Hosain Rahman at Startup School SV 2014

Thank you for coming. I hope to do not much talking at all, and I'm going to just ask you because it's sort of a strange thing you have this. I want you to sort of go through the chronology of the early days, yeah. Um, and talk about you really did have a lot of close calls, and I'd love for you to talk about them and how you got through them and sort of what did we call them the other day, sort of the not near-death experiences but defibrillator... defibrator, actually, had to sort of resuscitate it.

Yeah, and I've literally never heard of so many, so I'd love for you to give us a little story. Cuz it was back in 1999. I mean, that was a lot when we started the company; it was 1999. I mean, a lot of people don't know this, but when we started, we sort of started working on the ideas in sort of '97-'98. And for lack of a better way to describe it, we were trying to build Siri for mobile devices.

We've been inspired by Palm and Nokia and all these folks that were sort of at the front of this mobile device revolution, and we thought, why wouldn't it be great if you could talk to your phone? Um, and that would sort of transform the user interface experience and all of that. So we started with this kind of customer problem and that's always when we've been good. And when we stayed away, when we stray away from trying to solve that customer problem, we sort of fall down.

Um, and there's been a few of those along the way, but, um, you know, and so very early on, we were trying. We said, okay, voice would be the great way to interact with this. How do we make basically Siri this voice interaction layer on top of the operating system? And we set out to do that.

We realized everything we wanted to do with speech recognition wasn't going to work. Um, and at the same time, we sort of started to try to go raise money for this idea of a new way to interact with um, your device. And you know, I say now it was harder for us to raise our first half a million dollars than it was for us to raise our first $200 million.

Um, and it's because in those days, like, people weren't in the valley; they weren't so focused on mobile; they certainly weren't thinking about mobile voice. Um, and we really needed to prove that we had—everyone thought, yes, there absolutely has to be a better way to interact with your device—but what’s your hook? What’s your advantage? What's that kind of breakthrough?

And what we ended up doing is inventing this noise cancellation technology that turned out to be the biggest breakthrough in mobile audio in 30 years. It was a sort of serendipitous by-accident thing that we stumbled upon, um, mostly because we were trying to solve this problem.

And then when we did that, we said, wait, this is bigger than making speech recognition work; it gets to the core of what people do on their phones, which is talk at the time. Um, now we do lots of other things on our phones, but in those days it was a lot about talking—mostly about talking.

And so then we thought, okay great, you know, everyone’s going to want this. Everyone that ever heard the demo where we would do these A/B tests with weed whackers and blenders and all this stuff, and we'd sort of take out all noise, everyone’s like, wow, I want to have that in my device.

Um, and so we sat on this journey. We started talking to all the handset manufacturers that were big at the time, sort of Nokia and Motorola and the Korean guys, Sony Ericsson—or Ericsson, actually—and it was a real wake-up call: no one actually wanted to take this technology. They thought, yeah, it's a neat demo; we think it's cool.

So they loved the demo, but they loved the demo, but they didn't want to integrate the technology. It was too expensive to add additional microphones. I was with, um, somebody from Apple the other day who was telling me they now have six microphones, I think, in an iPhone.

Um, and at the time, we were just trying to convince people to put an extra one in to support our technology, and I was just like, wow, I can't imagine what we would have been able to do in terms of audio quality with that.

But, you know, again, we struggled to get our technology into that form factor, and handset manufacturers were just sort of concerned about pennies. So they didn't want to integrate what we had. And so at that time, we were struggling with, you know, how do we fund this? How do we take it to the next level? We know we had something pretty breakthrough, and what did we do?

And so Along Came DARPA. Um, and that was a great thing for us. The defense department, basically a government agency, would never fund you. We never thought we would do sort of work for the military-industrial complex. What they wanted you to do—they wanted us to make our noise cancellation algorithms the foundation of, um, their battlefield technology.

And so we thought, well, great, we have this kind of raw science; how do we take that and productize it and use these government grants to really sort of work through all the issues and take that science into a form that could be delivered in a product?

