Lessons learned building an anonymous social network

I built Formspring. For a while it was the fastest growing social network in history, but ultimately ended in failure. One of the defining characteristics of Formspring was that users could send questions to each other anonymously. There are a lot of reasons why we grew quickly and why we failed as a company, but it’s clear that anonymity was a significant contributor on both fronts.

As apps like Whisper and Secret have launched with early success, it’s likely that we’ll see more entrepreneurs experiment with products that incorporate anonymity. I think it’s worth sharing what I’ve learned in the hopes that it’ll help other entrepreneurs avoid some of the problems Formspring ran into.

Anonymity can be a good thing

I should lead by saying that I think anonymity can be a good thing. At Formspring, we spent a lot of time thinking and talking about ways to encourage expression. One of the reasons why I built Formspring in the first place was that I felt like the standard model for social interaction (aka Facebook) felt too rigid, too forced, too unnatural — it forces constraints on communication that don’t have any parallel in the offline world. Do I walk around with a name tag that shows you my birthday, my interests, and pictures of my mom?

Anonymity (and pseudonymity) is important because it’s part of how we naturally interact with others. We don’t have one “real” identity—we hide and reveal different dimensions of ourselves depending on the context of who we’re with and what we’re doing. Even more to the point, many of our interactions start off anonymously before growing into a relationship that’s structured by identity. I can strike up a conversation with a stranger between BART stops, stand up and ask a question during a conference Q&A session, or jump into a pick-up basketball game—and find those interactions rewarding without establishing my “real” identity ahead of time. It’s good to have equivalents to this online.

Of course, the problem we struggled with at Formspring was finding the right balance between all the ways that anonymity can be used in constructive ways without devolving into a product where anonymity is only used to say hurtful things about others. It’s tough, but I don’t think it’s impossible. With the right product and community mix, I think anonymity can be a very powerful tool for encouraging dialogue and sparking relationships. We saw incredible things happen on Formspring that wouldn’t have happened without anonymity—from the countless new friendships created around common interests, to support groups that formed to deal with difficult issues.

Fight perception early

One of the problems we faced is that it’s easy for people to instantly equate anonymity with abusive behavior. While anonymity can certainly breed abuse, it’s not necessarily the case. And it completely ignores the fact that abuse is rampant on apps that discourage anonymity.

In cases where we saw someone being abusive or being abused on Formspring, it was almost a given that we could find similar behavior on Facebook or other apps where that user spent their time. And sadly, we would frequently hear heartbreaking stories of abuse that was happening offline as well. The problem with bullying is bigger than any particular technology, and the solutions are much more complicated than might seem on the surface.

Given this, it’s not only important to find the right ways to discourage abuse, but to fight public perception early. Because people equate anonymity with abuse, it’s so much more important to get ahead of the conversation by being explicit about your vision for how anonymity should be used in the product in positive ways. I wish I would’ve tackled this much earlier at Formspring.

Perception is important for a lot of different audiences—employees you’re trying to recruit, partners you’re trying to work with, law enforcement, policy makers—but it’s most important with users. If users can’t see your vision beyond being a utility for posting abusive content, then it’ll be hard for the service to become more than that. That vision will need to be obvious as they explore the product, and in how they hear about the service from others.

Perception is probably most difficult in the press. Unfortunately, sensationalist headlines and soundbites work better than a nuanced discussion of complex issues. So it goes. This makes it even more important that your vision be clear to the public early.

The first mainstream media stories written about Formspring were about bullying. We hadn’t told our story well before that, so even as we worked on incredible things afterwards, it was was hard to overcome the initial perception about the company. When people read about abuse happening on Facebook, they already knew it as a place to see pictures of their grandkids. Without any context, a story about abuse happening on a new anonymous social network only cemented the impression that the purpose of the product was to encourage abuse.

Solve for long-term engagement

It’s hard enough for anyone to build a social app with long-term engagement, but anonymity introduces some unique problems.

Where anonymity was being used for abuse, we found that engagement had high amplitude but short wavelength. Like with any service, salacious content resulted in more activity in the short term, as users jumped into conversations to attack or defend each other. But most people don’t enjoy spending their time that way long-term. It’s easy to get burnt out and abandon the service forever. So combating abuse is not only the right thing to do, it also makes good business sense towards ensuring users stay engaged long-term.

Another problem anonymity poses for engagement is that it might create an unnatural barrier for continuing conversation. It’s easy enough for me to exchange phone numbers with the stranger I met on BART, but hard to reach out to the anonymous poster that seemed particularly interesting to me. That ability to shift modes of conversation can be helpful in keeping people engaged on the service long-term. It really depends on the core problem you’re solving, but most users aren’t looking for an anonymous utility—they’re looking for ways to express themselves and interact with others.

Work with the right people

The best advice I can give is that as you work on the policies and products to help shape the community, work with people who’ve dealt with these issues before. Even though I’d had a decent amount of experience working on sites with user-generated content, users used Formspring in ways that surprised me.

It’s easy to find people with strong opinions on product and community, but so often the things that sounded right at first weren’t really effective or had unintended consequences. Though we tried a lot of things early on, we didn’t get much traction addressing abuse until we started working with people that had direct experience working on these issues. Most often, they were people that spent their day thinking and learning about the type of problems online communities face.

It’s incredibly challenging to build a healthy community at scale. As challenging as it is, it’s harder still when anonymity is part of the product. Because of this, most people will tell you to rip it out. Investors will pass on your company, peers will criticize you, parents will yell at you. Maybe they’re right. And maybe not.

The best policy people I worked with understood the vision I had for Formspring, and why anonymity was part of that. They didn’t try to strip out everything in the product that made it unique, but worked with us to figure out how to fulfill that vision while building a healthy community.

Thanks to @sacchetti for encouraging me to write this, and reading an early draft.