October 8, 2014
Move Fast, Don’t Break the API
Amber spends most of her time building things at Stripe and figuring out how to scale their product engineering organization.
I think what we're doing is great, and it's working. The investment returns are really good, and people will probably start understanding that pretty soon. The brand that we built works great, on the entrepreneur side at least, for some people.
On the fundraising and the institutional side we scare people, and we look a little crazy. And they're a little cautious about that. It probably makes it harder for us to raise money from traditional sources.
The international work that we've done has made it a lot easier to raise money from international sources. So a lot of money we've raised in the third fund came from China, the Middle East, Southeast Asia, Japan and other places where people want to figure out what's going on in Silicon Valley. Or, money wants to get out of those places for political or economic reasons.
Eventually we'll get the institutional community to buy in. Maybe after I calm down or retire. But maybe not. I'm trying not to sell out.
I think the mission in vision that we're on is working. It's meaningful. And I don't really plan to change what we're fucking doing.
In fact, we're stepping on the gas pretty hard on the international investment and a lot of things that other people just don't agree with. And that's okay. I think we've gotten some people to pay attention. I haven't pissed off everybody in the venture capital community yet. Some, but not everybody. Stanford let me come and do a talk to the GSB, the MBA class. I didn't get in to Stanford, but they let me teach there. I don't know. Maybe that's progress.
I would say a lot of it was reactive strategy, not always intentional top-down strategy. I mean, when we started, we thought about, like, what's the name? We went through a couple different ideas about, you know, Pirate Ventures, because I had done the startup metric stuff. We looked at picking up methodolies from Star Wars, Star Trek, or Dune. I was thinking we were going to use Dune as kind of a lot of the mythology. But some folks hadn't heard of that. It's a little bit male, all those things. We're kind of male.
I wanted to call it Fail Factory, but Christine and Enrique were not comfortable with that. You've got to be really fucking confident if you're going to call your fund Fail Factory.
Although, I still think it would've worked awesome. We tried to do a TV show. I wanted to call the TV show "Fail Factory." Turns out, we called it "The Bazillion Dollar Club." The show got yanked after the first episode, so we should've called it "Fail Factory." That would've been perfect.
I think we were intentional about the name 500 Startups. I think we were kind of thoughtful about some of the branding being simplistic. Can 500 stand for us? Is it, like, short and iconic? You know, the way YC has become YC. It's not Y Combinator; it's YC. It's a single image.
I do think we weren't stupid about how we chose the branding. It morphed a little bit over the years. This is kind of stuck. The Mexico team actually changed the hashtag into this format. Which looks fucking cool. We've done variations on all that are in other languages. The hashtag was almost accidental. That one was pretty much, like, I can't remember who did it. I don't know if it was me or somebody else. We all leaned into it. It just happened, and it felt right.
I think I would give us credit for leaning into the stuff that feels right, you know? But some stuff we fuck up. And it's not perfect.
I think we try to live a brand that we are. I think it's tough to build a brand out of something you're not. You have to identify what you are that you feel different about.
There was a lot of competitive theory analysis around what YC was and what we were. And trying to be intentionally different. I respect the hell out YC. We are not YC. No fucking way, shape or form. Paul Graham's not going to get up here and swear at you. He's a thoughtful person who writes long-form, all black and white. And now I've toned it down, but if you look at my earlier blog posts, they look like a fucking ransom note.
They were all kinds of colors, and crazy. And they were like, "Oh man, you're fucking insane. Just from looking at your writing." Some of that was just me. Some of it was intentional. I do swear. My mom swears. That's probably where I got it from. But sometimes I'm intentionally dialing it up or down. I try to be aware of what I'm doing, but I try to be true to who I am.
They're all subjects that we wanted to learn about. So we had a passion for them. We thought they were interesting areas that we wanted to invest in companies. Sometimes we'd already done that investment. Sometimes we're trying to build the brand to get that deal flow. I think we've done a lot of conferences, so we know how that works.
Now people sort of trust that we're going to put on a good conference. So they'll come to one that we do. We've been doing one on growth marketing called WMD, or Weapons of Mass Distribution. I'm still questioning that choice of brand. Particularly we launched it right in the middle of the Israelis and Palestinians going to war. But they're distinctive, and they're unique.
I think really with COMMERCISM we totally had fun with it. With unSEXY we totally had fun with it. We took pictures of ourselves with broken glasses and tape in the middle. It was fun. We learned stuff, and great speakers came out.
