesting question because how long do you have to live somewhere before you could say that you’re from there, right?

Josh: We’ll say 22 years.

Michael: I’m getting there, pretty close. I grew up in Lancaster, Pennsylvania. Then I went to school at Dartmouth up in New Hampshire. Then after school, when I graduate from college, I moved to New York. That was in 1998.

Josh: Okay. Gotcha, so you graduated from Dartmouth in 1998, is that right?

Michael: Yup. Then I moved straight to the City and I’ve been there ever since. I was only supposed to be there for a year, but that’s what-

Josh: Happened.

Michael: Yeah, exactly.

Josh: All right. Cool. Actually, I’d like to touch on that. I’d love to hear your story pre-Trello, like way back. When you were a kid, what were you up to?

Michael: You know, I don’t know if you know, but they used to send out magazines that had computer code in them. This is how I first got interested in computers. I didn’t know anything about the code or what it meant, but I realized that if I typed all those letters without any errors into the computer, then into my TI-99 at the time, then cool stuff would happen. The worst part was I didn’t actually know how to use an editor, so it was just the command line. I’m typing these basic commands, so if at any point in time, I didn’t even know how to fix if I had done a type. It was like I really had to type the entire thing without error.

I started playing with computers very early on, playing computer games, all that kind of stuff. When I went through high school, I hd a bunch of jobs. I worked as a bus boy for a while at a deli place. I realized that I made my tips as a percentage of the waitresses’ tips and so I wasn’t getting paid that much because the checks for a deli are not that high. Then I went to become a bus boy at this fancy pants restaurant that was out in the country because the waitresses made a lot more money because the checks were a lot higher. Then I made more money.

Then I became a waiter out there. Then I got this job that paid really well, but it was in a freezer. It was like a grocery wholesaler, so all the trucks would come in and put all of the food into a refrigerated warehouse and a freezer warehouse. They paid you extra to work in the freezer warehouse because you had to wear a suit and this full-body suit and gloves.

Josh: You’d be freezing all day.

Michael: Yeah. It was like the middle of the summer, but you’re freezing all day. I did that job and I was in high school. It was the summer and I was working with these adults that were working this job. That was their life job. That was their career. I basically decided I do not want to do a manual labor job for the rest of my life. I better study really hard and do well in school because I was just looking at the struggle that these people went through and it was not fun.

I went to college and actually, at that point, I didn’t know what I wanted to do. I worked in the dining hall for a while. Then I also delivered newspapers. I had to get up super early and I had to skateboard around campus. I was the only person skateboarding around the New Hampshire campus. It was really weird at the time, but then I discovered that there was a job at the computing center to do computer support. That was like ringing this bell in me. It was like hey, that’s a cushy job, right? I don’t have to scrape dirty food off of dishes or get up at five in the morning delivering newspapers. I went over and I told the woman who ran the computing center. I was like I don’t really have that much experience with Macs. The whole campus was Mac at that time, but I’m a really hard worker. I’ll do a great job, and she gave me the job.

I basically just learned as I went and started taking computer science courses and one thing led to another. Then I graduated from school in ’98 and came to the City at a startup called Juno, which was a startup that was going to compete with AOL. AOL charged you $20 a month to get your email and we were going to give it to you for free, but we were going to sell it, advertising. We were going to make money off of advertising-

Josh: Juno, like the email client?

Michael: Yeah.

Josh: Yeah, I totally remember that.

Michael: I was a programmer and I wrote code on the Windows client that was the Juno application. Funny story, that was actually a project that was done by a hedge fund called D.E. Shaw. David Shaw, this guy in New York, he had a bunch of money and the internet was kind of hot at the time, and so he had smart people working for him. One of the guys had this idea for Juno to compete against AOL. This other guy had this idea. He was like I think selling things on the internet is going to be a really big deal, so maybe we should sell stuff on the internet. We’ll make a store on the internet. Maybe we’ll sell books. I think David was like ah, I might give you some money but I’m not really excited about that. That doesn’t seem like a very exciting business, so Jeff Bezos went and did his own thing.

Josh: Small, very, very minuscule success story.

Michael: Yeah, right, right. Juno kind of just floundered. It was actually, we went public. Juno went public and it was one of the first IPOs that closed lower than its opening price. The day before an IPO, you basically went around the office and there’s all these kids there, basically. I was one of them. It was ridiculous because we’re sitting there thinking wow, we’re going to open at $11 and we’ll probably close, our stock price will probably close at, I don’t know, $111 on the first day, because that was happening. It was ridiculous. You’re like okay, so I’ll work for a couple of years. Then I’ll retire. It was so crazy. Immediately, your dreams are dashed like a day later because the stock goes down. In fact, the only money I ever made was after I quit, I bought puts and bet on the price tanking and it did, so I made a couple grand on that.

Josh: That’s the emotional rollercoaster from thinking you’re about to be a millionaire. I assumed you played the what if game on the share price.

Michael: Yeah, totally, totally. Of course. You’re like the first I’m going to do is this. Then I’m going to just relax and not do anything and travel the world.

Josh: What was the tone in office as the day went on? Was everybody just like somber and [crosstalk 00:08:15]

Michael: Yeah, it was. It was … Basically, it started … It changed the emotional tone and it never really recovered because it also meant that it became harder to raise money. It became harder to run the company. People started getting laid off, and this was happening across every technology startup, not just Juno. That tide was shifting in 2000 and it was really starting to … It took a while. It took a long time, but first, all the people that were English majors that were in there working at startups, they got laid off first. The developers still had their jobs, but it wasn’t as fun because we were working with all of our friends and some of them were now unemployed and hanging out at Starbucks, although at that time, they extended the unemployment for a very long time. It was kind of this extended vacation for a lot of people.

