Evernote’s CEO on the company’s long, tricky journey to fix itself
"Silicon Valley is addicted to momentum," Evernote CEO Ian Small said, by way of explaining how his company had come to be, as he put it, stuck. "And to have momentum, you need to keep shipping things, and to keep shipping things, you need to keep pushing problems sometimes into the corner and look the other way. And eventually, the problem gets bigger than the room that you're in."
Sign up for Protocol newsletters
When Small joined Evernote in late 2018, the problem had eclipsed the room at the once-highly-regarded and still-hugely-popular company. (Even after years of problems, it still has more than 250 million users.) Evernote had five different apps run by five different teams for five different platforms, and each had its own set of features, design touches and technical issues. Internally, Evernote employees called the app's codebase "the monolith," and that monolith had grown so big and complex, it was preventing the company from shipping cross-platform features or doing much of anything in a short time. While other apps were becoming faster, more useful and more powerful, Evernote was slowly becoming a crufty, complex relic of a once-great app.
Evernote's been aware of this problem for a long time. "I think anyone could have told you that, looking from the outside in," Small said. "The only question was, what do we do about it?" In 2017, then-CEO Chris O'Neill moved Evernote's data to Google Cloud and started the process of unifying the codebase. It shipped something like an overhaul of the service, along with a few new features and big ideas about the future.
But the brand-new Evernote still had too many of the old Evernote's issues. Small felt the progress on this core stuff was slow and not sweeping enough. So not long after he started, he told his team that rebuilding Evernote was now their only job. They would stop building new features and new products for as long as it took to fix the core of Evernote from the ground up in a way that would work better going forward. The note editor, the apps, the search, the cloud infrastructure, everything had to change. Small originally thought it might take a year, maybe a little more.
Now, 18 months later, Evernote's launching a new iOS app that has the same name and look as the old Evernote, but is otherwise almost entirely new. New apps for Windows, Mac, Android and the web will follow "in the coming weeks," Small said. Why iOS first? "Oh, because the app is ready. We're not trying to hit some deadline, we're just putting the apps out when they're ready, and an iOS [app] is ready now."
Ahead of the launch, I chatted with Small over Zoom as he sat in front of an Evernote-branded virtual background and told me what life has been like inside the company as it has worked to rebuild itself.
···
I would assume you'd come into Evernote and everybody internally is aware of this reality, right? It can't be lost on people that there's all this tech debt, and things are overcomplicated.
Everybody at the company knew that. All you have to do is walk around and ask five random employees what the problem was, and they would all tell you the same thing. Or, if you go out and you ask five random customers, they wouldn't tell it to you in the same words, but they would tell you the ramifications of that problem.
So all it really took was, at some level, the constitutional fortitude to say 'we're actually going to do this.' And I'll be honest, I think on the day that I stood up in front of a company and said we were going to do this, I think there was skepticism in the audience: Oh, yeah, we've heard this before. This is going to last three months until he says, 'No, we need to ship something quick. Just patch this over here and put some more duct tape on and let's go.'
I knew when I came in that something like this was going to need to be done exactly how we did it. And from my standpoint, this wasn't something I needed to convince the board about. The board knew that this is where the company was. Everybody knew it. Someone just needed to make the decision, and that someone ended up being me.
It seems like doing it this way, the stakes get a little bit higher. To say, 'we are going to fix this' and then go dark for a little while, that ratchets up people's expectations. What happens when you come back out into the light?
Somebody internally described it as, what we're doing is we took a huge breath, we dove underwater, and now we're swimming. And the longer we're underwater, the more our lungs are burning, and the harder it is, but we don't get to come up until we're done.
That's absolutely what it feels like on the inside. We know that there's a tremendous demand from our users for us to move forward. We know there's real problems in some of the apps that are out there, both architecturally and from a standpoint of bugs and from a standpoint of consistency.
