In The Beginning, There Was A Pen
ReadWrite: Where did the concept of wearing technology on your wrist come from? When did that start to fascinate you?
Eric Migicovsky: I was a super nerdy kid, so you know how some kids would read Mad magazine? Well, I also read Mad magazine, but during high school and late elementary school, I also subscribed to a magazine called Pen Computing—which was as nerdy as it sounds. It featured articles about Palm Pilots, Newton Message Plans, Ericsson devices and stuff like that. I kept a stack of them under my bed.
Pen computing was the height of portable computing. It was the idea that you could have tablets. My first computer was a Newton EMate. They’re really cool—basically like an iPad, [but] 12 years ago. It was a green, translucent laptop that had super long battery life, very cool keyboard, and a pen. And you could write—it had handwriting recognition.
I bought these on eBay six years after they came out, when the price [dropped] and I could afford them with the money I made refereeing soccer on the weekends. I loved the idea. I loved fiddling with it. I would buy these computers, and that’s when I realized I like the idea of having technology get closer to the human body.
I read a lot of science fiction about human computers, brain-computer interfaces, and for me, pen computers was the height of that coolness.
RW: And that lead you to study engineering?
EM: Yeah. I went to the University of Waterloo. I studied in a five-year engineering program called Systems Design Engineering. It opened up an opportunity to try a whole bunch of different stuff—a bit of electrical engineering, computer engineering, software, a little bit of chemical engineering. The coolest thing about it was that every term, you worked together with three or four other people, and that would be an entire course. Like, instead of going to class, you’d get together with your group, and you would build something.
We built a miniature assembly line. We built a bridge. We built a device that sits in the bathroom of an elderly person with Alzheimers, and walks them through their activity of daily living. It had a little sensor under the pad in front of the sink that would remind them to brush their teeth, take their pills or wash their hands after they used the toilet. We built that into an Ikea bathroom set-up. It was a lot of fun.
It was a multidisciplinary team. I did a little of the electronics work, another person would do a bit of the software. And that’s why it was so natural, in my fourth year of engineering, to start working on InPulse, which was our first [smart]watch, with a group of my friends.
Out Of The Dormcubator, Into A Startup
RW: When did this go from school project to the realization that, “Hey we’ve got the beginnings of a business here.”
EM: So that was the fall of 2008. I was living in Canada’s first residence/incubator, which was in a dormitory. They called it a “dormcubator”—pretty terrible name. I was in the first class in 2008, which was during a downturn. You know, what were kids going to do once they graduated from university? Well, could they start a company.
This is Waterloo, Ontario, not Silicon Valley, but it still had a bit of tech going on. Blackberry was right up the road, and it was cool. I was working on this watch thing, showing a bunch of people, talking about it a lot, doing pitch competitions. I was still in my fourth year of engineering, and I didn’t want to drop out, so I decided to graduate. But more and more people were saying, “That’s a cool little invention. Have you thought about making it into a product?”
RW: If you weren’t thinking it before, you probably did then.
EM: I did a lot of co-ops. At Waterloo, you do a lot of co-op programs. You go into the workforce and have an internship for four months. And I did five of those in university. A whole bunch of random companies. I worked at Raytheon, I worked at a bank, I worked at an IST, but none of them really struck a chord with me. They were fun and interesting, and I learned a lot, but I didn’t want to make one my career.
So when I was going to graduate, I thought I could just continue doing InPulse. But I didn’t really think about it that much.
RW: Blackberry was so close to you. Is that why InPulse was compatible with that platform?
EM: Yeah, that is definitely why. [And] you know, 2008 was the first year of Android, and at that point the iPhone couldn’t support a watch.
RW: So when you left Canada and came to Silicon Valley, that was when you left Blackberry behind, right?
EM: Very close. We got accepted into Y Combinator and went to the Valley in the beginning of 2011. We were still selling InPulse at the time, but in Waterloo, everyone had Blackberries. When was the last time you saw 20 Blackberries in a room? That was, like, everyday in Waterloo.
Waiting For The Apple To Fall
When we got to the Valley, we discovered these new phones that everyone was using, called iPhones. Kinda cool? And the number one question from the venture capitalists—that everyone asked—was, “Do you support iPhone?” We couldn’t. At the time, there was no API [application programming interface], no SDK [software developer kit] that would allow you to build a watch that would work with the iPhone. That was pretty rough.
We literally said we can’t make it work. The big problem was that iPhones, before iOS 5, talked through Bluetooth to accessories like a watch. But the moment you put that app in the background, or you closed or turned off your phone, the app would not be able to communicate with the watch.
