Interview with Phil Libin (Evernote)
Evernote helps you remember and act upon ideas, projects and experiences and is synced between all the computers, phones and tablets you use.
To anyone who’s never come across Evernote, how would you describe it?
Evernote is your intellectual brain, it’s a service that lets you remember everything important that happens to you and use that information in a way to make yourself happier and more productive.
When you joined, it seems like you took some great projects and helped combine it all into a product and viable business. You joined a team with a largely scientific background, what were some of the difficulties you had at this time?
There was actually two teams that we brought together in 2007, a team of mostly scientists and researchers that had been working on image recognition and this concept of memory augmentation for a long time and I had a startup team mostly of engineers that were interested in making a product along the same lines. Making a product that would help people remember everything and combine those teams in the summer of 2007 and launched the first Evernote service in 2008.
I think starting with two teams presents slightly different challenges than a typical startup where you just go with a small number of people that are all together, but on balance it was definitely a very positive experience.
Could you tell me a bit about Evernote founder Stepan Pachikov and Dave Engberg CTO?
Well Dave and I have know each other for many years, he was the CTO of my last company CoreStreet, he was part of the original team that was with me at the beginning in 2007. Stepan started a different team of people in Russia in the late 80s, they developed a lot of the really cool technology for the Apple Newton, the very first tablet computer back in the late 80s early 90s and Apple actually brought them from Russia to California and then they sold their company to SGI. Stepan and his team have been working in this area for a while. So Stepan had one group of people and I have Dave, Andrew and a bunch of other people and we met up in 2007 and joined forces.
With around 20 million users you must receive a lot of feature requests and suggestions. How do you manage all of them?
I think we may have just passed 23 million. We get quite a bit and we really want to hear from users and see what they think. We’ve found that different types of user feedback is relevant in different ways. The least relevant is when you ask users what should we build, having users do product design or feature roadmap for you doesn’t really work. Users aren’t designers they don’t particularly know what they want so in terms of how we decide what to build. We are the target market, we are the users that we want to build for. We build software that we want to use and so anytime we get a recommendation or suggestion for something new, the filter we run it through is whether it’s something we want to use. That does happen sometimes and if not then we don’t build it.
What user feedback is extremely useful for, is feedback about particular things that we’ve done that aren’t working particularly well that people aren’t understanding or that they don’t like. The negative feedback and the feedback about how to improve specific features, that is the most audible feedback. Feeback about big ideas are just as vocal, but less ultimately valuable.
So have you had any feature ideas yourself which you’ve managed to push through?
Everything we make is basically made for me, it’s what I want to use. For me and for the rest of the team, we’re a bunch of fairly like minded people, it’s a diverse group in terms of backgrounds and tastes and ages and demographics, but we’re all nerds and passionate about the idea of remembering things. So we really build things for ourselves. I guess I am kind of king of the nerds, so a lot of the stuff I want gets built.
Who started the notion of Evernote as a 100 year company and how does this idea make itself apparent within the company culture?
Evernote is my third startup and a lot of the original team were with me or with Stepan for a couple of startups so it’s a pretty experienced team, almost everyone in the initial group had built 2 or 3 companies before. I think we had a common experience, theey were successful companies with good exits, but we were always building them for someone else. The products we built weren’t products for us, my first company built e-commerce software, so we sold to a lot of retailers. My second company built security systems for government and big banks. Those were all interesting, but we weren’t a government or a big bank, we weren’t a retailer. We never felt like we were building it for us. Ultimately we sold both of those companies, so it wasn’t even a company that we got to keep and when we were doing Evernote we thought, the first two companies we built for somebody else, the third one, lets build it for us, lets make a company we want to keep. I think from the very idea, this concept that this is the one we want and if you’re building something that you love and you’re building it for yourself you shouldn’t want to sell it, you should want to keep it.
I don’t think we thought about it as a 100 year company right away, but we knew we wanted to do something that was very long term, we weren’t interested in flipping it. We were interested in building something great. The 100 year stuff, probably most of it was influenced by Japan. We started traveling a lot to Japan, just because Evernote got very popular in Japan so we started doing a lot of work over there. We had a lot of partners and a lot of users and some of our partner companies were these Japanese companies that are more than a 100 years old and they’re still innovating and we felt really inspired by that. Let’s try to combine the best of both worlds. Let’s build a company that has the long term planning and thinking of some of these great Japanese companies, combined with the best of the Silicon Valley startup mentality. We don’t just want to build a 100 year company, we want to build a 100 year startup.