Joel, I met this guy, Joel Spolsky, a little bit older than me, had worked at Microsoft on Visual Basic for Excel. He was a program manager on Excel and came up with and implemented VBA for Excel. He worked at Juno with me and we’re kind of there. It wasn’t very fun. One summer, he and I said why don’t we just do our own thing. I was naïve enough to think that I could do that. I quit my job and Joel quit his job. We started consulting at the time. We started Fog Creek Software, was the company, and we just … We were trying to pay the bills, so we were like we’ll do consulting. Then we’ll build some software on the side. One thing led to another, and we … I don’t know how much detail you want me to go into-

Josh: Let’s hop to Fog Creek there. You’ve been doing Juno. It IPOs but that’s sort of a downer. Everybody’s getting laid off. Is the reason that you start what’s really just a consulting company or maybe almost like a holding company to just do stuff, why do that? Was it because of the failed IPO or the lackluster IPO? Or was the timing of that irrelevant?

Michael: I think that gave us the impetus, because it gave you an opportunity to do something different. I don’t think I ever set out with this mindset of I’m going to do my own thing, start my own company. I do have … When I look back at my life, I definitely had this naïve mentality that I can kind of figure anything out. Also, I’ll just go do it even though I might not know how to do it. Like saltwater fishing, for some reason that got into my bones and I was like I’m going to figure this out, and I figured it out. When out off shore 60 miles by myself and figured out how to catch tuna. I don’t know how I did that. It was a lot of days when I just came back and had nothing, but eventually, you start to pick things up. You talk to the right people and you learn and you can figure it out.

I didn’t set out to start a company. I think I just wanted to do something fun and Joel was an awesome guy to work with. We were stuck in New York City and it wasn’t really a great place for software developers because at that time, Silicon Valley was the place to be. If you’re in New York and you’re a developer, Juno was cool but it was really an ad company. Even though we were software developers, the business was seen as an ad business and so we were just supporting the ad business. We were just building that little thing that shows the ads. That’s how the company thought of us.

We were like hey, let’s build a software company where the developers are the most important component of the company. We’re going to do that in New York because we don’t want to move, and so we did. In the beginning, it was kind of … We were just barely paying for one of our salaries. In fact, we recorded Joel’s salary but didn’t pay him for a couple of years. I remember the day, it was probably like four or five years in where we were able to back pay all of the salary that we had recorded for him. That was pretty cool.

Josh: You mentioned you guys do this consulting thing or started this company, and part of that’s because you did sort of have this figure it out mentality. I think, to me that seems to be a pretty common trait among any founder or entrepreneur, is this ability to just … I don’t have any clue what I’m doing, but I’ll figure it out.

Michael: Yeah. It’s almost like … I don’t want to … It is a little bit of a naivete. You’re almost like oh, it can’t be that hard, which is obviously, it’s false.

Josh: Ten years later, and it’s like oh.

Michael: Exactly, but it lets you just plow right into it. I think that so many things are overwhelming but you look at it and you see another person doing it. I guess maybe some of it is like well, if they can do it, I can do it. Like back in that day, just simple things like just accepting credit cards online for your software to sell your software online was impossible. I remember going to the bank and telling them I want to sell stuff and accept people’s credit cards. They’re like oh, well have to come out and do a site visit and you’ll have to display these things on your retail establishment door. I’m like we don’t actually …

Josh: There’s no establishment.

Michael: Yeah. There’s no … But they couldn’t figure that out. We just had to hack the system. Okay, I guess we’ll put them up on the door and they even gave us a credit card reader. We had to get a credit card reader because they’re like yeah, there’s a gateway where you can send the credit cards over the internet, but most of your business will probably be through this credit card reader. Yeah. There was all this stuff, and just discovered as we go, and just learn as we go. There was no book you could read. Now it’s great because for every piece of running a business, you can just sign up for a SaaS service to do it.

Josh: Right, simplifies the whole thing. But I think, you think of somebody who goes to say business school or something. To me, there’s something to be said of just not knowing, like the naivete that you mentioned, just not realizing how difficult it can be, is really probably a good thing and there’s probably lots of companies that never would’ve gotten started had the people who started them known what on earth they were getting into.

Michael: Yeah. I think it’s the problem that most humans get caught up in it. I do too even, sometimes, which is that you think that the hard part is the idea and it’s not. It’s obviously the execution. It’s doing it day after day and failing and failing and failing and learning from your failings and then doing it again and doing it again. It’s only in hindsight, when you look back over a decade or 15 years, you’re like oh wow, I did come a long way. But when you’re doing it at the time, the amount of work that you have to put into it just seems like oh, I’m not making any progress.

Josh: You guys were doing consulting stuff, but at what point did you guys transition from the consulting part to actually building products?

Michael: Well, we always knew that consulting doesn’t scale. We wanted to just use it to bootstrap funding for selling software. We kind of were taking cues from what people were asking us to build and seeing if we could build something out of that. Actually, one of the first contracts we had was this company asked us to build … They had a bunch of HR people. It was a pretty big company. They had all these HR people. They were trying to put their HR documents up on their website, but they needed the HR people to be able to update those docs really easily. They basically needed a content management system, but at the time, that wasn’t really a thing. It kind of was but it really wasn’t at scale for consumers. They also, everyone used PCs and there as no such thing as SaaS.

We built that for them, this app that was like a VB app, Visual Basic app. We sort of made it more generic to be a content management system that we called City Desk, because we thought that people were going to be putting a lot more stuff online. The real blocker was that more people weren’t technical, so they didn’t how to get the movable typescripts into CGI bin and chmod 755, all the scripts so that they ran on the web server. It was like getting that set up was, you had to be a developer to do it. Our thinking was hey, we’ll just make it really easy. People just need their FTP credentials. We’ll write a desktop app that lets them add articles and then figures out all the HTML and then just uploads it.