And it was great, you know. It sort of kept the lights on. We were eight engineers; I think we were, um, after Dolby and Macromedia, one of the earliest technology companies—real technology companies—in San Francisco. We moved up to San Francisco in 1999. In those days, there just weren't a lot of startups in the city, and it's funny now cuz I love it: you walk around, you know, we're in the same block as Zynga and Airbnb and Pinterest, and we used to share an office with Airbnb, and it's just so fun that there's this vibrant community there now in San Francisco.

It wasn't like that. Yeah, I think I need to remind people, we're talking 15 years ago. God, it's like the stone ages almost before iPhones. So, I want everyone to go back in their minds and—and well, half these people were probably in grade school—totally, yeah, let’s hope they were all born.

Um, so then what happened? You get this money. We get the DARPA grant. And the other thing that was really tough is it’s so, again, also fascinating because so many of the ideas that were being kicked around in 1999. And I remember Larry and Sergey were like our TAs in computer science, and Marissa was my—you know, lived in the same freshman dorm, and she's now on the board. And so it was this crazy time at Stanford and Silicon Valley and all these interesting ideas and people.

But you had this huge run-up with everything—all the infrastructure being laid out for the internet as it stands today. Um, and all these ideas—like there were a lot of delivery services like Instacart, right? But none of them survived. Um, and you know, there were online-specific e-commerce things for niches; none of these things survived, and they were just sort of too early.

There needed to be a lot—either, you know, better internet connectivity, more distribution of it, whatever it was, it was just too early. And so the ideas weren't bad; they were just sort of ahead of their time. Um, and I think that’s another important lesson that we learned: you've got to stay your course through that.

We thought mobile was going to be a big thing. We thought that Palm devices would converge with smartphones; they ended up doing it sort of 10 years after we thought they would. Um, but they ended up doing that. Um, and so the sort of early lesson we learned is if you believe in what you think is right and what you think is going to happen to customer experiences and to the way people are going to interact with technology, then stay that course and put yourself in a position to do that.

And that was the sort of first thing: DARPA. We went to a really unconventional source of capital, right? Where who had DARPA funding at that time? Very, very, very few people. I just found out recently that Inktomi did too, but you know, it's not something that—who remembers Inktomi? No one. Um, they powered a lot of the infrastructure of the internet that Yahoo bought, right?

So anyways, um, what happened after the DARPA funding is again the whole technology industry had sort of cratered. Um, and it was impossible to raise money, and I remember we used to go up and down Sand Hill Road looking for money, and people would say to us, “What’s wrong with everyone who went to Stanford from 1993 to 2000? You guys have lost us billions of dollars; go get jobs! We don’t want to see you here!”

I mean, wow, it’s funny because I still run into some of these people. Um, but they like you now; it’s different. They don’t tell me to go get a job in the same way that they used to. But, you know, and so again, it’s interesting how the sort of tide had changed. It was not a happy, fun place full of new ideas, and that vibrancy wasn’t there.

There was a really bad feeling around, and so we had to figure out a way to keep going through that—to believe that, hey, mobile’s going to be a big thing. And at the time, the valley wasn't the center of the mobile universe; it is now for sure, right?

And so, um, how did you get through it though? You got some—so we had the DARPA funding, but that couldn’t have lasted forever. It didn’t last that long, and it couldn’t—it didn’t really allow us to scale in a huge way, but it did allow us to kind of productize, productize, and prove out what we had done.

And the next thing we did is we said, okay, how do we get this into a form or product that’s going to reach lots of people? And we've been in sort of licensing discussions because we wanted to be like Dolby. We said, you know, anyone that has an audio problem, we could be in the front end. It’ll be kind of like an Intel inside or Dolby, etc.

And because of various economic constraints for these very—for these guys and focusing on profits, no one wanted to pay us much to license our technology. And at the time, we’d been talking to the big headset manufacturers, Punch Rocks and Jabra, and we thought, wow, the technology we have in this space is really disruptive. There hasn't been a computing proposition there.

Um, and we can really transform what's possible there. Um, and so we’d been offered some deals from those guys; we backed out and said, you know, we’re going to go the course ourselves, and we're going to go make our own hardware device, right?

Wow! So this is in kind of late '02. Um, and we thought we can do this better, and that's why I have hardware backgrounds at all. Well, you know, I’d studied mechanical engineering; this was before there was a D school, so we were kind of in the D school, so we knew a little bit about—and some electrical as well, so we knew a little bit about how to make things, but certainly to be dangerous enough to think that we could do it, okay?