Am I correct in saying that roughly 5% of users upgrade to Evernote premium? What’s your philosophy on converting free members?
The overall percentage fluctuates quite a bit and isn’t particularly important because what happens is users convert as they use the service, the longer you’re using Evernote the more likely you are to convert. The conversion rates go up with the age of the user, so if you just want to take the overall conversion rate, you’re dividing the overall premium users by the total number of users, that’s primarily influenced by the ratio of new users to older users of the system. In a month when we grow really fast we get a whole bunch of new users, the overall conversion rate will be lower because more people haven’t been around for a few months, if we slow down growing then the overall conversion rate goes up because a larger percentage of users are older returning users.
The overall rate we frankly don’t pay attention to, but it’s probably somewhere around 4 or 5 percent range. What we pay a lot of attention to is what percentage convert in the first month, in the first year, in the first two years. Our oldest users, the ones who have been using Evernote for about 4 years, I think we first launched the service 4 years next month actually. Of the first few months worth of users, people who have been with us for about 4 years, around 25% of them are paying us right now, that ratio declines as you get new users.
We don’t care if you pay, we just want you to stay around and keep using it and get all your friends to use it. The longer you stay, the more likely that you’re going to fall in love with it and then pay.
In July 2011 you received $50 million in funding and have raised almost $100 million total. Do you think this kind of funding is necessary to create a world changing business/startup or do you have other motivations for raising such large amounts of capital?
We never needed all of this funding, we haven’t spent the majority of the money we’ve collected. We may very well take additional funding even though, we don’t need it for the cash. We really see funding as just another type of infrastructure we’re building, the goal is to build a 100 year company, when you start planning for a 100 year company you have to take that seriously and say well how do we structure everything for longevity. Part of that is how do we build out our data servers? How do we build out our architecture, how many people do we hire and how do we build out our management structure? A big part of it is investors and how do we put in the financial infrastructure that will last for 100 years. When we think about funding, we really think about in the context of wanting to isolate ourselves from having to make any short term decisions and being influenced by fluctuations in the market. We don’t want to have to time things. Is the IPO window closing, all that stuff, so we want to make sure we have investors who have the same time horizon. Investors who have this very long term vision and then whenever we can put together some financial infrastructure at good terms from the right investors, we’re always happy to take it because it’s just another tool to make sure we can achieve the company we want.
Where are you currently focusing your spending?
We spend money on the same things that we always do, it all goes to the product. The vast majority of the money we spend, the people we hire are building the product. So it’s developers, designers, QA, managers, testers, support people, operations people. We don’t have a sales culture, we don’t sell anything, we don’t have a sales team. Everyone is working on the product, we get more money as we increase spending, it all goes into the product account or the majority of it does.
How involved are you with fundraising?
We don’t do very much of it anymore, we haven’t needed to raise money in several years. The first few years of the company that was extremely involved, it was very time consuming, very difficult for us to get money the first couple of years. Really the whole team, but I suppose mostly me spent a lot of time and energy on it, now that it’s less important I don’t personally spend a whole lot of time on it myself.
We don’t go out and look for money, we get a ton of inbound requests on a daily basis from people who want to invest. Most of them we don’t talk to, it doesn’t take up much of my time these days. I spend most of my time on what I want to be doing, which is basically product design.
Back in 2007 everyone was pitching social startups. How did you convince investors that unsociable was right for Evernote?
We didn’t really convince them, we had almost no investment until 2009. 2007, 2008 we never had a few weeks of money in the bank, we were putting a lot of money in ourselves. I put money in, Stepan put money in, we were getting a lot of friends and family for funds. We were getting money from some of our more passionate users, from Russia, Japan, Canada, but we didn’t really have professional venture investors until 2009. Everyone was pitching websites and social, we were the opposite of both of those and we suffered because of it. We couldn’t actually get investors to pay attention until we had the data, it all flipped around in 2009 when we actually had a year or so of data and we could show the covert analysis and what was actually happening. Then it turned, almost overnight we went from having to constantly beg for money to having multiple term sheets.
