Inside the company that rebuilt Digg

But then Digg's fortunes turned south. The recession, management problems, and a botched software upgrade all contributed to Digg's downfall. By 2012, things had gotten so bad that the digg.com domain name was sold off to a company called Betaworks for $500,000. Digg, it seemed, was dead.

But Betaworks has managed to reverse Digg's seemingly terminal slide. Digg's new CEO says the site has grown from 1.5 million users to 8 million over the last two years, and he expects more growth in the coming years.

Digg is just the latest success story for Betaworks, which calls itself a "startup studio," a reference to the Hollywood studios of the 1920s. Like an old-school movie studio, Betaworks keeps a roster of talent on contract to work on Betaworks sites. The result has been a string of successful websites and apps, including Tweetdeck, Bitly, and Chartbeat.

The story of Digg's comeback — and the Betaworks strategy overall — tells us a lot about how the web has matured over the last decade. Starting a mass-market website used to be a risky, expensive, and error-prone process. But new technologies and business models have taken much of the guesswork out of website development, making it possible for the Betaworks team to reinvent Digg in a matter of weeks.

Digg this

In the 1990s, Silicon Valley was obsessed with "portals": websites like Yahoo, Excite, and Lycos that were curated by a combination of human editors and algorithms. Founding Digg, Rose envisioned a different kind of website that put users themselves in charge. In an era before the web was dominated by Facebook and Twitter, that was a radical idea.

Digg quickly attracted loyal users who believed in Rose's vision of a user-controlled news site. They scoured the web for interesting reads. If enough other users pushed the "digg" button on a story, it would get promoted to the Digg front page, where it would enjoy a flood of traffic. Publishers across the web recognized this opportunity and began installing Digg buttons on their own sites, helping to boost Digg's profile.

(BusinessWeek)

Popular topics included gadgets, politics, and video games, but there was always an eclectic mix. On one typical day in the late 2000s, the front page included links to articles about the health benefits of red wine, China’s emerging environmental movement, and the International Olympic Committee's announcement of finalists to host the 2016 Olympics.

By 2007, Digg had become one of the most popular ways to find technology news. But Rose wanted his site to be larger still, becoming everyone's favorite way to find interesting things to read online.

That proved challenging. For one thing, Digg was becoming a prisoner of its own opinionated user base. Digg needed its army of active users to help it find good content, but that user base could also become an impediment to broadening the site's appeal. There were rumors that active Digg users were colluding to control the content that appeared on the home page.

But when Digg management tweaked the voting system to discourage coordinated voting and promote a broader range of stories, it alienated some Digg veterans. The company never figured out how to balance these competing priorities.

"The brand name had gotten toxic"

Digg's leadership consistently overestimated the site's growth potential. For much of 2006 and 2007, rumors swirled of impending acquisitions. But Digg reportedly demanded more money than potential suitors were willing to pay. In 2008, Digg came within days of being acquired by Google for $200 million before the search giant walked away from the deal.

That proved to be Digg's high-water mark. When the recession hit in 2009, ad revenue plunged, forcing the company to lay off 7 workers.

Things got much worse in 2010. Conflict between Rose and Digg CEO Jay Adelson led to Adelson's ouster in April. Then, in August, Digg rolled out a new version of its website that was so bad that users fled in droves to a competitor, Reddit. Two rounds of layoffs, in May and October, slashed the company's headcount from 72 to 42.

After the software upgrade, "there was a lot more emphasis on the sites that were connected to Digg, and less about user-submitted content," says Alex Taylor, a longtime Digg user who lives in Toronto. "A lot of us checked out Reddit at that time."

The company never really recovered from that horrible year. As Digg's regulars deserted the site — and stopped performing their essential curatorial function — the quality of the content on the front page declined. In 2012, Digg's board threw in the towel, putting the company up for sale at a fraction of its value four years earlier.

Digg sold its patents — including one on the concept behind the Digg button — to LinkedIn for $4 million. The Washington Post paid $12 million for the Digg engineering team in a bid to beef up its social media technology. (Rob Malda, founder of Digg competitor Slashdot, joined the Post around the same time).

But shortly before the deal closed, the Washington Post decided it didn't want the Digg website itself. "The brand name had gotten toxic," says Andrew McLaughlin, the man who became Digg's new CEO. "It was seen as a failed site with a lot of scummy behavior, a lot of gaming of the Digg button. The Post didn't want it."

One company that did want the name was Betaworks, a New York firm in which McLaughlin is a partner. The company, which has given birth to hits such as Tweetdeck, Chartbeat, and Bitly, offered $500,000 for the Digg name and website. Digg's investors might have gotten more money if they'd auctioned the domain off to a content mill. However, McLaughlin says, "we made an emotional appeal that we would build something cool that they would be proud of."

A race against the clock

The Digg board accepted the Betaworks offer in mid-June 2012. That meant Betaworks had to scramble. Digg's servers — set up before cloud computing slashed the cost of running a website — cost around $250,000 per month to run. The old company was planning to shut them down on August 1.

So Betaworks had about six weeks to create a new Digg. The company already had an existing product, called News.me, that it had been developing with the New York Times since 2010. Engineers and designers from that project could be tapped to build the new Digg.

The company decided to ditch the user voting system that had been the hallmark of the original Digg. The Digg user community had become too acrimonious — and the front page too spammy — to be salvaged. Instead, the new Digg would use a combination of software and human editors to decide what would go on the home page.

