June 4th, 2018 | By: The Startups Team | Tags: Stories
Very few entrepreneurs have been are able to successfully create a viral, cult-like following around something as random as workplace collaboration software. But, if anyone could — it would be Stewart Butterfield, a two-time failed gaming entrepreneur.
When headlines circulated that Amazon was looking to spend some $9 billion to purchase Slack, it was another sign that this hot app wasn’t some overblown unicorn flash in the pan.
While Slack has a high valuation, it stands out from the current startup landscape. In a world of 24/7, party all night brogrammers, Butterfield says his offices are empty at 6:30 every night, and people work ~45 hours a week. He’s designed the company for people who have already lived the insanity of a boom and bust cycle but want to continue to build things — but, in a sustainable way so they can still see their kids grow up.
In other words: People like him. It took a unique journey to build a unique Valley leader. The following is Butterfield’s story of accidental success, crushing failure, and finally a bona fide financial hit on his hands.
The Startups Team: Tell us about where you grew up.
Stewart Butterfield: I was born in a small town called Lund in Northern British Columbia. It’s literally like the end of the highway that goes all the way from Chile, up the coast here, past Big Sur, and it just stops.
The Startups Team: You were the end of the line.
Stewart Butterfield: End of the line, my parents were “live off the land” hippies. My dad is from New York, and my mom is from Montreal, so they really had no f*cking idea what they were doing. They tried to grow wheat in a rain forest and tried to raise pigs and stuff like that.
Also, my grandfather, my mom’s dad, came between the wars from Poland, didn’t speak any English, only had 10 bucks in his pocket, blah‑blah‑blah‑blah‑blah. Worked his ass off, sent his kids to university, and then they end up going and living in the backwoods with a bathtub in the front yard, no running water and no electricity…
We moved to the city when I was five, so I could go to school. That was Victoria, British Columbia, and I was in second grade. We were the first class to have computers in the classroom. I was super fascinated by computers in general.
The Startups Team: What was it that you liked about computers?
Stewart Butterfield: I was seven, eight, nine years old and I really liked video games. It’s a weird thing. We see it now, and you have a three‑year‑old, and it’s impossible to resist a screen. Cal Henderson, my co‑founder, I was just having brunch with him and his wife on the weekend, and they have a five‑month‑old, I think. Even then, if Mom’s breastfeeding and on the phone on the side, he’s like, “I’m starving, but I want to see what’s on the phone.”
I don’t know what that is. There’s some primal thing that computers light up.
We’re so close to the beginning of a pretty big shift of what we are as a species. We don’t really think about it that way, but evolutionarily, pretty powerful stuff. We’ve able to communicate with anyone on the surface of the planet, no matter where they are, literally at the speed of light. I think it’s like the frog boiling in the pot of water, gradually heating up, we kind of don’t notice that stuff.
The Startups Team: What do you think that means, evolution as a species? What do you think is going to happen?
Stewart Butterfield: I don’t know. I think that other species really have no chance now. We have been breeding them just to kill them and eat them for a long time.
The Startups Team: The thumbs were just the beginning of it.
Stewart Butterfield: Yeah, exactly. If snakes had our technology, then we would be toast.
The Startups Team: Thank God they don’t. I hope they don’t. Do we know they don’t?
Stewart Butterfield: We know they don’t.
The Startups Team: Did you continue to do stuff with computers?
Stewart Butterfield: In high school, I didn’t really use computers that much. It wasn’t until I got to university ‑‑ this was ’92 ‑‑ I got a Unix account on the school’s mainframe and discovered the Internet. I was just like, [explosion sound ]. I realized at that time how powerful it was. Whatever you were into, if you were a breast cancer survivor, you’re a model train enthusiast, or whatever your deal was, you could find other people online.
In ’97‑’98, I finished my master’s degree, and I was going to go on to do the PhD, but ’97‑’98 was the very beginning of the dot.com era. A friend of mine finished his PhD, got his first job at the University of Louisville in Kentucky. He didn’t want to live in Louisville. It was a terrible job. It didn’t pay very much, and there was no job security because he had to renew every nine months. Meanwhile, friends of mine moved here and were making twice as much money as him, having fun, and doing all this cool new stuff.
The Startups Team: Changing the world.
Stewart Butterfield: Yeah, and I was good at HTML, so I said, “All right. I’m not going to do it. I’m going to go work on Internet stuff.” So ’98, ’99, I went back to British Columbia, and ended up head of design at the biggest design agency there. It’s a provincial capital, so it’s not a big deal, but we had maybe 80 people on staff, and 10‑15 million bucks in contracts.