So would that be your tip to someone looking for investment, create your product, make it successful to a certain level and then take it to investors?
I think the central insight is to match the investors that you’re talking to with the strengths that you currently have and the stage of the company that you’re currently in. There’s a type of plan that’s very common, which Evernote definitely falls into, it’s fundamentally not an investable concept. We couldn’t walk up to an investor and say here’s our plan, we’re going to make some software and it’s going to let you write stuff down and remember things and we’re going to give it away for free. We don’t have any experience in the consumer space either, please give us some money. That was the pitch. It doesn’t work.
The idea behind Evernote isn’t particularly novel, no ones going to hear, “we’re going to use computers to help people keep track of information” and be like “ok I’ll write you a cheque”. There are other ideas that are not on the face of it really brilliant ideas that you may be able to do this with, but certainly not 99% of what people pitch. For us it was about showing we had traction, we understood what our users were doing, we understood what they wanted. We could deliver a product, meet on time, deliver our plans and fundamentally we got the unit economics right. Every time we added a user, we made money. Once we showed that and we showed that in a very rigorous way, we’re fanatical about the data and how we present things. Then it became much more real, investors were much more willing to take a chance on us. It was still very high risk, everyone understood that it was still very unlikely, there was a very high degree of uncertainty, but we could show proof, that worked out well.
How did Max Levchin (PayPal founder) become involved with Evernote?
He was actually involved with Stepan. Max Levchin and Esther Dyson were actually both on the board of Stepan’s company before Stepan and I had even met. Max and Esther were probably the two people most instrumental in talking me into combining the two teams and making a unified Evernote.
Evernote has native apps for most operating systems, both desktop and mobile requiring varied skill sets and many teams. How many people are currently working at Evernote and how do you find talented people?
We have about 104 people right now worldwide, which has grown really quickly. We were only at about 40 or 45 people just last year, so we’ve basically tripled in size in a year and we might triple again this year, which is a little bit scary. We’re probably going to go from 40 to 130 to close to 300 maybe in a year. It’s difficult to get excellent people, it’s always the hardest thing. It hasn’t been nearly as bad as I thought it’d be because so many people that come in looking for work at Evernote do it because they love Evernote. People will come to us and say I love Evernote, I love what you’re doing, I want to help build it. We draw in a lot of people who specifically are not just looking for work, they want to work with us, be part of the adventure. That really helps, that gives us a very big qualified group of people and of course we hire people everywhere in the world. We have people in California, Texas, Switzerland, Japan, China and all over the place. We tend to attract the best and the most passionate product people from many of the top countries in the world.
How involved are you with the hiring process?
Not just hiring, but building the team is a fundamentally important thing, it may be the most important thing. It’s what I tell all of the managers as well. The product we build is built for us. Evernote could only have been built by our team. A different team wouldn’t have built it, they would have built something different. We’re not working off abstract customer requirements. We are the customers. To guarantee the long term success of the company, the only way to do it is to make sure that the team continues to be the right kind of people. If we become the wrong kind of team we can’t build this product anymore. It’s pretty important and all of us are deeply involved in it, luckily I have various executives that have become excellent at keeping and recruiting people and evangelizing. It’s working relatively smoothly.
The HR job is almost the most important one at Evernote because the HR department is in charge of building the company that’s going to build the products. It’s what has to come first.
How quickly did you realise that native clients would be important to the success of Evernote?
Realize is maybe not the right word. We bet on it. Our intuition, we didn’t know if this was going to be correct, but our intuition right form the beginning was that people are going to want the best possible experience on their devices. So much of the growth of the mobile industry, of app stores, so much of that is generated by rapid improvement in the quality of the hardware and in the devices from generation to generation. The shiny new tablet that you buy now is a lot better than the shiny new tablet that you bough last year. We thought while this really rapid improvement in consumer devices is happening people are always going to want the best possible experience. Evernote has to work better on your brand new phone than it did on your crappy old phone. If it doesn’t give you a better experience then what was the point? The only way to do that is to write all the clients natively. To go down the bare essence of what the device can do and say, how do we make the best possible experience on this great new device? If you take the lowest common denominator approach using cross platform stuff, you by definition create something that’s average. You get an experience that’s the same on your shiny new phone as it was on last years phone. If it’s the same experience you’ve removed the motivation for the manufacturers and publishers to promote you. You’ve removed the excitement of people who get their new device.
