Memo to VCs: Not every site with gargantuan traffic is destined to be a billion dollar public company
And you thought Dick Costolo had a rough time trying to build a “free speech first” media company into a super-fast growing ad business.
Ellen Pao is entering her second week of getting the absolute shit kicked out of her at Reddit.
For those who haven’t been following the story, the Christian Science monitor has a good TL;DR here…
The online community Reddit revolted over the weekend, following the dismissal of an employee who was particularly important to users, reports say. By Monday morning, a petition for the removal of controversial interim CEO Ellen Pao had garnered more than 160,000 signatures.
Victoria Taylor, Reddit’s director of talent and communications, was let go Thursday, and as a result, hundreds of “subreddits” went dark as moderators shut down their threads in protest.
Yesterday, Pao apologized to the community… Again.
We screwed up. Not just on July 2, but also over the past several years. We haven’t communicated well, and we have surprised moderators and the community with big changes. We have apologized and made promises to you, the moderators and the community, over many years, but time and again, we haven’t delivered on them. When you’ve had feedback or requests, we haven’t always been responsive. The mods and the community have lost trust in me and in us, the administrators of reddit.
Today, we acknowledge this long history of mistakes. We are grateful for all you do for reddit, and the buck stops with me.
Pao’s position had already been near-impossible: Many Reddit power users had apparently already decided they didn’t like her, for reasons largely unrelated to her actual performance in the job. Just look at the horrifying barrage of Reddit threads in recent days describing her as a Chinese communist, a “Nazi,” and a “cunt.”
Beyond that, even non-Redditors struggle to reconcile the idea of Ellen Pao as CEO of the famously cesspool-y Reddit with the woman who brought a sexual discrimination suit against Kleiner. It takes one hell of a commitment to free speech to file a suit against the “micro-indignities” of women working in the tech industry and then go back to running a site filled with creepshots and naked celebrity phone hacks.
But even if Pao had the thickest skin in the world and was more committed to unbridled free speech than anyone, her biggest problem was that she lost the support of Reddit’s core community, who accuse her of not knowing how to use the site correctly and fundamentally misunderstanding what motivates its users.
And that’s why she had to issue such a grovelling apology. Reddit might have millions of users but, without the work of a relatively small number of moderators and super-fans, the site simply wouldn’t function. It’s hard to argue the same thing about Ellen Pao, or any other Reddit exec.
I am not a Reddit power user. Or even a moderate Reddit user. What I do know is this: Reddit is a weird beast. A weirder beast than Twitter, and that’s saying something. It should have been left the fuck alone.
Reddit was a crazy comeback story which directly benefited from the implosion of Digg as the latter sought to become a mainstream advertising business. Digg always struggled with the dilemma of whether or not it should cave to a vocal minority or not. I’ve argued in the past that they did it too much, while Mark Zuckerberg early on set a precedent that he wasn’t gonna cave to user criticism of the News feed just because some people found it “creepy.”
Reddit doesn’t have the luxury of making any of those decisions on behalf of users. You can’t change communities midstream because you now want to build a scalable ad business with no hate speech.
The problem is not how you scale community generally. It’s how you scale your community. And some just can’t be “scaled.” Sure, they might scale in how they grow users. But “scaling” in the venture sense of growing revenue and valuation is another matter.
Pao admits the users have “lost trust in [her.]” That hasn’t been helped today by her total reversal in policy, even restoring shuttered sub-Reddits filled with appalling content. Anyone with a toddler knows how well giving in goes the next time there’s a tantrum.
That loss of trust -- and of authority -- is irreparably damaging for Pao in a way it isn’t at most startups. When executives lose trust in a CEO at any other startup, that CEO is replaced by the board or sometimes by investors. It’s the one socially acceptable time for VCs to turn anti-founder. In the case of Reddit, it’s the users who have to be thought of as management. They are the ones who made Reddit such a valuable property in the years before it took $50 million in venture capital. They are the ones who made it such an alluring bet-- even given the obvious issues with content that no advertiser would ever want to be put next to.
When a CEO cannot hire and fire employees as she sees fit; when she cannot control whether parts of the site are up or not; when she doesn’t know how to communicate with users because her posts keep getting voted down; when the only thing to stem a weeklong outcry is a no-holds-barred apology, after her first apology wasn’t enough... I just don’t see how she comes back from this.
In these situations, the community is either king or it’s a subordinate. At Facebook, it’s a subordinate. Boycotts come and go in the Facebook community, but the company rarely changes its policy. At most it may tweak it, as it has with the movement from “real” names to “authentic” names.
In the last week it’s been made clear that at Reddit the community is king. And today, that memo was explicitly delivered by Pao. Stakeholders, management, advertisers, and anyone else looking to get something out of the very unique thing that Reddit is needs to acknowledge that, decide if they are cool with it, and act accordingly. Because Reddit isn’t changing. Management and investors will have to change to accommodate Reddit. And even now, Pao can’t start to do damage control because she still doesn’t get the company-- by her own admission.
Pao wrote in her apology:
I know these are just words, and it may be hard for you to believe us. I don’t have all the answers, and it will take time for us to deliver concrete results… “I mean it when I say we screwed up, and we want to have a meaningful ongoing discussion. I know we’ve drifted out of touch with the community as we’ve grown and added more people, and we want to connect more.
This would have been a great spiritual quest before she took the job. Now it’s simply too late.
Here’s the thing: And it’s a rare time I’ll defend a lot of the vileness on Reddit. Reddit shouldn’t change. Reddit was doing just fine before a bunch of VCs decided to invest $50 million into it to make it “more commercial.” Simply put: Not everything large on the Internet should be commercial.