And then, sh*t went bananas in ’99, right into 2000. This is an interesting story. The guy who started this company in 1993 registered a ton of domain names, keyboard.com, mouse.com, brazil.com, vietnam.com, makeup.com, and hundreds of them that were suddenly big assets.
This is when Barnes & Noble bought books.com for $3 million or something like that, and someone bought business.com for $6 million.
For some reason, he rolled them into this company he started, which was a Web design agency, because this is 1999 mentality. We have all these domain names. Makeup.com is almost infinitely valuable. It’s much more valuable than Revlon as a brand, even though Revlon is an $8 billion company and everyone knows Revlon, but no one’s ever heard of makeup.com.
We are going to build a network of e‑commerce sites with these different domain names as the front end, and one common technological infrastructure behind them. All the clients that we had painstakingly sold on our services, and gone on sales trips and stuff like that, we now had to go and say, “Sorry. We’re in the middle of building your travel booking website, but we can’t finish it. Find someone else. See you later.”
No one knew what they were doing at all. If you have a company that’s 60, 70 people, and then you lose what it is that you do as a company, then it goes totally feral, like “Lord of the Flies.”
Everyone’s jockeying for position, trying to figure out what they are. You remember how weird of an era that was? You looked at companies where every single job title had “architect” in it. There was a data architect, there was a design architect, there was an experience architect.
Things went sideways. Every management meeting was a screaming match. Literally, two weeks before the crash, I couldn’t take it anymore, and I quit. I couldn’t stand working there anymore. I really, honestly, thought I was walking away from $10 million in equity, because…you remember what that was like.
The equity agreement was such that they forced the buyout. I got $35,000, and it was $35,000 more than anyone made who stuck around, because two weeks later, it was March of 2000 and [explosion sound] , Whoa.
That was an interesting time for me, just to watch it. It happened all at once, in terms of the public perception, but it took a long time for that to play out.
In 2000, we decided that we wanted to try to make a web‑based massively multiplayer game, which is a terrible idea. I’ve tried it twice. It doesn’t work. Don’t try to do it.
The Startups Team: Now, are you going to listen to yourself this time?
Stewart Butterfield: I’m done. I’m not doing that anymore.
It was myself, a guy named Jason Classon, who’s the guy who started Gradfinder, Caterina Fake, who I’d just married, and a couple other friends of ours ‑‑ Eric Costello, who went on to co‑found Glitch and Slack and all that stuff with me.
We were all fascinated by online communities in the broadest sense, just the human connection that can come out of this transmission of electronic bits back and forth. We were also fascinated by the role that play played in social interaction. My dad plays bridge. He doesn’t like to play bridge against a computer, because it’s kind of soulless, but he also wouldn’t invite the same three people over to his house just to hang out with no agenda.
When you’re playing a game, it’s competitive, it’s trash‑talking, it gives you like this kind of framework for interacting with people. It’s fun to play games with people.
We wanted to create that on a massive scale. It was primarily text based. We pioneered a lot of really interesting technologies that became Flickr afterwards. Actually, I don’t know whether it would have been successful or not. It was the middle of 2002 when we started and then into early 2003, and it was literally impossible for anyone to raise money for any kind of Internet anything.
We tried. A friend of mine who had an inheritance when his grandparents died put a little bit of money in. I put my money in. I convinced my parents to give me a little bit of money. But it was really that scraping together friends and family.
We got grants from the government, because the Canadian government likes to give grants for this kind of stuff. We got to the point where Eric, because he had kids, was the only one who was getting paid. We were desperate to try to find something else we could do with the technology that we had already developed that would be simpler and have a shorter timeline for development than this game would.
We had a prototype. The prototype was very well received, people loved it, but it would have taken us another year and a half, we figured, to build the real thing. We just couldn’t live and pay rent without an income.
This sounds ridiculous, but it’s true. I was flying to New York to go to a conference on law and virtual worlds, and on the plane, I got food poisoning. I got to JFK and was in the immigration hall and was like puking all over the immigration hall.
Got in the cab, made them pull over in the Van Wyck and was puking. Got to the hotel, stepped out of the cab and puked all over the carpet in the front of the hotel. I couldn’t sleep or anything. I was sweating, trying to drink ginger ale. The whole idea for Flickr came to me all at once.