Obviously the people that were building Typepad and Movable Type blogger and all those kinds of tools at the time obviously saw that same issue when people were trying to install their tools, and solved it by giving them a webpage that somebody could click to basically use their software. That became a way better solution to the problem that we were solving. That product, City Desk, we actually sold a decent amount of it, but it was the wrong it was the right problem to solve at the right time, but it was the wrong implementation.

Josh: Oh, gotcha.

Michael: We also had some bug tracking software that we had, Joel had written at some previous company and gotten the rights to. We were like eh, we’ll just throw it up on the website and sell that because we’re developers and Joel’s been writing some articles and developers were reading them. Maybe we could sell them this developer tool. People were using Bugzilla at the time. We were like hey, buy this thing. We stupidly named it FogBugz-

Josh: With a Z.

Michael: With a Z because that was cool at the time. That didn’t take off. We made like $10 one day and like $50 a couple weeks later. It was really slow but over time, when you looked at it in hindsight, it was growing pretty well and pretty fast.

Josh: Gotcha. You guys were doing consulting, but then you kind of figure out the product stuff, launch the bug tracking thing, and then it was like 2011 or so, Trello comes into the picture?

Michael: Yeah. We skipped like a whole decade. Let me just cover that decade real quick. One thing that happened early on was actually we were making pretty good progress. We had these consultants. We were also building software. We hired two people. We even had an intern at one point. We had hired two people, and then really the bottom dropped out of the market. It really crashed. It had been not doing great, but then it really crashed. The first thing that everyone cut was outside contractors. That just saves money immediately. You’re going to pay your employees before you pay people that aren’t really working for you.

All of our revenue basically dried up. We had just hired these two people, one of which, his wife had just gotten pregnant. It was actually awful. It was the worst part because we basically said to them, we went to them and said hey, we have enough money to pay you for another month, so you can start looking for a job now, or we can spend the month looking for more contracting gigs, and then that all … If we get one, then great, but if not, then that’s it. They were like let’s do it. Let’s look for the gigs.

We did it, nothing came up, and then they left anyways. We found both of them jobs with people that we had known, so it wasn’t the worst case scenario, but it was really sad. It was a really sad moment, really depressing. I would come to work and it would just be me and Joel. It was like starting from scratch and we clawed our way back, piece by piece, but really during that time, we just focused on the software and building the software. Joel was writing his articles and building up his brand, which was for us, a really good marketing tool. In hindsight, it was the right marketing tool at the right time, but it also prevented us from doing traditional marketing.

Josh: How so?

Michael: We never had to learn that muscle or that skill.

Josh: You were defaulting to Joel.

Michael: All we knew was content marketing. The only person that could do it was Joel. It was like his voice, his style. After a while, when he got ran out of ideas to write about or it wasn’t as interesting to him, there was nothing beyond that. What was our traditional marketing skills? Over that 10 years from 2000 to 2010, we experimented with a lot of different things, but FogBugz was really growing and was really the driver of a lot of our growth. We did have this culture of hey, let’s try new things. We built this app called CoPilot, which was a screen sharing tool we cobbled together with VNC at the time. We used it because we saw a need to control other people’s computers when we were trying to install FogBugz on their servers. They were trying to tell us what they saw on the screen and it was really difficult. You were like if I could just see what they, if I could just take over their computer, we can solve this in two seconds. We wrote some scripts that did that on top of VNC and started selling it.

That was actually also, that was actually the right product at the right time because LogMeIn and GoToMyPC didn’t exist then, but we marketed it to customer support people or people helping their moms with their computers. The marketing was all wrong if you look at what GoToMyPC did and LogMeIn. Also, I think we didn’t keep, we didn’t invest enough to keep up with the tech to make sure that it was performing. Really, for us, it was like hey, we’re making $30 grand a month. This is amazing. Really, we needed to plow all that money back into the product in order to build a billion dollar business, but there was a billion dollar business opportunity there because those two other companies did it, but we didn’t. We made some money off of it, but that was, not really a failure, but we learned some lessons there about marketing.

Josh: At what point did you decide CoPilot, which presumably of the products you had made thus far, that was the biggest success? Or …

Michael: No, FogBugz was a way bigger.

Josh: Oh FogBugz was still, had surpassed that? Okay. Well never mind.

Michael: Yeah. Nothing ever reached the pinnacle that FogBugz had reached. That was always driving a lot of our, the rest of our projects.

Josh: Okay, gotcha.

Michael: Until one day in 2010. We had this, Joel had this list of ideas. He had like a text file on his desktop that had all his ideas of things that we wanted to build someday. He was talking to Jeff Atwood who wrote a blog called Coding Horror. They were both kind of tech guys that had wrote websites that people followed. Jeff, I guess, had left his job and was looking to do something new. They were just chitchatting. Joel kind of pitched this idea to him because at the time, there was this website called Experts Exchange which always showed up in your Google Answers and it was like if you knew this hack where you scroll all the way to the bottom, you could get the answer. It drove programmers nuts. It was awful. It felt like such a scam.

Joel had this pitch that he wanted to build something to replace that. Jeff kind of got interested in that. We formed a little entity between Fog Creek and Jeff, although Jeff did most of the work and Joel was the PM, I guess. They would have these podcasts that they put out on how they were building it. Jeff brought in two of his friends and we formed a little entity. Stack Overflow was born and took off. It was funny because in the early days, we were so accustomed to selling software that when StackOverflow.com took off, our first thought was hey, we could sell this Q&A software to other people to run their own Q&A sites. We started doing that in beta and had people. It was kind of like when we looked at the sites, they were just like ghost towns.

