At the end of April, the Silicon Valley startup Secret opted to return a bunch of venture capitalists’ money and shut down. Secret had created an app that allowed its users to share anonymous thoughts. They raised more than $35 million and were covered by every major tech publication, as well as some mainstream outlets.
Although Secret did its best to close quietly, the controversial company was bound to attract postmortems, including an op-ed in the New York Times that blamed “cyberbullying” and “novelty.” Most of these postmortems are rife with misunderstandings of the startup ecosystem, social media, and digital anonymity. For instance, many commentators are claiming that Secret failed because anonymous sharing platforms can’t work. They ignore anonymous sharing platform YikYak, which has two million users and $72 million in venture capital, and competitor Whisper, which has ten million users and $61 million.
I’ve lived and worked in San Francisco/Silicon Valley tech for several years now, and for much of that time, I used Secret with my friends. I also have significant experience developing anonymous and pseudonymous digital identities; last year, in The Atlantic, I wrote about how I used social media to build a pseudonymous blogging brand. Drawing from my experience in the space, and from other Secret users’ commentary around Silicon Valley, it’s obvious to me that Secret failed due to ill-considered product decisions — it did not fail because of anonymity, cyberbullying, or other buzzwordy critiques.
As the Internet has become civilized (one might say colonized), there’s been a strong backlash against digital anonymity and pseudonymity. Lately, commentators often cite cyberbullying and trolling as reasons to enact “real name policies.” But there’s surprisingly good evidence that real name policies do not, in themselves, make lasting improvements to the civility of online discourse. For example, as tech journalist Greg Ferenstein explained in TechCrunch back in 2012, the entire country of South Korea has passed multiple laws intended to enforce real name usage on digital media. Yet after several years of attempts, the laws were ditched because they didn’t meaningfully reduce abusive and malicious comments.
This is not to say that cyberbullying and trolling aren’t real problems. Indeed, these are problems that can destroy communities, unless they are addressed by moderators with specialized skills and tools. But those problems are not unique to platforms that allow pseudonymity and anonymity. Plus, anonymity platforms have a lot more going on than vicious gossip — including concrete business applications.
In other words: Secret was not a wasteland of spite. During 2014, I took many screenshots of Secret, some of which you see in this piece. I’ll use these screenshots to give you a sense of my experience on the platform, and then I’ll explain why Secret’s product decisions killed the company.
As you can see, the secrets of Secret were tagged with “Friend” (meaning the person was among my cell phone contacts) or “Friend of friend.” My friends used Secret to share tiny, poignant moments. Occasionally, they scheduled dinner parties. These sweet moments get lost when people focus on the seedy, salacious side of Secret.
And what was the salacious side? There was the inevitable sex, drugs, and money talk, including plenty of tech industry gossip. But some of my favorite “industry gossip” from Secret demonstrates its users’ knowledge that none of us could trust what we saw there.
These clearly fake “rumors” showcase Secret users’ collective understanding that anything we saw there could be a lie. In general, we traded a lot of jokes on Secret — jokes about our lives, our work, and about Secret itself. Yet in the midst of this madness, tech employees managed to discuss important but taboo business topics, and to share information under the veil of plausible deniability.
Take this secret as another example. Was it true or false? I don’t know, but if it was true, then this was valuable information for everyone involved.
There were many Secret threads where users traded advice and anecdotes about salaries, workplace norms, and tough negotiations. On one memorable occasion, a user posted that the company she co-founded had been acquired by Google, yet Google had hired everyone but her. She then noted that she was the only woman of five co-founders. The secret went viral and eventually reached journalist Kevin Roose, who managed to get in touch with the author. He confirmed her identity, then wrote up her story in New York Magazine. As Roose explained at the time:
The Secret thread quickly became the talk of Silicon Valley. It was seeming proof that even within the happy-go-lucky world of tech start-ups, there are winners and losers, and more often than not, the losers in situations like these are the designers, who are more likely to be female than their engineer counterparts, and whose “soft” skills are seen as less valuable than coding chops. … Making offers to four-fifths of a company as part of an acqui-hire, while legal, is nearly unheard of in Silicon Valley, where mergers and acquisitions are still generally governed by a certain type of decorum.
Just in case this isn’t making Secret’s value clear, here is a final shot from my collection:
Here, one user seems to need advice about male VCs who trick female founders into going on dates. In response, another user offers thoughts on how to avoid such men, then goes the extra mile and offers more advice in private (over an external anonymous messaging platform called Anonyfish). This demonstrates how Secret formed an important safe space, even while being laden with ridiculous comments (like “Me”).
In short: Secret enabled users with a sufficiently rich network to ask a bunch of their social connections for help, quickly and in relative anonymity. Of course that made it useful for scattered groups handling stigmatized, socially complex problems — such as female founders attempting to raise money in a pressure-cooker environment where they are marginalized, they are never supposed to complain, and the occasional powerful man will want something quite different.
To be clear, Secret was not always fun. Scrolling through the feed was sometimes stressful, and I occasionally uninstalled the app and took a break after reading things that left a bad taste in my mouth. But commentators who discount Secret’s rise and fall with buzzwords like “cyberbullying” and “novelty” are doing the platform a huge disservice.
So what really happened to Secret? No one knows for sure, but many users have observed that the company changed the product frequently and drastically. A number of Silicon Valley insiders have offered thoughts about the moments they themselves stopped using Secret, which often coincided with big product updates and feature changes. One person used the phrase “death of a thousand cuts” as a metaphor for how Secret “revised itself into obscurity.” (Full disclosure: I have done contract work for Quibb, the media platform hosting the conversation I linked above.)
For instance, in response to concerns about cyberbullying, Secret removed users’ ability to post custom photos — a popular feature that they later added back. That was a big change in itself. Yet in the same product update, Secret introduced a completely new polling feature, which proved unpopular and was quickly removed. (Interestingly, Facebook and LinkedIn have both developed and then removed polling features, which suggests that social media polls seem like a great idea in theory but are hard in practice.)
Shortly before the startup folded, Secret re-hauled its product to be more like the anonymous chat app YikYak, which is currently popular with college students. The changes were fundamental, and some cite them the final nail in Secret’s coffin. As the tech publication Re/code put it, Secret became “unrecognizable” — it “looked like a lot of another anonymous chat apps.” Around the same time, Secret’s co-founder and Chief Product Officer left the company.
As a general rule, useful product changes can be unpopular at first, but Secret made so many unpopular changes that it’s hard to imagine the process behind them. How were those decisions made? Were they grounded in careful user research? From outside the company, it doesn’t seem that way. After creating such a popular product, Secret could have slowed down, and they could have carefully gauged each change separately. Instead, they seemed to overreact to media criticism and successful competitors. They moved too fast, and they broke too many things.
The lesson of Secret is not to stop building anonymity-focused media tools — again, Whisper and YikYak are doing very well. Nor should we conclude that the product failed because it was a fad and everyone got bored.
Instead, the lesson is do thorough and continuous user research while developing a product. As a product begins to succeed, the team ought to be cautious in pivots, ensuring that they consistently respect and express the central value that users find in their product. Identifying a famous product’s true value among the noise of pundits, detractors, funders, and superfans can be challenging — but it’s crucial to a company’s long-term success.
Editor’s note: The author would like to thank Alex Kawas for his feedback on the first draft of this piece.