We see this over and over again, so let’s look at a few examples: Craigslist, Wikipedia, BitTorrent, and CouchSurfing. The first two had leaders who knew they didn’t want to raise money, become a large company, become more commercial, or in any way capitalize on their size for financial gain. Wall Street personalities, VCs-- and in the case of Craigslist, arch-rivals over at eBay-- have gone apoplectic trying to figure out how companies could leave so much cash on the table. And while parts of Craigslist have certainly been disintermediated by rental sites, ride sharing sites, job boards, and other verticals, Craigslist is still around and a core utility for a lot of cities.
Ditto Wikipedia. Media mogul after media mogul has written blog post after blog post about why Jimmy Wales should take advertising money instead of pleading for donations on every page. He hasn’t. He’s surely left capital on the table. But Wikipedia has continued to grow and for the most part, hasn’t gotten replaced.
Craigslist looks like a place where design and functionality go to die. And Wikipedia looks like it’s been in a time capsule. But both websites are clearly functional cornerstones of the Web, and both are still around.
Compare that to -- admittedly smaller -- community phenomenons that did raise money. BitTorrent raised $40 million in the early days of Web 2.0, off the back of YouTube mania.
Here’s what my then-colleague Heather Green at BusinessWeek wrote about the move:
The idea? Gather movies, music, games, and software from Hollywood studios and indie producers alike and distribute them to consumers. BitTorrent would generate revenue either by selling ads within the content or charging a fee for the files.
That's a pretty dramatic turn for a technology that up to now has been so closely associated with illegal file sharing. BitTorrent was developed and released in 2002 by independent programmer Bram Cohen as a way to efficiently distribute the free Linux operating system that competes with Microsoft's (MSFT) software. But it was quickly commandeered by file sharers for illegal swapping of copyrighted movies and TV shows.
That strategy pretty much came and went along with rounds of senior execs. The last round was in 2008. It’s hard to argue the company is better for the dilution, distraction, and detour.
See also: Couchsurfing. It made a huge shift from non-profit to “startup” when it raised $22 million from six VCs including a lead from Matt Cohler at Benchmark. Couchsurfing never had the heft of Reddit, and when its non-profit status was called into question it had to do something. So it’s not a perfect analogy. But it certainly had a strong community that was eroded by the change. CEOs came and went and the company had many missteps.
From an indepth GigaOm piece by Pando alum Carmel D’Amicis:
Before and after its venture funding, Couchsurfing cycled through CEOs, acquiring and discarding them like ill-fitting t-shirts. Each one tried hammering the organization into some semblance of professionalism, efficiency, and money-making and each encountered intense push back from the community. Couchsurfing was founded on the ethos of cooperation, not capitalism, and its most involved users were intensely suspicious of the ulterior motives of the service’s overlords.
“Imagine if you’ve been contributing to Wikipedia for years and one day the founders say they are selling it for a large personal profit but you’re still free to use it. Yup, it’s like that,” one former Couchsurfing ambassador explained on Medium in 2013.
Users made meme videos poking fun at the corruption of the organization’s leaders and published cartoons to represent them. The community that had once volunteered hours to run Couchsurfing could not bring itself to trust leaders overseen by a venture capital firm.
Sound familiar? It went from bad to worse. In 2013, latest Couchsurfing CEO Tony Espinoza was out, and 40% of the staff was cut. The company hasn’t raised money since 2012.
We saw the same phenomena over and over again in the early 2000s as open source mania swept the Web, and VCs scoured European countries for popular open source apps that could be “startup-ified.” It mostly didn’t work. The only $1 billion exit that came from that entire frenzy of deal making was MySQL.
The common denominator in these examples -- both the holdouts who didn’t go the mega-growth route and those who tried and stumbled-- is a popular product that was content to be beloved and sustainable… until a startup version of it appeared doing something similar but with a total Silicon Valley mega-growth mindset.
BitTorrent got dollars in its eyes when it saw YouTube. Couchsurfing did when it saw Airbnb. Craigslist refused, even when eBay tried to take it on with Kijiji and appropriated Craigslist trade secrets to bolster it. It was called crazy by everyone with a capitalist pulse and a sense of design.
In the case of Reddit, it was the surge of free-speech, news-driven, user-generated content which fueled everything from Facebook to Twitter. There’s been a belief for the last 15 years of the Web that eyeballs represent dollars, and if you can route around millions of eyeballs then you’re a gatekeeper of the dollars. Surely that’s worth even more dollars right?
The very core of Reddit’s unpredictable resurgence and staying power is in that bizarro community of volunteers and moderators that keep a 170-million-monthly-uniques site running despite a paid staff of just 80 employees.
The reason this kind of thing keeps happening is no one charged with making the decision is incentivized to do otherwise.
For the VCs, it was well worth the gamble. It’s their job to risk millions on ideas or nascent properties that may never catch fire and develop a passionate engaged community of more than 100 million people. So funding an existing massive property where the “only” risk is making it into a company? Easy bet.
For the founder, who frequently gets to cash out some shares in these situations? It may be a thorny bet in terms of legacy and emotion to take money that ultimately destroys a valuable asset, but, in terms of personal security, at least founders can take some money off the table and enjoy some spoils of their hard work.
The trouble is it’s not such an easy bet for employees and even less so for a site’s loyal community, who view their labor as more than a simple financial risk/reward. Unlike a VC, they can’t just write this off as an experiment if it destroys a beloved asset. And unlike the board and founder, they don’t get a vote.