And so for a day we built a little branded community, and in some cases that brand continued over a series of events and years.
I think probably I shouldn't of been breaking up, fragmenting the brand so much. They were still all 500 events. Somebody else is like, "Why do you do all that shit? Why don't you just like call everything 500?" I'm like, "Well, because it's fun." I think once you sort of get the hang of it you can do this repeatably. I kind of feel like in another life I should've been an advertising executive, or scriptwriter. Something like that. Probably would be a little bit formulaic after a while. But it's super fun, and I really enjoy it, particularly when you put on a good event with good speakers and content.
We try and webcast all our events. We try and make everything free. So if you don't attend, you can still get the material. You can watch it remotely if you don't want to come. Or pay. And so I think for those of you who are Grateful Dead fans, they sort of gave away the product for many years and built a huge audience.
The Grateful Dead basically invented freemium, like, 30 fucking years ago. Look it up. It's kind of true. Now it's fun, easy and it works.
I was doing a fair amount of writing between 2005 and 2010. When I started 500, even when I was at Founders Fund, I got busy. I didn't have as much time to write. I flipped into doing a lot of short stuff on Twitter and Facebook. I'm completely addicted to both of those environments.
The hack that I usually do, most people probably know this. I read Techmeme and Nuzzel constantly, browse it. Do you guys know Nuzzel? Do you know Techmeme? How many people know what Techmeme is? Wow. See, Gabe's doing a shitty job marketing. Techmeme's been around for 10 years. Nuzzel's been around for three years.
So, Hacker News is another great source of content, obviously, that tons of people watch. I used to read stuff on Techmeme. I read stuff on Nuzzel, and I just curate it into Facebook and Twitter streams, sometimes without even reading it. A lot of people see stuff through my feeds on Twitter and through Facebook that are articles that I care about.
A lot of me and a lot of 500 is asymmetric familiarity. Here's who I am. I'm going to push it out to the world.
For 50,000 people on Facebook and 300,000 people on Twitter, they get to see what I like, do and experience. Whether that's technology, comedy, art, politics, or my kids, or just silly stupid shit, or me swearing. But I actually try and take a fair amount of time thinking about the pictures and the links.
A lot of times I'll re-curate other peoples' stories who didn't put the right pictures in. A simple trick on Twitter that I use all the time is the three-picture format. It looks like there's a picture on the left and two on the right. And if you put charts in there, people will read that shit all the fucking time. Doesn't matter what it is.
There's some really simple tricks that I think, that I know, work. And because people experience who I am and what I like, I don't actually have to write the long-format shit. I'm totally cheating. I'm copying other peoples' content. But they experience my view of reality, and they feel like they know me.
People walk up to me all the fucking time, "Oh, I saw you did." And I'm like, "I've never met that person before." But they've been stalking me for like the last five years. It builds an interesting sense of connection. Fred, and Brad, and Suster do that amazingly. Even now, with all the people like writing and blogging, I still think it's one of the easiest ways to create connection: slides on SlideShare, pictures on Instagram, Tweets, Facebook feeds, whatever. Your ability to connect with people now is just astonishing.
And so if you have something interesting to say, if you have a take on anything that's kind of unique, people will hear that. In some sense you know brand-building and community-building is easier than ever. I think you just have to be authentic, and you have to connect with other like-minded souls.
That was the point that I was making about being vulnerable. I think telling your stories of weakness and insecurity, in some sense it's bravado. I feel confident enough to tell you my stories about weakness and insecurity, but I think people feel connection on that front. They want to be with someone strong, but they want to also be with someone who has the same insecurities and weaknesses. And so, yes, we're in it together.
"Oh yeah, I got beat up by that motherfucker, too." Or, "Yeah, you're crushing it. I want to hang out with you." I don't know. I think it's just all about sharing who you are. As long as you're not trying to fool someone, people are like, "Okay, maybe I'll relate to that." But I think people worry too much about being perfect or trying to compete on every feature set.
I do take credit for, we thought about what YC represented and what other VCs represented and chose to pick the angle slightly differently.
You don't have to beat the other people at everything. You just have to be different or better in one or two things.
We kind of want to see people hustle to getting a basic product or prototype out and getting early users or customers just to show that they can do it. That they don't need our money to do that. So, people are like, "I'm waiting to raise money before I quit my job, before I build a product." Well, fuck you. You're not taking any risks. Why should I invest in you? Right?