Um, but knowing what I know now, it’s totally insane that we did it, right? Um, and I can't believe that—you know, anyway, that's so—I can't believe that people thought we might actually be able to do it, right?

Um, and, uh, and so, you know, again, long story short, we kind of put together some parts and were able to raise some capital to go build a hardware device. Now again, in those days, building consumer hardware as a startup was totally insane.

Um, I think you know we did a study at that time. The only other sort of venture-funded consumer hardware company—well, there’s two—had been Apple, right? Whenever that was in the late '70s, um, and then Palm. Oh my God! And Palm didn’t really stay around very long; it became part of US Robotics even before they launched the company.

Um, and so there had been no examples for people to look at and say, okay, this is something that we took through the cycle, built the product, scaled it, built distribution, got it out, and then there was an exit. So there was no sort of pattern that people could point to.

And so talking to anybody—I mean, there were big failures like Go, and some of the Bill Campbell was involved with that, and John Doerr and all these guys. Um, and so just convincing people that we knew what it was going to take to do that was really, really hard.

And I remember when we raised, you know, venture capital in 2003, um, it was like one of the only deals that that firm at the time had done. Um, and it was insane to go build this stuff. Um, and then we set about going to build our first kind of consumer headset, and that's when we started to really think about design.

Um, and before that, you know, in the valley, design was a sort of thing that once you had a little momentum, you might put like a cool case on something or a different color or whatever, and if you wanted to appeal to women, you made it pink. And, um, this is sort of what people thought of design.

The only other company that was really sort of saying this was this intersection of technology and design was Apple, um, in sort of '01-'02, really with the iPod, um, at a mass scale. Um, and so for us, we were fortunate to have such a great, you know, kind of tip of the arrow in Apple sort of going and cutting down the path in front of us.

Um, and again, so long story short, we put these elements together, and we realized how hard it was to take these complex algorithms and to figure out all of the little details that you needed to think about in order to make this product great.

Um, and you know, sometimes I sort of sit up and wonder. We make lots of products and now have been fortunate to ship tens of millions of things, but I look at, you know, Tesla that makes cars, and I think about all the details that we have to think about to get it right.

And I can't imagine how Elon sort of processes all of that all the time from, you know, there’s just a million different things that can go wrong that you have to think about when you build a car, right? And it's not dissimilar when you build these very complex consumer products where you've got software interacting with hardware and now talking to the cloud and how do these things work?

And there’s very little patience people have for things not working. Um, anyways, so kind of flash forward, we ended up launching our first consumer product, U, which was a headset. Um, and it was a—and there was a sort of a precursor where I knew that we might be going wrong.

Um, on this first product, I had the first time I was fortunate enough to meet Steve Jobs was in mid-2004 right before, um, Walt and Kara’s D conference because we were going on stage to debut our product. And it had been set up to meet him—he’s sort of the god of consumer electronics, right?

And I went and saw him, and it was sort of 45 minutes of absolute just—I don't even know how to describe it. We just got killed—absolutely killed on every single decision, every single edge, every single technology decision.

The crazy part is I knew everything that he said was right, and we had known that deep in our hearts, and we didn't because we were trying to meet a timetable and because we were trying to, you know, not run out of money and ship the thing on time. We sort of made all these trade-offs, and we compromised a lot of things that we in our hearts knew better on.

And for me, it was a sort of really galvanizing moment because I remembered that, hey, you know what? This guy who’s really good at this is telling me that that was wrong and that we shouldn’t have compromised on what we know was true, right?

And so that was the first moment I was like, we should have gone with our gut. Sort of harsh but valuable feedback—totally okay. Um, it was unemotional but like brutal. I mean, just really, really, really, really brutal, you know? Just cutting—like, oh, the only place I remember we were talking about, like, this thing was crazy.

I don't know if anyone's ever—certainly probably no one in this audience remembers it, but it was a big headset connected to a clip on your belt, and then it plugged into your phone. And we were arguing about the clip, and the product guy at the time that we had hired was like, “Oh no, but people clip this to your belt.”