That was the bet we took, at least for the next decade or two, consumer devices are going to be improving so rapidly that it’ll be important to take advantage of that. If that ever slows down, maybe we can go to some sort of lowest common denominator approach, but there’s certainly no evidence that the improvement in phones and tablets is slowing down anytime soon.
You gained your first half a million users in around just 10 months. How did you get off to such a quick start?
Our user growth has actually been really smooth from the very beginning. It’s actually quite surprising, I just dug up the original model we used when we closed our first Silicon Valley round with Morgan Capital in 2009. I found the original plan, we built this model for them that shows how we’re going to grow, both users and revenue. It was from 2009 and that plan went to the end of 2011. The end of 2011 back then was unimaginably far in the future for us. I just found it, I wanted to compare how we did against what we projected and we were within 5% on both users and revenue, that’s kind of amazing. The growth has followed this really smooth exponentially growing upward slope. It’s just driven by word of mouth. We don’t spend any money on user acquisition, we don’t do any SEO or SEM. There’s no tricks. We don’t pay money for users. We don’t pay for incentive downloads or any of the stuff you hear people talking about, we don’t do any of it.
It’s all just word of mouth and people finding Evernote because of friends who love it recommending it. We’ve probably added more than 2 million users in the past month and it just keeps increasing because of that.
What were some of the initial successes you had with gaining users, up to to get say the first 20,000 users?
We launched closed beta on TechCrunch, we were lucky enough that TechCrunch wrote about us right as we were starting the closed beta and we gave away 100 invites, that was the first spark. We had a couple thousand people within the first few days just because of that really early spark and it just grew from there.
What we did, is we really killed ourselves in the first couple of years to always be in all of the app store launches on day one. Whenever a new device or platform would come out, we would work days and nights for months before that to make sure Evernote was there and supporting the new device or operating system in the app store on the first day. So that it could be one of the showcase apps for all of these devices as they launch. We were mostly successful. When iPhone launched we were one of the very first iPhone apps, so we were promoted and had a lot of visibility. When iPad launched, we were there on day one, not just with a port of our iPhone client which a lot of other companies did. A completely new designed version for the iPad even though we’d never seen an iPad before we stood inline with everyone else. Same thing with Android devices and Kindle Fire.
In the early days the we put in a lot of hard work synchronizing so we could follow in the wake of all these massive new releases and benefit from those. Now that’s less important to us because the momentum is already so high, but in the early days that was fundamental.
So do you think the timing was perfect because that period was when apps as we know them now were taking off? You were there at the beginning of a big change in mobile.
We definitely were very lucky with the timing, absolutely. I think that’s mainly true of many successful companies, in hindsight you look back, it’s always easy to point to things where the timing worked out. We were very fortunate that the app stores were launching right as we were getting ready. If we were 6 months behind in our development, we would had missed all of it.
The logic works the other way as well, there’s many events where the timing was horrible. We ran out of money, we came within days of shutting down the company in October 2008, this was the worst time in the history of the universe to raise money. That was when Lehman collapsed and the markets were crashing. No one was giving any money. That’s bad luck, needing to raise money at the worst time for raising money is also not something we anticipated. In the long term you have to deal with both the good luck and bad luck that’s dealt to you and what you make of it is ultimately what determines the success of the company.
You had over 125,000 users before you came out of closed beta mid 2008. What was the main takeaway?
The main purpose of the closed beta was doing the user testing, making sure we understood what people actually wanted and that it worked. This was a new experience for most people. We had native apps that were synchronizing in real time, we had a new type of server architecture that we were tweaking, so a lot of it was really just for technical reasons. We couldn’t support a lot more people, but it also started as a good way to generate interest. The fact that you have to sign up and send around invites to get in actually generated some buzz. That was never our intention, we never thought of the closed beta as a marketing exercise. We were frankly terrified that everything would crash all the time. It actually turned out to be pretty good as well as just a way to generate interest. I wouldn’t recommend that any company starting up today go through anything like that purely for the sake of generating interest. I think users can smell that. They can smell marketing tricks and you tend not to respond to it very well. But for us it worked quite well.