We realized the reason Stack Overflow succeeded was because Joel and Jeff brought their entire audience there on day one and created both the supply and the demand in the network. That’s the whole key. It wasn’t really the software. It was the community that was so valuable. That wasn’t working out for these other people and we realized that was kind of not great. We were going to sell this software to these people for $10 a month and their sites are just going be vBulletin sites or PBBB. It was just a continuation of that.

At the time, Stack Overflow was pretty successful so some VCs had found out about it. Union Square Ventures came and talked to us a little bit and said look, the value that you’ve created is this network of developers, not necessarily the software. Selling the software is really not the direction that you want to go. If you want to build these larger sites, you have to nurture them and seed them. We canceled the selling Stack Overflow software business and spun Stack Overflow off into its own company and took VC for it.

Josh: Then ultimately that spawned Stack Exchange? I mean, just the whole …

Michael: Right. Then it was like hey, we built Stack Overflow’s Q&A for developers. Let’s see if we can build Q&A sites for adjacent things that those developers might be interested in. We could spread out. We’re sort of cognizant of the fact that Yahoo Answers was this place where you could ask any question, and it just became a cesspool. If there’s no quality filter on the questions, and everyone can ask any question, it just, it’s a broken window. It just makes the site

 feel like …

Josh: It will eventually end up at the bottom with just the worst of the worst.

Michael: Yeah, right. You have to have some rules around what is allowed and what is not allowed. The not allowed is actually just as important as what is allowed. We also have learned that it’s also important to be nice. There can be rules but you also have to be nice to people. That’s an important component of it too, but …

Josh: What as your role in the Stack Overflow, Stack Exchange bit?

Michael: I was like the business guy. I took care of the payroll and the sales tax and the Quickbooks. I just did all the operational stuff. I wasn’t really product … In fact, when we spun Stack off into its own company, I became the CFO and just did all the operational stuff. It was more like a COO but I was not just finance, but I don’t really have a finance background, so I just had to learn it. It was that same thing again where I’m like I don’t know what deferred revenue is, but I guess I need to learn.

Josh: I’ll go look it up.

Michael: Yeah. I did that for a while and eventually, we realized maybe we should hire somebody that has some real finance experience at some point, and so we did. It was fine. Also at that time, Fog Creek had been humming along and had built this Trello thing. I’ll tell you how that came about. That was kind of interesting. We started Stack Overflow in 2010 and in 2011, Joel had this idea for Trello. The idea came about because we kept doing this thing in FogBugz where we’d get all these cases of things that we wanted to build. A year later, you’d realize you have this giant backlog of stuff you’re never going to do or it wasn’t relevant anymore. You just declare bankruptcy and you close it all. You’re like, why did we do all that work to put all that stuff in there anyway?

The problem was there was no forcing function on the work that we put into the system to deal with it. We just saw it as this unending bucket that we could just keep adding to. That created a cognitive load on us that didn’t allow us to know exactly what we were working on. It was … We basically said … We were also managing a company that was kind of bigger at that point in time. We were trying to figure out, I mean now we have this database of all this stuff, but what are people really doing? You’d go in their office and you’d see they’re writing code, so they were clearly doing stuff, but it wasn’t clear how it fit in the strategy.

Joel, he came up with this idea. He was like okay, what we’ll do is everyone have a to-do list and, I don’t know, five slots in it. You only get one to-do list. Josh, you’ve got a to-do list. It’s got five slots in it. I can see yours. You can see mine. Two of the items are things you’re working on right now. Two of the items are things you’re going to work on next, and one thing you’re just never going to do so don’t even bother asking me to do this thing because I’m just not doing it. That was just a random idea. At the time, I was like this is a dumb idea. This is not a good idea. I don’t get it, and also because I think Joel wanted to call it Five Camels or Five Things. It was something like, I was like … It’s just that the more I thought about it, it was not … I just didn’t get it.

The idea swished around for a while. Some people played with it. We were also looking, as developers we were looking at, there were a bunch of kanban boards out there. We had post-it notes up all over our walls in our offices and all those kinds of things. It was sort of this mish-mash of lists, but also post-it notes, which weren’t really tasks. They weren’t defined as tasks. They could just be whatever you wanted them to be. Also, trying to think about our marketing mistakes in the past where every time, we kept building developer tools and at that point in time, Atlassian was huge. It was like hey, can we build something even bigger than our current market because what’s really exciting is to build a tool that 100 million people could use. That would be so cool, but also so audacious and ridiculous.

All that stuff swirled together in a pot. That’s how Trello was born, was that it came out of this idea of well, we’re going to borrow from kanban, but the cards are not tasks. They’re just cards. You can use them for whatever you want. Just think of them like sticky notes, but also there’s not any kind of concept of dependencies or hierarchy or structure or, you’re never going to see a Gantt chart in Trello because this thing is not built for developers. It’s built for humans. That was actually pretty tricky because early on, we knew all the people that were going to use that tool were developers, and every time we’re going to hear feedback, it was going to be some feature request for some project management feature.

We were going to have to say that’s not really what we’re building. I know that’s 100% of the people that are using it today but we hope some day, we’re going to have sales people use it and marketing people use it and people to plan their weddings and people using it for their little leagues. The idea was that it would give structure to any kind of process that you’re working on. The code name for it was called Trellis, which is a structure for plants to grow on. That’s how it came to be.

Josh: Was it hard to be dogmatic about the, obviously developers who can be vocal. Was it hard to have to constantly push back and say yeah, that’s a great idea but we’re not doing that?