There was a shoebox of photos, and technologically super cool. It was pretty ground‑breaking. You could drag a photo onto a person. It would pop up in their screen, and you could talk about it live. It had a real opposite of the network effect, a cold start problem, where it was great if you knew the other person on the other end of the line.
Eric worked in New York and he would upload photos of taking the family to the beach for the summer holidays and we would talk about them in real life. You could annotate little regions of the photos and stuff like that.
But the odds that someone you actually cared about would be online at the same time as you on this brand new social network was zero. If you log in and you’d be like, “Oh, well, nothing for me here,” and log out and never come back. It’s kind of like trying to start an instant messaging app or something like that in the time when everyone used Jabber Messenger, AIM, and ICQ.
But it evolved it really quickly, and we started getting some traction. There was a bunch of happy accidents. One of them was we knew all the people at Blogger. Blogger didn’t have a photo hosting capability at the time. In fact, there was no one on the Internet, there was no Imgur, so you couldn’t upload a photo and link it from anywhere else.
One of the things you wanted to do was let people use Flickr to host photos on their blog, but you had to make a link back to Flickr for the photo, and we did that. On Blogger, if you clicked on Help and said, “How do I upload a photo,” it’d say, “We don’t do that. Use Flickr instead.”
I think also because I was from this community… I was one of the maybe 200 first bloggers ever. It was a really tight community of people. I still know most of those people, where every blogger knew every one else’s name. Because of that, we got a huge amount of, at the time, very famous bloggers using us.
The Startups Team: That’s interesting. In those early, early days before anyone was even calling it Web 2.0, companies like yours, Blogger, Delicious, and a few of these other ones were so nascent. It was really deemed as something insane. Everyone was so scorched earth against consumer Internet, at least out here, and I imagine everywhere else, too, even more extremely.
Was there something really special about building that company? The people doing that were so small and so tight, and you could have a casual relationship like that, that would wind up driving huge usage, because there wasn’t this crazy fragmentation.
Stewart Butterfield: Yeah, it was, and people who are friendly tend to help each other. There’s a big boost. Also, it was kind of nice, because the initial thing that pushed us to start Flickr was desperation, because we had to pay rent.
But once it started going, we actually did get a little bit of investment. We got Reid Hoffman to invest. We got a guy named Brett Bullington, who was a pretty well known angel as well. Esther Dyson, who was maybe the best‑known angel investor in the world at that time.
Anyways, we got some good support and it was fun. It was like, not being a rock star in the sense that you have groupies, but like playing music with other really talented musicians, or like working in a really well‑run restaurant. Those are still my favorite analogies for making good software.
And it was exhilarating. Every day we would log in, and we would grow like one percent a day, which doesn’t sound like very much but it compounds pretty insanely. We would double this month, double the next month, and then double the next month.
It was “the thing.” For that end of fall of 2004, into the new year 2005, it was the only cool thing on the Internet or that’s what it felt like. There was Delicious. Facebook hadn’t opened up yet, so there wasn’t much. It was pre‑Digg. MySpace was just starting to ascend. It was really fun. It was an awesome time.
The Startups Team: The companies that came right after that, like Digg, Yelp, and YouTube, and a lot of those, they would always point to you and Delicious. Those were sort of the two, because you guys had those two exits, which seemed crazy at the time.
I was working at Business Week and I remember being in New York. People were like, “Yahoo bought Flickr for $30 million! $30 million!” People thought it was crazy. What was that acquisition process like from your point of view? Did you think it was crazy?
Stewart Butterfield: No, I didn’t think it was crazy. I don’t have the chronology perfect, but we had started charging money a couple months before, and we were close to break‑even. It was on a growth curve and a trajectory. It had the attention of enough people that it was obvious that it was going to be successful. when Flickr was acquired, and that was, we announced in March of 2005, that was almost 10 years ago exactly.
It was the only thing like that on the Internet. The only thing that had been an exit, and it was even until about a year after Flickr was acquired that YouTube was founded. People forget how far back and people give Yahoo a lot of sh*t for “They totaled Flickr. Flickr was so awesome before Yahoo bought it.”
Almost everything people know about Flickr happened after the Yahoo acquisition, maybe in those first 18 months, two years after the acquisition. Yahoo didn’t do as good of a job as it could have done in supporting Flickr at that time, but we were nine people when we were acquired. We had barely any of the features that people knew us for.
The Startups Team: So you actually think Yahoo wasn’t a bad steward of Flickr?
Stewart Butterfield: No, I think it was a bad steward to Flickr. I do think that it wasn’t like Yahoo bought it and then instantly it turned to sh*t.