And Steve was like, “The only place anyone would ever clip that is in your mind. Get it out of your mind!” You know? And he was right. He was right. We probably sold more units sitting here right now than we did of that first product because it shipped, it got all this like great buzz, we were product of the year in Time magazine and it was—we sold like five of them because it had had all these problems.

But we knew that there was this cool technology, and that was this like really interesting intersection of, wait, we didn't stay true to the customer experience and the problem that we were trying to solve for the user. We compromised all these things to meet all these schedule criteria—funny, and they’re real constraints, right?

Um, we didn’t follow our instinct—didn't follow the gut. That gut instinct we knew better, and that was probably the hardest thing about my conversation with Steve. I did send him an email after that saying, “You were right on everything. I appreciate that you actually took the time to be honest with me because few people are.”

Um, and respond—yeah, and it was the—it was sort of he was always like that with us. And um, then we launched several other products with Apple like the Jambox and all those things. We had a great relationship with that company and continue to, but we sort of learned a lot from my experience.

And so going a little bit further on, on what happened, we ended up launching this thing; it was totally stillborn—didn't sell. We ran out of money, and now it was like, oh my God, what do we do? Um, and again, very long story short, people at the time that were supporting us, our investors, had sort of lost faith, and they shut the company down.

People don’t realize this: like Jawbone was shut down; all the employees were laid off. They literally chained the doors, and they chained the doors—chained locks; the whole thing. I have a letter that says everyone, “We don’t see any value in this thing; we’ve terminated all the employees.” It was gone, right? Um, because in those days too, you didn't—as entrepreneurs and founders, it’s a lot more friendly now than it used to be from investors.

Like, you actually get to control your company. Actually, we were kicked off the board. We weren’t sort of kicked off the board; we didn’t—you know, have a say in these kinds of corporate—was like, thanks very much for your ideas; like, we’ll take it from here. So what—that’s changed a lot.

So what happened? Show up in their chains on—chains on the doors! But the thing that I had—and my co-founder really had—is we still believed that we had created something special. We still believed the mission, and we still believed that we could improve people’s lives and that we could allow them to communicate better and that we could sort of push this stuff forward.

And so what we ended up doing is we kind of wrestled it back from those investors, and I think I took it back. We had $60,000 in the bank and $600,000 in debt. Um, we took another DARPA contract, and we focused in and said, what are all those mistakes that we made that we knew better on?

Um, and often failure is the best teacher that you can have, right? Because you learn so much more because you think about every single angle in 540 degrees; you’re looking at it again and again and again, and you’re like, what went wrong, and why did that happen? And are you independent at this point, or are your investors still involved?

Well, they just sort of go—they said just go take it. Everyone would sort of go off the board, lost… just whatever it was. It wasn't even they were—it wasn’t even a considered decision; it was kind of—you're sort of starting with a blank slate. Blank slate, right?

And we had this cool technology that had all this promise, and people resonated with it when they got it. And so we figured out that it had to get into the right package or the right form or the right way that people could access it and interact with it, right?

And so we had failed on what I think about sort of is all what design is all about, which is the attention to the right details and resolving those details to make the best possible experience for the customer, right? And we’d missed that, right? And we missed it in a big way, and it nearly put us out.

And so, again, you know, we focused for the next two years. We worked for no salaries, kept DARPA—sort of kept the lights on, as did, you know, our friends and families. And then we focused on our first headset, and that was when we started to think about wearable computing, actually.

And it's funny because now everyone talks about wearables every day. Probably every hour I get a question about wearables and where’s it going and what’s happening. And that’s when we started to think about these things on your body that have all this computing power. How are they going to tie to services and all that stuff?

So anyways, we plowed through that. That was two years of probably another, you know, two days of crazy stories. Can you talk at all about the shipping story?

Oh, you love this. Um, we went through two years, no salaries. We had, you know, Venod and David Widen from Coast Ventures had helped us get a deal with AT&T to launch our, you know, kind of next generation Bluetooth headset.

And everything’s lined up. We have this big customer; we still couldn't raise that much capital to get it off the ground. Um, it was tough, but Venod and David sort of saw some problems, so they’re kind of just advising us to see what would happen. And we were trying to launch; it was late in December.