Michael: Well, it’s interesting because we put up a Trello board where we took feature requests. Then you could vote on the cards, because this is around the time that user voice was big and talking to your customer. In the early days, I think that was very valuable because we had a small set of people, but as time went on, that board actually became a really bad idea because what it did was it essentially told people tell us what you want, and if you vote for it, we will be compelled to make it, which is not something we wanted to say.

We were actually trying to say we have a vision for what we’re going to try and build. We also want to hear from you, but actually, what we want to hear is your story and your pain point, not your idea for how to fix it because we’re trying to make something that works for you and for somebody that’s nothing like you. You might come up with a feature that’s going to fix your problem exactly, but really what we need to do is figure out a horizontal solution for everyone. It might only fix 80% of your problem.

Eventually we took that board down because there would be these cards that had thousands of votes, because that number just keeps going up over time. Then the more things you don’t do, those are the things that stick around. Then people start saying this feature was requested four years ago and you didn’t do it. You’re like I did this to myself because I told these people to tell me what they wanted by giving them a tool that made it seem like that’s what I wanted. It’s not what I wanted. What I wanted is to have conversations with people and figure out their pain points.

I’ll tell you one example of a story that comes into that. One thing people kept asking for was they wanted to be able to mark a card as done because they’re coming from this project management world. Most of the boards they have have cards which are tasks and they’re like I want to mark a task as done. I want to mark a card as done. For us, the problem there was that cards are not necessarily tasks. If we made them tasks, then that compartmentalizes what Trello could possibly be and narrows its breadth. We struggled with that. I would ask people why can you just archive the card? They’d say no, but I want to see it. I would say well why don’t you move it to a list called done? They’d say no, I … There was always this-

Josh: Some excuse.

Michael: … like what is going on. As I drilled down more and more, you could see a pattern which was if you actually looked at what people were doing, when people added a due date to a card, if the due date had passed, Trello would make a little badge on the front of the card which looked like a red bell, like bright red. When you looked at your board, you noticed it, but they’re like but I did that thing. Trello doesn’t give me a way to tell it to stop yelling at me because I did it. I was like oh, it’s actually just the red color that’s bothering them. Trello is causing them anxiety and poking them and that is the pain that they need to be solved.

When I understood that, it was like okay, I can bring this down to a very finite … I don’t have to change the whole system of Trello. I can just focus on that red bell because that is the problem. We had a couple different ways of fixing that. We had a spec that let you, if you added a due date, you could add a completion date. If you didn’t have a completion date and it was after the due date, then the bell went red. If you added a completion date, then it would turn green. Even that felt a little heavy handed because even the concept of due dates felt too project management-y to me. I always wish we had just done dates, but one of our developers, like a couple months ago. This feature’s been around four years. A couple months ago, he came up with this idea and just shifted. Everyone’s like oh right, that’s it. That’s perfect. What he did, all he did was he took a due date and he added a check box.

Josh: Check box. I’ve noticed this in the past month or so and it felt so good to be able to check it off.

Michael: It’s just check box. That’s it, right? It doesn’t even make sense because you can’t mark a due date as done. That doesn’t even mean anything. The due date’s not done. The task is done, but everyone just gets it. The requests for that feature just basically went to zero. It was like all I needed was to stop the red. Why is Trello yelling at me? I did this thing. Stop yelling at me. That was one of those cases where it was totally useless to listen to the feature request and very important to …

My mission is really, I think our mission at Trello is really, it’s successful because the on-ramp to figuring out what’s going on is so easy and matches a metaphor that everyone gets. They’ve put sticky notes on a wall. You’ve used lists before. There’s not a lot else going on there. The more abstract concepts that we make in the software, it would make it more powerful, but it would also mean that the learning curve would get more complicated.

I think about tools like a very vertical tool like even Jira or Salesfoce. There’s all these concepts. In Salesforce, you’ve got leads and contacts and opportunities. Those different things mean different things in the real world, and you have to learn what that means in Salesforce in order to be useful with it. Once you learn that-

Josh: That’s why there’s people who sell Salesforce training, right?

Michael: That’s right. Exactly, but once you learn that. It’s super powerful because if you adopt that into your data workflow, even you can customize the workflow, you’re really adopting this schema. If you adopt that schema into your work life, when you give Salesforce that information and it knows what your data looks like, it can tell you all kinds of amazing things. It can be like hey, your pipeline is way too short. It’s like telling Baremetrics the same stuff. If you get your data into the right format, I can tell you oh, you want to know what your magic number is? Done.

You get all these benefits from out of that which you don’t get with Trello because Trello is agnostic about the structure of your data. We let you, it’s very fluid and it sort of does, but I don’t know the card in a certain list, I don’t really know if that represents code that’s going out to production or not. I would have to analyze the name of your list and that kind of stuff.

I think about, one of the cool things about Excel was that when you go into Excel, you write, you put your data into the sheet, but you also put your model of how the data relates to itself in the sheet. When you open up a sheet, it’s just blank. It doesn’t have a pre-ordained knowledge of how all your data fits together, but over time, you build it and you change it. You add sheets and you delete stuff. Also, any old cell, you could just write a comment. You know what I mean? It doesn’t even have to fit in the model that you’ve defined in another piece of Excel. I think that’s where Excel’s power comes from, and that’s how I see Trello when you look at processes.

These are some Lego blocks. You get what a card is. You can make it, you can get some labels. You can play with some people’s faces. There’s some checklists and some dates. You put that stuff together in the way that you think represents this process that you’re working on. Then you’re done. It’s solved, right?

Josh: Yeah. I think it’s almost like … Excel’s the same way as Trello is in this. It’s almost like a really powerful sheet of paper. You just, if you put the data on there and then you define what that data is and then all of a sudden, you can do a lot of powerful things with it.