The Startups Team: So it was a patient bad, a cautious bad steward.
Stewart Butterfield: That was really eye‑opening for me, because it was the first time I’d ever worked in a the big company. I’d done a lot of consulting work for really big companies, but I had one point of contact and met maybe eight people at the most, but usually three or four people that I was working with.
Now, I arrived at this 10,000‑person company and all the terrible software that we had to use inside it, which has been in my head since then. It’s not like that was unique to Yahoo. Enterprise software on the whole, generally, still is pretty terrible. It’s getting a lot better, but it’s still not as good as it could be.
Yahoo was kind of messed up. There was a lot of senior vice presidents, all of whom at least had a significant portion of their motivation coming from defeating the other senior vice presidents, as opposed to making the company successful.
The Startups Team: It was like a slug fest for pixels on the home page.
Stewart Butterfield: Yeah. It was chaotic, because of management in‑fighting and stuff like that, so it was really hard. Yahoo News was by far the biggest news site of the Internet. Yahoo Finance was the biggest finance site. Yahoo Sports was the biggest sports site, and so on.
They were used to, if you grew 20 percent over the course of the year, that was really good, whereas, we were doubling every month. The three‑month timeline to order a new hardware didn’t work for us. The nine‑month budgeting process to get new headcount didn’t work for us. We had to hire people now and get hardware now.
It was super exasperating, because we would have literally 30 people in this room for a hardware review to figure out whether we should order new servers. The answer was “Yes, we should have new servers because otherwise the site’s going to go down.” There was just a lot of that.
It wasn’t any one particular person’s fault.
The Startups Team: A lot of the people that seemingly you met at Yahoo! wound up being angel investors in Glitch/Slack, Jeff Weiner and Rob Solomon and some of those people, right?
Stewart Butterfield: Yeah. Because I spend half my time in Vancouver ‑‑ we have the office in Vancouver and we have the office here and I go back and forth ‑‑ I get asked a lot, “What can Vancouver do to become the next Silicon Valley?” I’m just like, “It’s impossible.” If you want to do inter‑bank finance, you have to go to London. If you want to start a giant media company, you have to be in New York. If you want to do movie production, you have to be in LA.
That’s a great example because the people, the immediate circle of people I worked with, one of them was Andrew Braccia, who left Yahoo before I did, became a VC at Accel, and was our first institutional investor, and supported us all the way.
His boss was Jeff Weiner, who’s now the CEO of LinkedIn. Rob Solomon was the CEO of Groupon. James Slavet became a VC at Greylock, Bradley Horowitz, vice president at Google. Everyone, the people who sat next to me is a valuable contact.
If you’ve worked at Google, Yahoo, eBay, PayPal, Amazon, or wherever ‑‑ Facebook ‑‑ you get this network, and the people that you sat next to go on to work somewhere else. When you need help from that company or need an introduction, you have it. It’s so powerful.
The Startups Team: Why did you decide to sell, though? There’s obviously a lot of uncertainty at the time, generally, when it came to consumer Internet. It was not like today at all. That said, you said earlier that it was clear that this was going to be a successful company. You were close to breaking even.
Stewart Butterfield: The acquisition price was actually less than the $30 million that was reported. For whatever reason, we decided we weren’t going to reveal the actual price. I don’t remember what the actual price was, but let’s say more like $25. That was pretty f*cking successful at the time. In the context, $25 million is a lot of money. It doesn’t seem like anymore, because of this crazy world that we live in, in this particular geographic region. The advice that we got, and I think it was good advice, from Reid, from Esther, and from Brett was, “Look, these people at Yahoo are good people,” and many of them actually are good people, “and you never know when another plane is going to fly into the World Trade Center or there’s going to be an Asian currency crisis or any of the other things that could happen.”
Again, there were no successful consumer Internet companies for that whole period, from 2002 to early 2005. There was no example. “You sell now. You’ll learn a lot going there. It’ll still be successful. You’ll have enough money that you can decide what you want to do next after that, and that’s a pretty powerful thing. And so there’s not really any downside.”
We didn’t have like now the model that Evan Spiegel can say no to $3 billion because he thinks he can take it higher. That wasn’t the mindset that anyone had at that time.
The Startups Team: Yeah, and no one was doing partial liquidations at that time, either. You couldn’t take a little bit of money off the table to keep growing it.
Stewart Butterfield: No. It’s funny. The change in VC terms makes a big difference, too. We had a bunch of acquisitions, offers, and a bunch of interest, but the two it came down to was Yahoo or I think it was $5 million from Accel.