Um, we had product in a dock in New York, and the manufacturer wouldn’t release it to us because we didn’t have any money in the bank. We literally had like $2,700 in the bank, and we had to figure out how to get those products released out of U.S. Customs so they could be delivered. AT&T was singular at the time to get into stores.

Which for like Christmas—because for anyone that's launching a product, like December 21st is not a good time to launch a product for Christmas; it’s kind of too late. Um, and so that was a whole another story.

So we convinced AT&T to still take it in, and then one of our earliest angels, these guys Chris Burr and Austin Hurst, they gave us a letter of credit to unlock it. I had to like go to JFK to go find the customs people and figure out what the harmonized code was for this type of product and get them to release the stuff so it could get in stores; and Moss wrote about it in The Journal, and it sold out in a few hours.

Wow! I think we went from like $0 to $70 million in revenue in our first year. Wow! Um, so that’s the thing that I learned that I hope people take away is when you actually go focus on ruthlessly what those things are that you're trying to solve, um, and those problems, you can transform your business like that.

The other stuff—sort of you got to figure it out; it's not easy, but that's the core thing, right? Um, and if you don't do it, everything gets really bad, and if you do, then you can build from there, and things transform overnight like that. So literally, it was like this. It was like overnight, overnight—overnight.

Then what happened? Then we had to go build a business. We had to scale; we had to build a team. We weren't fortunate in that situation where a lot—you know, we would raise a bunch of money, built out our team, made the product; it was a success, and then we could scale from there. We had no infrastructure.

Right, so we went $0 to $70 million with a consumer hardware product with—at that moment, in the beginning of December of—oh, sorry, beginning of January '07, we had six people. So we scaled that Z level revenue grew to like $22 million, and that’s when, you know, Vode and Mark and Ben and, um, Ben Horowitz and Mark and Dreon and Sky guys all kind of came in, and we started to build the company.

Um, and then, you know, we had this sort of screaming 2008; business was growing, growing, growing, and then all of a sudden, you know, you had the big meltdown of 2008 where Lehman Brothers exploded. And we found ourselves with a million units on back order, and all of a sudden those orders just got canceled, and so we had like $70 million—or $50 million—or $7 million worth of inventory in China we had to go then learn to work that down.

Um, and spent a lot of 2009 working that down to zero, did a big deal with Costco to move it. So it was just, you know, all of these bumps and bruises, and again, we experienced it in the fall of 2011. You know, I had announced this product on stage that—foray—using all the stuff we learned about sensors and headsets and now translating it to health, and it was this massive problem.

I literally debuted the thing on stage at TED, um, and we had like a billion media impressions around it, and it was—was the most hyped product. I still think it was the fastest selling third-party product in the history of Apple retail that fall, and it was going out there, and all of a sudden we were starting to hear issues of them breaking, um, at scale, and we didn’t—we couldn’t even get units back to figure out what was wrong fast enough.

Um, and here was this sort of incredibly hyped product. We had tried—we were thinking about levitating in that category even before we had learned how to crawl, let alone sort of walk. And, you know, all this hype—and you don't even know what to do with respect to the consumer.

I think we're fortunate because we've been through some hard times, so there is a DNA in the organization where we could sort of take a step back and say, you know what? Like, let’s just try to figure out what’s happening. First principle is let’s make sure we take care of the customer, make sure users are happy. That, you know, the first question I ask is are we doing any physical harm to people? Is there anything wrong? Is it like hurting people?

If it's not, okay, what's—Catching Fire? Is it Catching Fire? Is it burning people’s wrists? That happened to one of our competitors. Not good. So, you know, um, you know, I think that you got to sort of go from there, and we’d been fortunate that we’ve been through hard times, and it was never this sort of straight shot of success.

And I often tell people, like, look, the problems that you experience when you’re five people with, you know, tens of thousands of dollars in the bank versus when you're, you know, worth billions in market value and got hundreds of millions of dollars in big channels and customers—they are no less existential or nerve-wracking; they're just at different scales.

And so if you learn to sort of persevere through those things, and you focus on the things that matter, you'll carry those skills with you through the whole journey, right? And don’t forget those lessons of what it was like to be on the ground with, you know, your teeth kicked in and the taste of that blood and how you go through it.