Michael: Yeah.

Josh: Trello’s launched 2011. Then in 2014, you guys spun it out from FogCreek, right?

Michael: That was around the time that we hired a CFO at Stack. We were also thinking about spinning off Trello because we were running Trello with the profits from selling FogBugz, and so it was only growing as fast as we could grow according to how much profit FogBugz was throwing off. We basically had four different things that we wanted to implement, but our dev team was only big enough to do one at a time. We were like okay. We have investors that we know from Stack. We could, they’re interested in Trello. They see the vision. We could turn around a term sheet in like a day. Let’s just, we spun off Stack Overflow. We learned our lessons and took our lumps. If we do this a second time, we’ll probably figure out how to do it a little bit better.

Then we were like but the problem is who’s going to run it? Joel was running Stack. I was running Fog Creek. It was like how do you find someone to own your baby and grow it? We struggled with that for a very long period of time and eventually ended up in the camp of okay, I was going to do Trello and we kind of would collectively, Joel and I would put some of our time into Fog Creek and rely on the managers there to keep it going, which was kind of okay. I guess it worked for a year or two, but it was not the best situation until we just hired, Anil Dash is our CEO last year which was a huge win for us.

Josh: Sure. What was your role? Prior to breaking out Trello into its own entity, what was your role on Trello the product?

Michael: Oh good question. I don’t remember.

Josh: Were you like a project manager kind of thing or …

Michael: No. I think I was still just doing the business stuff, like the operations. I think basically that, the vision for the product was really Joel’s and he was the product visionary for that, and also the team of people, the core team of people that put it together, there were like five people at the company that were this group of startup-y people that grew it, went to TechCrunch Disrupt and launched it on stage and grew it into a product and figured out what to do with it. But I was … Yeah, go ahead.

Josh: Joel’s mainly focused on Stack Exchange, Stack Overflow, right?

Michael: Yeah.

Josh: I mean, at that point, so was it like you became the default because you had been the co-founder of Fog Creek?

Michael: Yeah. I think so. I think in hindsight, you have to have … There’s a certain, what is the word for it, but there’s a certain buy-in you need from the employees, right? There’s only, there’s certain people that can take ownership of that. Because I was a founder of Fog Creek and had worked so closely with Stack Overflow, I was able to be the CEO of Trello, even though probably at that time, some people were still skeptical of whether I could pull that off, including myself. But when you looked around, that was the best choice that we had. It wasn’t, there was going to be nobody else that showed up that somebody said oh right, this person’s perfect for taking this startup and this series A and turning it into a big, successful company.

Josh: What was that transition to CEO like for you? Was that a difficult thing? Or does it kind of feel natural?

Michael: The pieces of the operational pieces of the business, getting the money part working, that part was easy because I had been doing that. That was really, I totally knew how to get the whole business side of the business set up. I think the biggest challenge for me was for a decade, Joel had been the voice of our products and our company.

Josh: You had been sort of behind the scenes, really.

Michael: Yeah, right. Now, it was like oh, there’s somebody from the press who wants to talk to the CEO of Trello. I was like oh, I have to-

Josh: That’s me.

Michael: I would have to be on video or go on stage. It’s funny because it’s the same thing again. As an entrepreneur, you’re like oh well, I guess I’ll just figure it out. You just do it and eventually, after you’ve done it a bunch, you’re like oh yeah. That’s not that hard. If I’m really talking honestly and saying things I believe, then it’s easy. If I have to put on a show, that might be hard for me because then I would have to practice and rehearse and know what I was going to say beforehand. I found talking to people, when I go to conferences, I love doing Q&A because it means I don’t have to prepare. I just get on the stage and I’m like sure, I’ll talk about anything.

I think for me, that works really well with my personality type and people, I think they like that openness. The transition to CEO was first can you be, so there are the three things the CEO does. This is what Joel told me. He says you got to keep the vision. You got to not run out of money, and you got to hire the best people. That’s it. That’s all you do. The not run out of money part was pretty easy. I mean, at least I knew the scope of the problem and how to approach it. The hire the best people was relatively easy given that Trello was so popular and we were hiring people remote, so the recruiting was not that difficult.

I think the keeping the vision was the trickiest part and something that I’m still working on now. I think it’s pretty easy for me to talk to PR and the press and outwards, but it’s … Being coherent in the story internally to the company, it’s easy when it’s a group of 10 people and you can sit down and have lunch and everyone’s head is in the same place. But once you get past 100, the communication has to be a little bit more … Written communication is much more important. Editing that down and being consistent in what you say, so it’s something I still am perfecting.

Josh: You mentioned the vision part. Trello’s model from the beginning was to have this huge number of users. You mentioned 100 million. Ultimately, you’d charge some smaller percentage of them. That 100 million user number, what was the driving force behind picking that number? Was it the revenue part that would come from charging a small percentage? Or was this desire to build something that touches the lives of a huge number of people or something else?

Michael: I think it was just a big number. It set the scale of the thing that we wanted to build because if you’re trying to get 100 million people, well one, you have to build a global tool. Your software is going to be used globally. Your market is the world, not the US, not the English speaking world. If you want 100 million people using it, you’re not going to be able to target just sales people or just marketers or just HR. You’re going to need everyone at the company to be able to see value in your tool, and probably people in their own personal life to see value in your tool.

That was more setting the stage for who is our market and where does this tool fit. The 100 million, 1%, $100 million, it’s all just guidelines for yeah, there’s going to be a lot of people that get value out of our product that maybe don’t pay us, but we’re going to focus on the people that get a ton of value. The value that we’ll get from the other 99%, whether it’s 97% or 90% or 80%, the other percentage will be those people will be fans and advocates for us and continue that growth.