Then in 2011, when WWDC came around, Apple released an API that enabled third-party Bluetooth applications to talk to an app in the background.
EM: iOS 5 opened it up, and we were like, “Sweet, this is awesome!” Within a month or two, we built a prototype that would work with iOS.
We had already sold about 1,200 InPulses by then, and there were already over 100 apps for it. We learned a hell of a lot. The InPulse had a color screen. It wasn’t waterproof. It had some very rough edges around it. But it had an SDK.
People loved writing apps around the watch. So we went back to the drawing board in the fall of 2011. We said, we’re making a watch that works with iOS and Android. We said, it’s going to be waterproof. It’s going to have an e-paper screen. And it’s going to be a drastic improvement from our first version.
So we put together this design, started working with our industrial designer, and at the beginning of 2012, we had a prototype. We had a plan, we had a name—Pebble—and we tried shopping it around to VCs.
We had quite a few connections through Y Combinator, to some of the hot VCs in the Valley. Everyone still loved the idea, but we could not get any traction. We took meetings and we could not lock them down.
Scraping The Barrel
RW: Initially, you got funding from Canadian government programs and, later Pebble took off on Kickstarter. Did the different sources of funding affect the way you built the devices?
EM: Yeah, totally. The very first InPulse prototype was funded from winning pitch competitions. We were constrained from the get-go. I didn’t even know about raising money [then]. We were just a couple of kids in my garage in Waterloo. So we worked within our means. I borrowed $15,000 from my parents. I borrowed a personal loan from the government. We got some funding from the Canadian government.
We didn’t have much cash. We worked with a small prototyping shop in Waterloo, and we started making really cheap circuit boards in China. When we came down to the Valley in 2011, we ended up raising around $300,000 right after Y Combinator. That was our first big dose of money. Until then it was $15,000 here and there.
We ended up making some mistakes. Up until that point, we had always been building within our means, but now that we had a little bit of money, we made some mistakes, and we built too many products of our first generation—more than we could sell.
RW: Supply chains can be tricky. Earlier this year, it seemed a number of Kickstarter backers were getting restless about receiving their Pebbles. What’s up with that?
EM: We shipped all of our Kickstarter orders. [In fact,] we shipped many more. We had 85,000 orders on Kickstarter, and we shipped way more than that.
RW: That was through your relationship with Best Buy, right? And AT&T start carrying Pebbles recently.
EM: Yes, that is correct. We also sell on our website around the world.
RW: Ultimately how did you figure out the supply chain issue, and the pricing of components? That must have been a big learning curve.
EM: It just takes time, and a lot of effort. During the Kickstarter campaign, I went to China about seven times…. Literally, there was one person from the Pebble team in China, basically 24/7 for eight months. Like literally driving around China and trying to find the right combination of stuff.
RW: How do you decide what goes in there and what doesn’t? For example, Pebble doesn’t have certain fitness-oriented functions like a pedometer.
EM: Pebble does have an accelerometer, which is the core sensor inside a pedometer. So it’s really a software problem at this point. This fall, one of the things we’ll be talking about is the ability for developers to use the sensors inside Pebble. We’ll be announcing that very shortly.
Making A Smartwatch Platform
RW: Seems like you have quite a commitment to openness and the developer community.
EM: It’s an open platform, which means anyone and everyone can hack on top of Pebble. We’re not limiting it to just the people that we want, or [forcing developers to] authorize apps. If you want to build something cool, Pebble is just the best wearable platform [for it].
Back in 2011, when we were doing a redesign of our architecture, we made sure that Pebble was a platform that other people could build apps on top of. When we announced it on Kickstarter, we posted that. It was part of the pitch—customizability.
So when we launched our initial alpha SDK earlier this year [the PebbleKit SDK, in April], people started creating different watch faces for Pebble, we knew that we made the right decision in opening it up. There’s a great site called MyPebbleFaces.com, and another called WatchFace-Generator.de that lets people who don’t know how to program create their own watch face.
RW: I heard that Pebble isn’t planning on building its own watch app store. Is that true?
EM: That’s not true. We’ve always talked about how we can help people find apps for Pebble—that’s part of our core competency. We haven’t done that yet, that’s definitely an area where we’ve been lacking. But our goal is to make it as easy as possible for users to find new apps for their watches.
[But] we’re still a fairly small team. When we launched Pebble, we were three people. At the beginning of 2013, we were 11 people. Now we’re over 30. We choose our battles, and we have an immense number of things that we could be working on as a company, but we choose to work on the most important things for our users at any given time.
Right now, what we’re working on is improving the developer experience on Pebble. And part of that is incentivizing developers to write apps for Pebble.
Future-Proofing The Pebble?
Source and read More on