The Rise of Evernote: An Interview with CEO Phil Libin (Part 1)
The paperless office is a pipe dream. But these days, the bulk of the information that knowledge workers encounter every day is born digital, not on paper. As long as there’s an easy way to store and retrieve the important snippets, there’s a diminishing need to print any of it.
That’s where companies like Evernote come in. The Mountain View, CA, startup runs a digital notekeeping service with more than 3.2 million users, a number that’s growing by 5,000 to 10,000 per day, CEO Phil Libin says. The customers are gravitating to a service that lets them upload as much as 40 megabytes of data per month to their Evernote notebooks for free. All that data can be accessed a number of ways, either through a simple Web interface, the dedicated Evernote programs for Mac or Windows computers, or customized Evernote apps for Android, Apple, BlackBerry, Palm, or Windows Mobile devices.
Once it’s on Evernote’s servers, users’ material—whether it’s Web clippings, e-mails, receipts, recipes, voice memos, photos, maps, or shopping lists—can be searched and browsed from any Web browser or Evernote-capable device, on demand. Perhaps Evernote’s coolest feature is its ability to index even handwritten, scanned, or photographed text. This makes it easy to snap a picture of your office whiteboard with your camera phone, upload it, and suddenly make every kooky idea you and your colleagues have searchable on Evernote.
Longtime Xconomy readers know that I’m a major fan of Evernote, having relied on it as my main digital repository ever since the company’s combined desktop, Web, and mobile service debuted in 2008. In fact, so much of my life is stored on Evernote now that when I set out for San Francisco later this month, I’ll be putting far fewer file boxes on the moving van than the last time I moved a few years ago.
Earlier this month I caught up with Libin by phone. He’s a former Bostonian who studied computer science at Boston University. He also spent long parts of his career at pioneering e-commerce companies Art Technology Group before founding Engine 5 (acquired by Vignette) and then CoreStreet. Libin was recruited to lead Evernote in 2007, five years after its founding by mathematician and software engineer Stepan Pachikov. Under Libin, the company hit the fundraising trail (it’s collected some $25.5 million in total angel and venture funding, from an eclectic group of investors that includes Japan’s NTT DoCoMo, Russian investment bank Troika Dialog, and Menlo Park, CA-based Morgenthaler Ventures). It also went back to the drawing board on its technology, spending about a year redesigning its original Windows storage software to work across multiple platforms, including the Web and mobile devices.
That decision was vindicated, Libin says, by the huge upsurge in users that followed the release of the Evernote iPhone app in July 2008. (It seems appropriate that the iPhone, and now the iPad, are now such important vehicles for Evernote, given that Pachikov is known as one of the fathers of automatic handwriting recognition, and was involved in the creation of the Apple Newton back in the early 1990s.) My talk with Libin, who is one of the newest members of our Xconomist network, ranged from his early interest in “memory prosthesis” technology to the mobile revolution to Evernote’s business model, which, interestingly enough, is not dependent on converting lots of free users to Evernote’s $45-per-year premium plan. Today, in the first installment of a two-part conversation, Libin talks about how he got involved with Evernote and crafted its strategy in 2007 and 2008. Tomorrow, we’ll continue with details about Evernote’s revenue model and Libin’s vision for the future of the company. Here’s the first part:
Xconomy: You became CEO of Evernote in 2007 after about six years at CoreStreet, a security company. How did you first connect with Evernote?
Phil Libin: The business [at CoreStreet] had morphed into sharing with the only people who cared about security, which was basically government, people who buy crypto stuff. It was a nice, profitable business, but I started getting pretty bored. I became concerned one day when I realized how much of my brain was taken up with information about DoD budget cycles and the German defense establishment. The company was at profitability, and just needed adult supervision, so we hired a new CEO, and I started thinking about my next thing.
I decided to do something radically different, staying with computers but getting as far as I could from government cryptography stuff. I wanted to do something in consumer Internet. And if it was successful, I wanted a billion people to wake up in the morning excited about it.