In retrospect, obviously, we should have taken the VC investment. We should have gotten longer and all that. It’s not like I have any regrets about that. The terms were not good. It was $5 million, but on terms that would be ridiculous today. I don’t remember what the details were, but a very high liquidation preference, very strong anti‑dilution provisions, and stuff like that. It was much less attractive.
It was also a bunch of squishing down the original investors, the outside investors. If it was better terms, it might have gone a different way. MySpace was a ground‑shaking acquisition at $680 million, “heard around the world” kind of thing. It took a while before…YouTube I think was late 2006?
Then, YouTube at a $1.65 billion seemed absolutely mind‑blowing, but then this whole trajectory started.
The Startups Team: Do you worry at all? You look at the list of billion‑ dollar companies and it’s something like 0.1 percent of companies ever exit for north of a billion dollars. It seems to me there’s this 1999 thinking out there that, “No, we’re just going to have more billion companies than we did in the past.” Do you think that’s the case or do you think there’s more bets on who those billion‑dollar companies will be?
Stewart Butterfield: I think it’s a little bit of both. I don’t know how inside baseball people want to get, but interest rates are at historic lows and aren’t going up anytime soon. People need to have a return on their money. There’s a lot more money in equities generally. There’s a lot more money in alternative investment categories broadly, and that means there’s a lot more money in VC funds.
VCs that would have raised a $350 million fund eight years ago, now raise $1.5 billion. They can’t invest in five times as many companies as they used to do because it’s just not feasible. It’s not possible. So they have to write bigger checks, which means that valuations go up, and a bunch of other stuff like that.
At the same time, people are asking me why Slack has taken off. I got a lot of headlines, a couple of weeks ago, for saying, “I have no f*cking idea.” I have a little bit of an idea, but I feel like it’s very speculative and also it will take me 45 minutes to explain it, so we don’t need to get into that.
Part of it is this shift. When I think back to 10 years ago or even 11 years when we started Flickr, hardware is 95 percent less expensive now than it was then. Open source software generally, across the board, is 50 times better than it used to be. More importantly probably, in 2004, there was a couple of thousand people who had experience building Internet‑scale applications.
You had to have worked at Yahoo or eBay. Even Google wasn’t quite there when we first started. Google was still significantly smaller than Yahoo. Now, there’s lots of people. There’s tens of thousands of engineers who have had really good experience. They know what they’re doing. The technologies are much more robust and mature.
People have done experiments, documented the results, and shared ideas at conferences. There’s much more of a community. Even more importantly, there’s literally 10 times as many people online, but they’re also online in a different way than they were 10 years ago.
I remember, back in 2000, after I got my $35,000 after I quit that company, I bought the very first AirPort and the very first PCI card, AirPort card for my laptop, basically so I can move four feet off my desk to my couch and sit on my couch and use the Internet. It was like maybe $600 or something like that.
That was like a miracle.
The relationship people have to each other, as mediated through technology, has changed in a really profound way. It’s not obvious to us now, but maybe in 10 years or so, we’ll be able to look back and say, “Wow, from 2008 to 2014, sh*t really, really changed in a big way because of the iPhone.”
Now, my mom sends me texts. I don’t mean to use my mom as a canonical example of someone who isn’t technological savvy.
The Startups Team: She was a hippie living off the land, though.
Stewart Butterfield: She was a “living off the land” hippie. Everyone does it, and it’s not weird. I used to say this all the time about Flickr, in 2006‑2007, going back 10 years from then to the mid‑, late‑’90s, the whole idea of someone using Internet at all seemed creepy, weird, and deviant.
If you talked to someone on a chat room, that was pretty suspect stuff, like you’re probably a pedophile or something. Now, you see the commercials all the time, grandma FaceTiming with her grandkids. That’s just becoming a normal, everyday thing. I’m not even sure if I’m identifying the right individual characteristics, but a billion people on Facebook, everyone in the world using messaging.
The Startups Team: More people having phones than have toilets.
Stewart Butterfield: Yeah. At least in developed world and much of the undeveloped world, or the less‑developed world, fantastic Internet speeds, no matter where you go. That changes your relationship to technology. Going back to being able to speak to anyone, communicate with anyone literally at the speed of light. Something about technology changes.
The Startups Team: This leads into Flickr/Instagram, and then we’re going to get into Glitch and Slack. Was the big different between Flickr and Instagram mobile? Would you have had $1 billion photo sharing app in a pre‑mobile world?