And then what we ended up doing with UP is we figured out that we had an issue, and I sat down and wrote a letter and said to people that I’m sorry; we made a mistake, and we screwed this up, and give you your money back, and you keep the product. Um, and, and you know, we’ll take you through it, and we're going to go fix this stuff, and we’ll be back.

And when we came back, consumers embraced us, and you know, people—I was surprised because the reaction of that letter went from I was getting death threats on Twitter to, um, literally my wife will remember this—to, you know, all of a sudden people saying, wow, like that's the example of how you treat customers.

And all we were—we weren’t trying to think about that; we were just like what’s the right thing to do, right? So I feel like I tell people never to start a competitor to Jobon or any of your products; you're like the toughest founder who’s been through so much they’d never survive.

So we have a couple more minutes, and I know there’s so much I wanted to get through. Hopefully, that was the good narrative; that was good. We talked about some of the near-death ones that I loved.

Um, can you talk about how hard it is to take software and hardware and data and bring them all together? You're so good at that.

It's cool; I mean what we're doing now is—I feel like everything over the last, you know, 14 years has been preparation for it. I feel like we've been going to school where we learned how to build these great high-design products, um, and we've been fortunate enough to be sort of recognized for it.

And one of the things that was interesting about the first UP launch is we had built our own application from scratch. We'd never done that before. Um, you know, in 2011, beginning of 2011, we didn't have an application team; we had to build that from scratch.

And so one of the things that the hardware defects matched was how crap our app was. Um, it was really bad; a lot of my friends like Zach and all these other guys said, “What are you doing making software?”

I said, "We've always made software, just at the algorithm level, firmware and all this stuff.” But you know, it was a really tough learning experience to get people who come from the application world to work with people who work in hardware and not to have them work in silos.

And then some magical point for it all to come together—you realize that doesn’t actually work. And so I used a lot of 2012 to kind of rip it all down and say, look, you know, what we learned in hardware is you have to make stuff that people will pay you more than what it costs you to make, which is a really interesting discipline because you end up being really focused on is this good enough that someone would take a dollar out and pay you for it.

Right? In app development and in lots of web development—mobile apps—people don't have that discipline because you don’t—everything’s free, and it just sort of moves quickly, and so people iterate, and you just sort of figure out stuff; but there isn’t that discipline of like is it good? Are we solving that problem? Will people pay us for it?

Right? And so we took that sort of a sensibility from the hardware team, and we kind of tried to get our software guys to think about their experiences in that way around trying to resolve it to a point where, yeah, we’re making decisions, and we’re making judgments on where this is going to go and how it comes together, but then we got the speed at which the software guys could iterate to sort of infuse our hardware team.

And so they kind of tried to get what the best of each of these things were, and then ultimately what I did is I organized it against the user problem, and I said we are solving this customer problem, and we are going to use these things as tools, whether it be the software app or we’re going to solve that in hardware, or we’re going to solve it in the cloud with data.

That was another new thing for us: to build a whole data science team. Um, I think we have a world-class one. We started to publish a lot of those results; I measure it as world-class. You have Google and Facebook and all these guys trying to recruit out of our team, which I think is a sign that you’ve arrived when the big guys are trying to steal your people.

Um, but, um, you know, putting all those elements together and getting them all to talk to each other because they speak different languages is really difficult, and I think that’s the new tip of the arrow, right?

And I think this comes back to that advantage that we have against people who've traditionally been successful in CE. CE is totally different than what it was. They're no longer these dumb boxes; they're boxes where the application experience in software is just as important as what you physically touch, and there's a melding of those lines.

I tell people when we think of design in mobile applications, it’s something between three dimensions and two dimensions. It’s not like just a little visual thing; you’ve got to—you’ve got to interact with, and that’s not quite three dimensions where you can feel the button, but it’s something in between.

And so we’re sort of putting all these pieces together, but again, it's all about here's the problem we're trying to solve, and how do these pieces—how do you move around to actually go solve that?

So, well sadly, we are over. Um, thank you; congratulations on all of your hard-won success. I'm obsessed with my Jambox mini, and thank you for coming today. It was awesome.