Ev Williams has co-founded a handful of Internet companies and seen Blogger and Twitter from fledgling products to household names. But even successes have some snags along the way. With his new venture, Medium, he’s striving to make every step of the process, from putting together a team to fine-tuning features, even better. In his recent Bridge talk, Ev shared his thoughts on building a company and product—and why good design is so vital to today’s tech.
Making something initially difficult to use or access can make it more rewarding. Getting started with Twitter, for instance, was tougher than just making an account. “If it’s hard to figure out, and you figure it out, you’re cool,” Ev explains. If there’s some friction to the process of figuring out a new program, people feel like they’ve earned it. Once people had gotten the hang of how to use Twitter, Ev says, they tended to keep using it for a long time.
Similarly, writing on Medium was invite-only until recently. This friction in accessing the site wasn’t just enticing to users; the slow rollout also provided the team with a small but steadily growing test group. That let them see how the system really worked at an earlier stage, without getting stuck doing things a certain way. At Twitter, Ev and his team rolled out the suggested user list early on, intending it to be temporary—and found themselves stuck with it for a long time, even though they didn’t like some of its effects on the site. This time around, “we wanted to reserve the right to change things drastically, and the more people there are, the more difficult that is,” Ev says.
After 20 years of working with the Web, Ev realized: The greatest benefit of going online isn’t that we can do new things. It’s that we can do the same stuff we always do, with less hassle. Good products help us do just that, tapping into what he calls “the internet of convenience”. When we’re online, we’re mostly going about the business of daily life, “whether it’s buying stuff, finding some information to help you solve a problem, learning something, or connecting with people,” he says. “The [internet companies] that succeed are the ones that help you address a widely held human desire more conveniently than you could before,” allowing people to accomplish something faster or in fewer steps. As part of a startup, your goal should be to help people do their own thing, better—not to give them new things (even cool ones) they have to do.
A single product often has a multitude of features. But Ev is hashing out a framework for sorting those features into three types: definitional features, improvements, and transformations.
Definitional features are the things that make a product what it is: profiles and friending on a social network, for instance. Anything else you add is an improvement or a transformation. The difference between the two is how much they change the product. Improvements make the basic product better. Transformations radically alter it, taking it far beyond its competitors—but they can backfire if they don’t work for users. “The things that are extremely high reward—potentially transformational—of course have to be high risk,” Ev says. “Improvements are like putting your money in bonds. They’re probably going to return,” but rarely dazzle. Any startup has to make choices about how to invest its time and resources, in risky but possibly game-changing transformational features or in surer but more incremental improvements. At Medium, Ev and his team are aiming high: 70 percent of the features they’re working on have the potential to be transformational.
As Twitter grew under Ev’s direction, he found that scaling the company made it harder and harder to get quality work from each person. So at Medium, he’s leaving behind the traditional corporate structure and forming teams using holacracy. It’s a management system with its roots in Agile software development, where things are built in sprints rather than extended projects.
Teams at Medium work the same way. No one’s a product manager from day one onward. Instead, each team is geared toward some aspect of the product, and each person on has a specific role that leverages their talents to help the team. As needs shift, teams and roles do, too. “You dole out the rules and responsibilities [within each team], and then you do what you need to do until it’s time to change,” Ev says. “It’s much more dynamic and fluid.” People can focus on what they’re doing this week or this month, not what their title is.
Long gone are Internet’s early days, when design was an afterthought. “If you’re starting a company and thinking about where your value comes from, it’s increasingly design,” Ev says. There’s still a lot of difficult engineering to be done, but in many spaces, multiple companies have mastered the basics. The way to get the most out of that functionality, he says, is “by building great experiences.” And there, design has the opportunity to set a product apart.