Stewart Butterfield: We forget that iPhone was announced in 2007, in January. It didn’t actually get released until the middle of 2007. I don’t know how many people remember how crappy the first iPhone was. There was no App Store. That wasn’t the thing that was released in the first version. There’s no MMS and all this stuff.
It took until 2008 that there was an App Store. It took until 2009 that there was even really the concept of apps. Now, it feels like it has always been this way. At the time that Flickr was acquired, the coolest phone that you could not get, because it was flying off the shelves and they couldn’t meet production capacity, was the Motorola Razr. People would take it out of their shirt pocket and put it on the table when they’re at lunch, be like, “I have one of these.” People would flip it open and that crappy blue LED, which just looks so ridiculous. That was 10 years ago. That really was the absolute height of our mobile technology, so yes, definitely mobile make a huge difference.
That started to make a difference about five years after we started Flickr.
The Startups Team: All right, so I want to make sure that we get to Glitch and Slack. You failed trying to do a big game the first time. Was the thinking, “Well, we didn’t really get to run that experiment because we ran out of money and because circumstances were different, and so now, let’s try it again with crazy resources?” Or was there some bigger insight in that that made you think this time it’ll work?
Stewart Butterfield: No, it was really about, A, we’re all a lot more experienced than we were. We’re a lot more capable and smarter. We had learned a lot in that process. Everyone wanted to give us money. We could raise money instantly and easily. All the things I said about hardware being cheaper.
There just being AWS. All those factors have made it easier. People’s individual computers are way better. People have way better Internet connections.
It was not at all obvious, even in 2009, even in early 2010, how much of people’s discretionary computing time would shift to mobile. I know that sounds crazy now, because we started building the client technology on Flash, which is one of the things that just made impossible for us to port and was a total catastrophe.
So we missed that, but in every other respect, I think it was…Computers are more powerful. People have Internet. There’s more people online. Hardware’s cheaper. We’re better. Open source software is fantastic. Every part of what we would have to do was fantastically easier. I think partially we overreached because of that. We got a little bit too ambitious in what we’re trying to build.
This is something that I have failed to learn many times, and I think I’m starting to get it. This is going to be my only piece of advice for entrepreneurs besides be really lucky. If what you’re doing is something you can’t explain to another person such that they can go and reliably explain it to someone else, it’s probably hopeless.
Let me try to explain Glitch, the combat‑free, massively multiplayer game that’s absurd and surreal, where you collaborate with other people in building an environment while…it’d take me like 15 minutes to…
The Startups Team: Yeah, every description I read of Glitch just sounds like it’s a sketch in a comedy show or something. It just sounds like it’s like “Brewster’s Millions.” It’s like you raised all this money, and they were like, “You’ve got to deplete all of this money without a successful project in a year, and then you will win all this other money.”
Stewart Butterfield: If you were someone who was a fan of online culture generally, you were a heavy Internet user, and you were intelligent, then Glitch was fantastic. It was a great community. The people tended to be really creative. Unfortunately, most people are morons, and have terrible taste and stuff like that.
That was definitely a factor. but also I think for games, in particular…this was at the time when Zynga was starting to take off, but even before Zynga, there’s like this matrix of, “What type of game is it in terms of mechanics?” You have real‑time strategy, first‑person shooter, and simulation, or whatever. Then you have this other axis, which is the window dressing you put on it.
Stewart Butterfield: So it’s World War II or Post‑apocalyptic SciFi, or something. That’s Dungeons & Dragons kind of…
The Startups Team: Gangsters.
Stewart Butterfield: Yeah. You put those two together and you’re like, “Real‑time strategy game based on World War II? I’m in, that sounds awesome.” You can almost make the purchase decision in advance of even trying the game. People had a really clear concept of what it was and whether it would appeal to them.
Again, if you try to explain this thing, it’s a massively multiplayer game, but unlike World of Warcraft, there is no combat. You cooperate with other people. It’s absurd and surreal and everything. It’s like Monty Python meets Dr. Seuss, intellectual net person in paradise.
We had like 100,000 players who were super committed. The average revenue per paying user was off the chart, like 80 bucks or something like that, but it was very obvious that it would never become the kind of business that would justify $17.5 million in venture capital investment. It would just never have that kind of return.
If we had done it on the cheap, and we wanted to do this as a lifestyle business because we liked it and we enjoyed interacting with the community, it would have been fine. Maybe if we had switched off Flash earlier and had some foresight about what was going on with mobile, maybe that would have saved it. Maybe it was a doomed concept in the first place. It’s kind of unclear.