I thought that some kind of memory prosthesis or augmentation could be really cool. No one is satisfied with their normal memory. It seemed that maybe the technology was getting to the point where it would be possible to build a consumer-facing service or product that really would help you remember everything. So I started doing a bunch of research, and I very quickly ran into Stepan Pachikov and his company Evernote, which was making this stuff and had a very similar vision about what they wanted to do with it.
I went to California, and we went out to a sushi place and talked stuff over. Stepan gave me a demo, and I was really amazed by his vision. He had really refined a lot of these concepts to a much higher degree than I had. And the technology was fantastic.They had built all this image recognition technology that let you search through printed and handwritten matter, and it was years ahead of the competition. He had a team that went all the way back to the Apple Newton, and had kept it together for years and years.
X: Exactly how was Pachikov involved with the Newton?
PL: The original Evernote guys had founded a company in Moscow called ParaGraph, which provided all of the handwriting analysis for the Newton. The cool stuff that the Newton did was all from ParaGraph. It was right at the end of the Soviet Union, and they were the first company [that] went to the West—SGI bought them right as the curtain was starting to come up, so it was part of the first wave of Russian entrepreneurial resurgence.
So I figured these guys had at least a five year head start, but they didn’t really have much of a product. It was kind of a science project. There was a business plan, but it wasn’t really credible. It was a bunch of disparate little shareware applications, not really a mature product. And they were looking for a CEO. I didn’t initially have any interest in working at somebody else’s company.Both of my previous experiences had been with companies that I started. But the board members included people like Esther Dyson and Max Levchin, and they kind of talked me into it. That was three years ago this month.
X: Most of the press materials about Evernote make it sound as if the company was founded in 2008, but obviously it goes back a lot farther than that.Wasn’t there a Windows note-taking application, for example?
PL: Relatively few people know the company from before June 2008, which was when we launched the Evernote service, with a full rewrite of the components, the user interface and the architecture. There is not a whole lot that’s left over from the original days in terms of the technology. The image recognition stuff that’s running on the server uses all the same algorithms that were developed back then. But all of the consumer-facing stuff is new. And we had to do a full recapitalization, so in effect it was a re-launch of the company.
X: It sounds like your first focus when you came onboard in 2007 was on fundraising, not technology.
PL: Yeah. The company had no money when I came on. I think we had three or four weeks of payroll left in the bank. All sorts of things were a mess. It was run by scientists, who had all sorts of fascinating software and great people, but none of it was particularly simple or coherent. I was kind of in a weird position, because I was explicitly brought in to be the discipline guy, which was kind of funny because I had always been the punk kid before that, the one who came up with stuff and then the real adults would come in and run it.
X: But even if the technology had been in more coherent shape when you got there, it seems to me that all the pieces you needed to make this “memory prosthesis” work, including smartphones and 3G wireless, weren’t really in place before 2007 or 2008.
PL: That was exactly my thought when I started researching the business. I’m a giant nerd. I’ve been following lifelogging and Gordon Bell’s work and the MIT Media Lab and all of that stuff for years, and I’ve always been a big fan of the idea of a cognitive prosthesis. Every other aspect of our bodies’ capabilities has been magnified thousands-fold by technology in the past 10,000 years. How fast I can travel, how much weight I can move—all that stuff is many orders of magnitude better. But the number of things that I can keep in my head is still pretty much the same.
So the universal demand was always there, but as you said, until recently you did not have the right factors in place—such as affordable cloud-based computing, good image recognition, and smartphones and other devices—that were powerful enough so that it was almost as good as having it in your head. In the 1990s, it was nowhere near consumer-ready. You had to be very far out from the mainstream to take advantage of this stuff, like when Gordon Bell started MyLifeBits. But it was getting to the point, when I was thinking about this in 2007, where it felt like in the next five years everything was going to line up.
Continued in Part 2.