We were almost forced out of beta by Apple. The iPhone app store launch in July 2008, Apple doesn’t allow betas, you can’t have a beta in the app store. We knew if we wanted to be there on day one, we had to be out of beta, so we worked 20 hour days 7 days a week for a few months before then to make sure that we could have something that was genuinely available and out of beta.
Everything about Evernote, from the product to the business model seems to be based around simplicity. How important is it for you to keep Evernote simple?
That’s a real balancing app, I think the core Evernote app is too complicated right now. I don’t think anyone understood design philosophy, especially mobile app design philosophy in the early days, we certainly didn’t. A lot of initial designs were very feature oriented. We had check lists of features we wanted to implement, we went through and put them in. What we’ve realized since then is that isn’t really how you should build software, you really need to be experience driven. Focus entirely on what is the experience of using that and then add the features almost as a secondary criterion. So I think a lot of our newer products are very straightforward and simple and a lot of initial products aren’t. But we are in the process of tweaking the designs, sometimes in major ways to simplify them. Of course you don’t want to lose any power from those apps while doing it, so it’s very difficult.
It turns out it’s much much harder to make something simple than complicated. Making something complicated is easy for developers, because it’s always easier to add things. Making something simple is very hard. We’re just starting to get good at it.
Do users tell you how they’re using Evernote? What are some of the surprising ways in which it’s being used?
I think it’s an early sign that you might have a successful product, when you see people using it in ways you totally didn’t anticipate. That started happening right away at Evernote. Use cases are very broad.
I was just in Tokyo a couple of weeks ago and went to get my haircut. The hair stylist was using Evernote to basically run the business. So she would take pictures of all of her clients and their hair styles every time they came in and take notes on them. She had Evernote on iPads and would flick through so whenever someone came in they could see what they’re hair styles looked like before and what some of the other people had gotten and choose what they wanted. All the notes, shampoos were in there, so she was using Evernote as a source of both inspiration and management for hair salon, which was certainly not something I ever dreamed of, but a really cool use case and there’s thousands more as well.
When you look at companies like Dropbox, do you think that’s the future now, as in flexibility and simplicity? Letting the user decide how they want to use the product.
I think Dropbox is doing something really great. They’re the ubiquitous hard drive in the sky. They made this industry and got a head start on everybody else, so now Apple, Google, Microsoft, Amazon and everyone else, every cellphone carrier, they’re all trying to compete with Dropbox. Dropbox just did it really well and got a lot of love and deserves it. The primary use case is fairly simple, you shouldn’t have to think about the concept of computers and files. What we do at Evernote is philosophically pretty different. We’re not an infrastructure play, we’re really about how you experience your life. The experience of your memories, what they mean to you what can you use them for to make yourself more productive and happier. We really focus more on making a beautiful user experience, but I think the core philosophies are quite similar and I’m a big fan of Dropbox myself.
What do you wish you’d known when you became CEO?
We certainly made lots of mistakes. The fundamental thing that I wish I’d of known, is I underestimated the importance of simplicity and design. The first couple of years I really didn’t feel that, I didn’t have a good enough appreciation that the most important thing for consumer facing software was that it was beautiful, simple, people immediately and intuitively understood how to use it.
It really wasn’t about making it more powerful, it was about making it more natural. I think I’ve only internalized that in the past couple of years and it’ll take another lifetime to really master it.
What are you most excited about at the moment?
Wow, I’m generally a very excitable person, I figure if I need a one word description for myself, if I need to put something on my tombstone. I just want it to say I’ve certainly been an enthusiast. I’m excited about just about everything, I think this is the best time in the history of the world to be starting a technology company and running a technology company.
We live in almost a geek meritocracy right now. Where all you have to do is make a great product and you can be successful. That was never true before, that wasn’t true for my first two companies. You could make a great product, but you still had to worry about everything else, all the channels and logistics, advertising and relationships. It was very tough. Right now, especially in the consumer space and because of things like app stores and tablets, if you’re a nerd and you focus passionately on making a great product you could be really successful and change the world and that’s a deeply exciting thing for me.
Finished reading? Check out Evernote!