The Startups Team: You’ve talked about, we’ve written about this moment where you had to fire all these people, and how emotional and how painful that was. Was there a point when you realized that was inevitable? Did it start dawning on you, even when you were working on development, but you kept pushing it? What did it take for you to realize, “This is just not going to work?”
Stewart Butterfield: It took several years, first of all because the first years is developing core technologies and stuff like that, and starting to recruit. We didn’t even start testing for a while. There’s a long lag between the time when you start investing in it and when you actually have some useful, constructive feedback.
As an entrepreneur, you always feel like, “OK, I see. That’s the one thing that we have to fix, and once we address this problem, everything will be cool.” Then you address that problem and things aren’t cool, then you’re like, “Oh, OK. No, it was actually this thing,” and then you try this other…
I went to bed one night, and I was pretty sure. It was like, “I think we have to stop,” and then I woke up in the morning, and I was like, “Yeah, we have to stop.” There definitely wasn’t consensus around that either at the board level or with the other managers. To a certain extent, it was like, “If I, as the CEO, don’t believe in it anymore, then, by definition, we won’t succeed, and so just because of that.”
We could see how long we had gone. We could see what our burden was every week. We could see how much runway we had, and how much time we had to figure out some solution to this. We could see what the growth rates were like, and all that kind of stuff. In the end, there was consensus around the decision to shut it down. It didn’t make it any less horrible to actually do.
The good news for us was we had five million bucks left in the bank when made that call, so we could afford to say, “We’re going to give everyone a refund if they want a refund, or they can let us keep the money, or they can donate it to a charity. We’ll give them a list of charities, and we’ll pool all the money together, and we’ll donate it.”
We could give generous severance packages to everyone who’s getting laid off. We spent a month writing reference letters and getting interviews, and every single person who got laid off got a job, so that was the good outcome.
It was super cathartic to announce it, deal with the press, deal with the employees, deal with the community. It’s also really sad. I don’t know how many people would identify as being members of online communities, because that, even still, sound a little bit weird, but if you’ve ever been a part of one that has gone away, it’s really, it’s traumatic.
There’s relationships that you establish with people that don’t exist outside of the medium of that particular community. When it gets shut down ‑‑ and I think this is especially true of massive multi‑player player games, but all other kinds of communities ‑‑ that group, that circle of friends, that whole little universe disappears.
The Startups Team: Was that the hardest thing you’ve had to do professionally?
Stewart Butterfield: Yes, definitely. In a large respect, this was a happy ending, A, because people got jobs, and B, because we’ve hired back several of those people, and some of those people exercised their options at that time. Some people didn’t, but it’s all OK.
But there’s one person in particular I remember. He was a pretty big guy. He’s not a tough or aggressive person, but he’s substantially bigger than me, and I had just convinced him to move, with his family, to a new city and start this job, three months ago. It’s not simple. You’re moving away from your support network, with her parents lived in the same city. They had, I think, a two‑year old or something like that.
First of all, I felt terrible about it, and second of all, I was like, “He looks super angry, and when this talk is over, I hope he doesn’t beat me up.”
Now he works for us, and everything’s cool. It was horrible. I hope no one here ever has to go through that.
Also, it’s f*cking humiliating.
The Startups Team: That’s a big part of it. Here you were this guy who, when you were building Flickr, no one would give you money. Flickr’s like, big idea, maybe a little too early for its time, but woefully undercapitalized in order to achieve the vision and the growth.
It sounds like Glitch was a little bit the opposite of that ‑‑ huge amounts of money, huge amounts of attention and faith in this thing that wound up being a good idea that people loved, but ultimately was a lifestyle business, ultimately shouldn’t have been a billion‑dollar company. Amid that, you were this guy on the ascendancy everyone wanted to fund.
In a sense, is it more embarrassing when you fail after a decent success that people think is your first act than coming out of the gate and failing?
Stewart Butterfield: Oh, yeah, I’m sure it is. There’s way more at stake. We raised more money than we would have if we were first‑time entrepreneurs. It was more ambitious of a project probably than we would have elected to do if we had to scrape together every little thing and make the compromises that you have to make when you don’t have a lot of money.
When you think of every musician who’s had a hit single, and then everything after that has been…
The Startups Team: Did you worry that was you?
Stewart Butterfield: A little bit, yeah. I said yes instinctively. We’re really good at doing that kind of stuff. This was a failed business, but I never thought we were bad at what we did professionally.