Josh: Was there a singular inflection point where the number of users got really big for you guys? Same for revenue?

Michael: No. It just sort of grew over time. Things would happen and you’d get a bump up in new users per week. Then it would just stay there. Then that would get higher. For example, a couple years ago, Life Hacker wrote a big article about Trello. It was very long article and it came out in the new year which was also a time when people come back to work and want to get organized. It was this boom, we saw tons of signups. We naturally see that seasonally, but also because of this article, and so huge spike in weekly signups, but then when it came back down, it settled at a point that was way higher than previous points.

This is because we have never spent a dollar on acquiring customers. We don’t do any advertising. We don’t do anything to … We would do some content marketing. We employ marketing people, but the people that coming in to Trello are coming because somebody else told them about it. If we get more sign ups, then that begets more signups which begets more signups. I don’t think there was one point in time. There were just these spikes that you would see, mostly around press or big feature releases that we would do. Then they would, they’d come back down but they’d be a little bit higher. That would just continue to go up.

Josh: Was there a day, whether it’s in one of these inflection points, like a Life Hacker article or something else, was there a specific time you felt like you realized okay, this is not just some little tool? Our 100 million user vision actually could work out.

Michael: Yeah. There was. There was a couple moment like that. I think one of them was, so when we launched at TechCrunch Disrupt, we got pretty … We were in this competition. It was called Startup Battlefield where you could pitch your idea and then I think Roelof from Sequoia and Marisa Meyer was on stage. Michael Arrington was on stage. They were judges and we got to the final round, came in second to this company that basically ceased to exist after a year, but after that day at TechCrunch Disrupt, our signups went through the roof, a huge blip on the graph. They settled way below that a couple weeks later. Then it grew and it grew and it grew. It was like there was a moment where every week, we were signing up the number of people that we had sign up during TechCrunch Disrupt, and that was kind of … When you looked at it, you were like whoa. That-

Josh: That became the new norm.

Michael: Yeah. You’re like every week is a TechCrunch Disrupt. Then it became every day is a TechCrunch Disrupt. That was, it took a couple years for that to happen, but it was a big deal.

Josh: The inverse of that, was there ever a time that you thought no, done, not going to work, like some epic mistake or that felt detrimental?

Michael: No. I think that whenever … I think competition can always feel very scary, like when people copy you or when people do something very similar, or you hear about successes that people have, even if you think that they’re not anything like you. I’ll think back to when Wunderlist got acquired by Microsoft and it was like oh no, Microsoft’s getting in this game. They’re going to come after us, even though that’s a to-do list and I would … If anyone called Trello a to-do list, I’d throw some water on them or something. I don’t know. That would be the biggest insult I would ever hear if somebody said that. Today, they just discontinued Wunderlist, and so that was kind of interesting.

There was a lot. Over the last year, I think people saw a lot of our success and so you saw this plethora of people that are building project management tools or adjacent to project management tools would be like oh, we’ll just add the Trello feature, which is we’ll just add boards. That’ll put a stop to Trello. Maybe they’re right. I don’t know, but I think it’s a little bit … In one sense, you’re like oh, that’s kind of scary because those are, that’s a lot of people that are adding that feature. But you spend too much time watching the competition, and I think it distracts you from building an awesome tool.

Josh: I think it makes problems feel bigger than they really are, sort of a big distraction.

Michael: Right, right.

Josh: Last year, you got approached by a little company called Atlassian that wanted to acquire you guys. How did that even come about?

Michael: We’ve been competing with Mike and Scott, the founders of Atlassian for like 13 years. We released FogBugz back in, very early and they came out with Jira two years after that. For a while it was a very fierce competition, neck and neck, and they plowed a lot of their profits back into the company and grew really quickly. Also, I mentioned that marketing muscle before. They were really good at traditional marketing. They grew to be this huge, huge company, IPO’d. It wasn’t this out of the blue contact, I don’t know who these people are.

I think it was either Scott or Mike sent an email to Joel and said that they were going to be in New York and wondering if I was around and I could have lunch. I said sure, I’ll meet you for lunch. This was back in the fall at some point, so last year in the fall. I was thinking about raising another round. We had only done one round of fundraising for Trello, A round, so I was in the midst of talking to VCs about B round. I kind of knew what the conversation was going to be about.

I sat down with Mike and at this little hole in the wall Mexican joint near our office and had lunch. We talked about Trello. It was kind of interesting because all the questions that he asked me were about the vision for the product, like where wanted to take it, about the culture, the people that were working at the company, the kind of company that we built, and it was not what I was expecting. Here, I’m expecting he’s going to show up and be like well, we want to buy your company and here’s the offer. Here’s the numbers and here’s what we think it’s worth or whatever. We literally didn’t talk about that at all, such that when I left lunch, I was kind of like wait, was that, were they trying to acquire us? I don’t know because I didn’t get any details.

They had done like 18, 19 acquisitions already. A lot of them were smaller than the Trello acquisition, but they had learned a lot from those acquisitions. I think when they went into this, the things that they knew about how successful Trello was. Trello was successful doing what it was doing. They weren’t going to show up and try to tell us to do something different, but also they wanted to make sure that our vision was aligned with their vision. There was this one moment when he’s talking to me and he’s like … I sort of talked about the 100 million number. It’s like we’re trying to build this product for the entire world and we want, since day one, we’ve been trying to get 100 million customers or 100 million users. He starts laughing. I thought he was laughing at that number. I was like wow, how rude. I was like-

Josh: He was a jerk.