We did something unconventional on the inside, and then we matched it with an unconventional communication strategy, which was to just go out like in my first few months and say, "Right. Here's the state of the world: The app's kind of broken, we're going to fix it, it's going to take a while." As you go through the process, being transparent brings the users on your side and lets them into the process so they can be excited about what's coming. Of course, when they get excited they get impatient, so it's a tricky balancing act.
···
Evernote underwent this transformation at an interesting time. A new generation of productivity and note-taking tools, like Notion and Roam and Dropbox Paper, were becoming popular choices in part because Evernote had lost its way a bit. Many of these new tools built Evernote importers, specifically hoping to capture dissatisfied users. Arguing about the merits of Evernote versus another tool became the productivity aficionado's version of the "Why I'm leaving New York" blog post.
There were also lots of new features being implemented into these other tools, from integrations with other services to things like bidirectional links and interlinking databases. Evernote, in this new launch, will have none of those things. Small said he's been fighting the urge to bake loads of shiny new features into the first iteration of the new Evernote, and he's helping his team do the same.
···
I'm sure you're watching Notion and Roam and all these other apps come up. Has part of this process been about, OK, does Evernote need to change what it is going forward?
Thinking, yes. Is this moment in which we're shipping representative of that thinking? No. Is what you'll see over the next 12 months? Yes.
It's hard enough to rebuild a platform and fix it to the extent that we did. If you try and reinvent it at the same time, you really face a challenge in getting people over to that platform. And so our goal has been to improve everything that Evernote is, but the Evernote that will ship on iOS in the coming days is incredibly familiar.
Six months ago, when we could start to see things were coming together and we knew we'd be at the place we are now, we started forking off small teams in the company to start working ahead on this new base on the stuff we want to be shipping. That's one of the reasons I can be very confident about the next 12 months, because there are already things in flight that are really significant shifts in the way you think about Evernote, and what it does and does for you.
I know you're not going to tell me the new features that are coming —
Nope!
— but tell me how your thinking about what Evernote is and means to people has changed over the last year or so.
Stepan [Pachikov, Evernote founder]'s original vision was to be an extension for your brain. And, I mean, it would be hard to ask for a larger vision than that! There's still a lot of room in there.
Very quickly that vision turned into a single unifying mantra, which was "Remember Everything." And Evernote got very good at capturing things: obviously notes, but web clips and images and cameras and scanning and all this stuff, because it was all about Remember Everything.
We have extended the idea of remembering to say that what we really want to do is let people remember everything and accomplish anything. We now want to start to double down: If we have all this stuff for you, how can we help you accomplish things? And we have some very concrete ideas on how we can do that, based on how people already use Evernote today, but also based on pushing the envelope for what people might think Evernote is.
···
I asked Small what Evernote was doing to avoid falling into this position again, how it was avoiding yet another reboot in 2030. He said there's not much he could do, at least not yet. "You know, if you go back eight years or something, the Evernote architecture made all the sense in the world!" Going forward, the team needs to continually invest in solving its tech debt problems, rather than papering over them to build something new.
That's been a big culture change for Evernote. The first step, Small said, was to get everyone working toward a single goal, rather than each split off on their own product. He reorganized the company to focus teams on features and functions, rather than platforms, in an attempt to keep everything thinking about Evernote as a whole rather than a single platform. Small quoted Steven Sinofsky's famous advice: "Don't ship the org chart."
Going forward, the team has to figure out how to branch off again. The note-editor team can go back to building new stuff, as can the folks working on search or design or any of the new features coming soon. But now they know that everything has to be cross-platform, multifunctional, and designed to last.
The thing about a reboot is you only get so many chances. How many times will users stick around waiting for the New New New Evernote that yet again claims to fix everything? Small said he understands the stakes, that after a year and a half of promises, Evernote had better deliver. That's why, he said, the new apps are launching now. Not because people are clamoring for them, not because Evernote needs to start getting back some of the momentum it's lost in the last two years. Because it's ready.