I felt especially bad because I’m the CEO. I’m the business one, and I make the call that we do this or we don’t do that and what the business plan should be, do the forecasting, raise the money, and all that kind of stuff.
The software wasn’t bad. Our messaging server was amazing. We don’t need to get into it, but it had all these cool technologies that no one else had developed yet. Given how crappy Flash is we built an incredible client. We had really phenomenal musicians, animators. It definitely felt like it was me.
I took it very personally. It didn’t take that long to bounce back, I don’t think. That was the fall of 2012, end of 2012. We took maybe the holidays off, and then regrouped in January and started development of Slack. That was just over two years ago.
The Startups Team: If you think about Flickr was “big idea,” under‑capitalized, a little early. Glitch, smaller idea, over‑capitalized, Slack seems the Goldilocks of the three.
Stewart Butterfield: Just right, yeah. It seemed like this was a good idea.
One of the things we realized, and I didn’t really mention, when we started Tiny Speck, the company that made Glitch and then later changed its name to Slack Technologies, Inc., and now makes Slack, there was four of us, two in Vancouver, one in New York, and one here in San Francisco.
No one was going to move, so the very first thing we did was install an IRC server. Every person we added to the company joined that IRC server. This phenomenon happened where that was the way we communicated, so that’s what people paid attention to. Everyone had it open next to whatever else it was that they did.
Because of that, we added more information to it. When someone uploaded a file to the file server, it would get announced into IRC. When there was database alerts, those would flow into IRC. When our daily stats are ready, those would flow into IRC.
Because there wasn’t a good iPhone client for IRC, we built our own HTML 5.0 front end. Because there was no way to send a message to someone when they were offline, we built a way of archiving the messages. Once we had archives, we built search.
This wasn’t something that we did really deliberately or consciously, but at the end of the lifecycle of Glitch, before we shut it down, with 45 people working on it, we had, on our company‑wide email list, which was the only email list we ever created, 50 messages over three years, a little bit less than one every three weeks.
That was not ideological. That was not a decision that we ever made.
We were sure that other people would like this, too. We were sure we could build it, because we just built something that was f*cking 50 times harder to build, technologically. We all liked working together. There was eight of us.
Three months after we started, we had a version that was good enough for us to start using. We switched off IRC. Maybe two or three months after that, we started begging our friends to please try it. That was painful.
The Startups Team: You came out in a Fortune article saying, “We weren’t going to raise money unless we got a billion‑dollar valuation.” Why was that so important to you?
Stewart Butterfield: It’s definitely a psychological threshold. People like round numbers, but also it made that article about the unicorns and it became a term. It’s arbitrary as f*ck. There’s no logic to what you get valued at.
It’s the enthusiasm of the individual partners at the VC firm and their ability to convince the other partners. It’s the amount of money they raised. It’s what story they told their LPs before. It’s the trajectory of the business.
The Startups Team: How many people wanted to get in.
Stewart Butterfield: Its dynamic that drives it. We wanted the press to talk about us. We wanted our customers [to have] the comfort of knowing that we raised a serious amount of money. Because we’re well‑capitalized, we’re not go out of business on the one hand, but also we just made ourselves super expensive, and so it’s much less likely that we’ll be acquired. They want a company that’s going to be independent and be around for the long term.
There’s also a certain class of employee who, just because of their personality, they’re more risk averse. They currently work at Google or Facebook or some huge company, they have kids, and one of their driving decisions in how they evaluate whether they would take a job is, “Is this company going to be around for a long time?” They’re kind of people who wouldn’t go work in startups.
Some people who we want to hire are young and cowboys, and other people are older and more mature, and just for whatever reason have that personality type that they want to know that it’s a stable environment.
I should say, one of the ancillary reasons that I think we’re successful is of the four cofounders, at 41, I’m the third youngest. We probably have more engineers over 40 than any other company. Of all the billion‑dollar companies, I suspect that we work the fewest hours in a week. The office is pretty empty by 6:30.
The Startups Team: Really?
Stewart Butterfield: Yeah. People generally work 45‑hour work weeks. It wasn’t ideologically thing. It wasn’t something that we did necessarily consciously, but it is a more mature, respectful environment. There’s not a lot of cowboy people.
The Startups Team: No brogrammers?
Stewart Butterfield: No brogrammers. We have maternity and paternity leave policies, which recognize the realities of being a parent and having to work as well. I think that’s something that makes us more successful.