Whereas people who are like passionate, like, "Okay, I already built the fucking product. I already got customers. The economics are working. I just need your money to buy some paid acquisition, or maybe get in the door with somebody else." I'm like, "Okay great."
A lot of our model is really around entering at a point of d-risk, but a relatively low point of valuation. It's actually pretty stupidly simple, like functional prototype, early users or customers, unit economics. That sort of makes sense, or at least could make sense. And then demonstrated growth or potential for growth. And then we invest hoping that growth happens, that unit economics improve, that money starts coming and other investors jump in.
We're wrong 80% of the time, but we're right enough of the time. Probably about 20% of what we invest in gets to some kind of exit.
Probably 5% to 10% gets to a big exit that might be, you know, 5, 10, 20X. We think 5% will get to $100 million plus exit. We think 10% to 20% will get to, you know, north of $5-$10 million exit. We think one to 2% will get to a 50,000X outcome, whether that's a billion-dollar company or not, we don't necessarily care.
And so a lot of our economics for the fund happen from the 2% that become 50X and from the 5% that do 20X. If we're not idiots, we should be able to make 2X or better on that. Depending on the time frame, that kind of works.
The interesting part is probably getting 20% to do 5X and getting liquidity on that and getting exits. I think that's actually kind of interesting. If we could promise that a significant number of entrepreneurs that we invest in could get to $10 million businesses that employ 100 people and are worthwhile, that doesn't make that much sense from a venture-capital perspective. But it makes a hell of a lot of sense from kind of an overall economics, jobs creation, life satisfaction. Some people call that lifestyle business.
I think if you build a $10 million business, employ 100 people, and have a sustainable business, you've done a fantastic job. That's not a lifestyle business. It's not easy running a company with 100 people. Believe me, I'm trying it right now.
I think a lot of what we're doing is not very crazy. It's just betting on a lot of people who might be crazy. It's actually pretty conservative strategy. Our model is absolutely the most riskless, conservative strategy, probably on the planet. Most people don't understand when I say that. It's super conservative. We're betting on there's fucking five or six hundred companies in the current portfolio. The next fund is going to have close to 1,000. We're like the most conservative fucking investor in venture capital on the planet.
The people doing 20 bets and expecting that they're going to get one or two unicorns, those people are fucking crazy gamblers. They're drunk with power, you know? I'm a mathematician. I'm just sitting here looking at the numbers going, "Yep. Two percent's going to get to unicorn. So we better have at least 200 companies. Maybe 100 will be enough, but probably at 200 to 500, statistically speaking, we're going to get three to 10 unicorns out of that.
Then there's huge amounts of flexibility that you can have with the strategy. You can do it all over the planet. You can do it for weird fucking reasons. You can do it for exploration and testing, and you get marketing benefit and network effects out of doing that many investments.
A lot of people just think we're silly and crazy, "Oh, Dave's just throwing shit at dartboards." I'm like, "No, motherfucker, there's actually science and mathematics to this. And it's actually working, and it's really straightforward."
Our three companies that are over a billion dollar valuations are Twilio, Credit Karma and GrabTaxi. Most of those companies we got in at slightly later valuations, between $20 to $60 million. I initially did the investments in Twilio and Credit Karma at three and five million dollars. I did the investment in Zimride, which became Lyft at about three million dollars. Not because I'm smart, just because I'm lucky.
Rather than worrying about unicorns and billion-dollar companies, what I think is more interesting for us is we can probably, between five to seven percent of the time, find people who end up creating $100-million-plus value companies. That's pretty amazing.
Talkdesk was a company we invested in about four years ago. Guy from Portugal came over, won a Twilio platform developer contest. We put some money in. Three, four years later, he's doing $10, $15 million in revenue. Company's valued probably over $200 million. He raised $20 or $30 million. Probably more now.
That's pretty amazing. It just doesn't happen very often. But, you know, it does happen. I don't want to belabor the point of billionaire dollars and other things.
People build cool products that don't cost very much money, solve problems and create happiness or productivity for other people in the world. And that's happening pretty frequently all over the place.
If we're not idiots about building products, and we actually pay attention, you can really change a lot in the world. Possibly. So, that's cool. It's fun. Anyway, I'm done. Thank you.