Michael: He’s like no, no, no. We had this goal internally. We have these big, hairy, audacious goals to focus the company around. A long time ago, we had 60,000 customers. We thought that was a huge number. About a year ago, we passed it and so we had to come up with a new number, a new goal. We came up with we wanted 100 million users to use our products. He’s like you could go look. It’s documented. I’m not, I swear I wasn’t copying you. It was a funny moment but it was also very clear to me that as you dug in, they started their company in the same era that Joel and I started our company. They started it in a very similar way. They were almost like our doppelgangers from Australia. The way that they built the company, the culture that they’ve built, the values, the way that they treated their people, the goals that they had, and the way that they ran their business.

There was tons of alignment and tons of things where you’d go in and you’d be like oh, they have this problem. It’s like yeah, I know exactly why they have that problem. I had that same problem. It felt very familiar and it still feels very familiar. I think that was one of the biggest components to this acquisitions, was is this going to let us keep doing what we want to do and build the kind of product that we want to build. That was a resounding yes. Of course, we also had to make sure that it was going to be a good financial outcome for the employees and the investors and that sort of stuff, but that’s how it happened.

Josh: I can’t imagine this was the first time someone had approached you guys about or at least with some sort of interest in acquiring Trello. I guess was it just the alignment of visions mixed with the right price tag? Were they really the only ones who ever showed legitimate interest?

Michael: I’ll give you a non-answer answer, unfortunately. I think any time you’re in that situation, you’re running a company, you’re trying to build a very valuable company. The way that that value gets expressed can be many different ways. You can get acquired. You can build a great company to work at that pays your salary and a whole bunch of people that you love, their salaries. You could go IPO and extract value that way. The way that you extract the value, that’s not the goal. That’s just a mechanism once you’ve built something that’s really, that people see value in. They are multiple ways that you can extract that value.

I think that one of the things that I thought about was I had done a bunch of reading about people that had sold their companies before and the things that they afterwards, and whether people were happy or not. One of the things that came up over and over again was that the people … People that start companies. They do it because it’s just inside them. That when they get up in the morning, the way that they … If they stop doing that, like there were a bunch of people that sold their companies and went off and traveled the world or something for six months, those people were inevitably always depressed afterwards. They just lost their purpose.

When I thought about this and thought about the opportunity, part of it was trying to figure out are we still going to be able to do what we set out to do. Is Atlassian going to help us do what we wanted to do? Also, is this a good outcome that had invested in the company and the employees? In fact, when we spun off Trello, the cap table for Trello was a copy of the Fog Creek cap table. Everyone that had owned equity over the last decade in Fog Creek had ownership in Trello because of the way that we spun it off.

This looked like a good outcome, but also for me, it was really important that these were people that I wanted to work with and that we were going to keep doing what we were doing, because I didn’t feel like we were even close to being finished. That felt, talking to Mike and Scott and just that history and the places that they had been and where they had come from, I felt that alignment and it just all came together.

Josh: You’ve now transitioned to Atlassian. After 15 plus years of building your own companies, you’re essentially now an employee at a public company. What’s the transition for you been like personally? Is that been a hard adjustment or just sort of a non-issue?

Michael: There’s a lot of … We were a startup, 100 people. Atlassian’s a company that does many, many different things, 2,000 people, bigger company. What we’ve tried to do is there is a policy in place right now which is do not disturb and trying to let Trello do what it needs to do without slowing it down. As I talked before about the acquisitions and learning from the previous acquisitions, I think that they’re adamant that they do the right things to make sure that Trello’s growth continues on the path that it was on. They would only do things that would accelerate the growth, not slow it down.

Somethings are inevitable. We got to switch payroll systems. We got get a new health insurance. You go through this period of oh, I got to come up on speed on how I submit a health reimbursement expense or something like that, so there’s that stuff that we had been through. I think the acquisition closed on February 2nd. Now it’s been like two months basically, and I think we’re coming through that, but generally day to day, not much has changed for pretty much, for most of the people at the company.

Josh: I mean you guys are still same offices and everything, right?

Michael: Yeah.

Josh: Gotcha.

Michael: They’ve been very clear to make sure that the streams of … Sometimes you can have the best intentions. The people at Atlassian have the best intentions and want to help, like really excited about helping and learning, but the problem is there’s 10 of them for every one of us, or actually there’s 20. 20 to one. You can get just overwhelmed. You have to do these things very slowly. I think we’re benefiting from the fact that they’ve done these acquisitions before and learned so many lessons from that. I think day to day, it hasn’t really changed that much. I might use different tools and I might submit my reimbursement in a different way, but as far as building the features for Trello and what we’re going to do in the future, it’s the same as it was six months ago.

Josh: To wrap things up, what do you feel like the future holds for Trello now that maybe … Is it any different than it was pre-Atlassian? The 100 million user goal sounds like it’s actually still probably the same.

Michael: Yeah.

Josh: Has anything otherwise changed as far as end goals?

Michael: I think one thing that’s changed is, I said keep the vision, recruit really good people, don’t run out of money. The don’t run out of money part becomes a lot easier. It’s way easier for us to, if our burn was like $250,000 and we had $2 million left in the bank, it’s like okay, we only got eight months of runway. In a $6 billion company, if they want to spend $10 million in investment, it’s like okay. We just decide to do that. It changes the dynamics around that piece. It’s much more strategic. You’re not worried about hey, can I just keep going for three more months so I can get better numbers so when I go to the VCs, I have better charts so I don’t get as much dilution. I get a value … It just takes that piece out of the equation. It’s like well, what do we want to do? What can we do? What’s going to move the needle the most? And let’s focus on that.

Josh: Cool. That sounds awesome. Well, that’s all I’ve got Michael. Thanks for hopping on a call, man.

Michael: Thanks Josh.