In some ways, this marked a return to the "portal" vision that the original Digg had sought to replace. But this was a portal with a lot more intelligence under the hood. Betaworks has developed a sophisticated set of software tools to scour the web for interesting content and then use a variety of signals — including attention on social media sites like Twitter — to judge which articles are most interesting. The software produces a feed of promising candidates that are then curated by human editors.

Digg editor David Weiner chats with managing editor Anna Dubenko. (Digg)

And in another way, Digg is the anti-portal. Portals were created in an era where it was hard to find interesting stuff to read online. They aimed to cram as much useful content as possible on one page. In contrast, the Digg team prizes simplicity.

"The internet is too noisy," McLaughlin said. "The typical emotional state of an internet or mobile phone person these days is to feel overwhelmed. There's too many stories, too many blog posts, too many status updates, too many tweets. Too much demanding your attention."

In contrast, he said, "the theory of Digg is to build something that's very clean, very pristine." Digg's goal is to "distill this noisy, cacophonous internet down to a much smaller set of things, almost all of which you find interesting."

Digg's comeback

"We didn't even realize we needed editors until a couple of weeks into our redesign," said Digg design director Justin Van Slembrouck. "We had this prototype of a general news site that was all data-driven. It was pretty good, but it was a little bit behind. It added up to something that was boring. That's when we realized we needed an editor."

Betaworks hired David Weiner, a journalist who was an early employee of the Huffington Post, as Digg's editor. Today he manages a small team that — with a lot of help from software — decides what will appear on the site.

Weiner knew he faced a big challenge salvaging Digg's tarnished brand. By 2012, Digg had suffered from years of bad press. In many peoples' minds, the sale of digg.com for a paltry $500,000 was just the final nail in the coffin.

"The attitude that I wanted to get across from the beginning was that of 'we're in on the joke. We have a sense of humility. We can laugh at ourselves,'" Weiner said. "We were immensely proud of what we'd built, but we had to take some of the punishment for things we had not done."

For example, during the new Digg's first month, President Obama did an "ask me anything" post on Reddit. Obviously, Digg couldn't compete with that kind of star power. So it responded with humor instead:

We asked Dukakis but he turned us down... :(

— Digg (@digg) August 29, 2012

Users appreciated this self-deprecating approach. They also appreciated the eclectic fare Weiner and his team began surfacing at digg.com. Slowly, the site began to regain user trust, and traffic grew.

Alex Taylor is one Digg veteran who has warmed to the site's new incarnation. "They've won me back in terms of content," she said. "It's not really an interactive community anymore. I like it the way that it is, but it's not the same."

Taylor attributes the success of the new Digg to a growing amount of content on independent blogging sites such as Medium. Digg helps to surface this content, which users might not otherwise see. Digg, she said, "appeals to me more because it's not quite so heavy-handed with mainstream media."

According to McLaughlin, Digg has grown from 1.5 million users per month to 8 million. That's nowhere near as large as Reddit, which was once smaller than Digg but now serves 115 million monthly visitors. But it's a surprising comeback for a site that had been abandoned two years ago. McLaughlin said Digg has a realistic plan to reach profitability, though he wouldn't predict how soon the site would be in the black.

Digg CEO Andrew McLaughlin. (Digg)

He argued that there's lots of room for further growth. He believes Digg's software tools and minimalist aesthetic will appeal to users in other countries, so international expansion is in the cards. There have been experiments with original content. And Digg hopes to "revitalize the Digg button," in McLaughlin's words, by making it a way for Digg users to signal to their friends that they like a particular article.

Last year, when Google announced it was shutting down its popular RSS reader, Digg saw an opportunity. The team quickly created an RSS reader called Digg Reader that picked up where Google had left off. Creating Digg Reader was made easier by the fact that the company already had software to scrape thousands of RSS feeds to help Digg's editors find good content.

McLaughlin said Reader now has more than a million users. And they're very active, he said: "At any given time on a typical weekday, about 20 percent are in Digg Reader."

Inside the startup studio

Digg's parent company, Betaworks, defies easy classification. Like a conventional startup incubator, Betaworks seeks to build companies that can eventually become independent brands. But it takes an unorthodox approach.

In a traditional incubator, each startup is run by independent founders who work primarily for equity in their own companies. If a startup fails, its founder-employees are out of a job. In contrast, the people running new Betaworks companies are salaried employees of Betaworks. They move from company to company as needed.

If a Betaworks company fails, the designers, developers, and other talent behind the company are re-assigned to other projects. If a Betaworks company grows large enough to leave the nest, some employees go with the new company but others will return to the mothership to work on another project.

McLaughlin describes Betaworks as a "startup studio," an analogy to the studios that dominated the movie industry during Hollywood's early years. These studios had directors, writers, and actors under contract who would be moved from project to project as the need arose. Likewise, Betaworks has a stable of designers, developers, and project managers that can be tapped to help with new opportunities as they come along.

This model seems to have helped Betaworks churn out a series of modest hits. It owns Giphy, a search engine for animated GIFs, and the URL shortener Bitly. It owns Chartbeat and SocialFlow, which are tools popular with web-based publishers like Vox. Last year it acquired Instapaper, a popular app for saving content to read later.

What the model hasn't done is produced a multibillion-dollar company like Google or Facebook. One of the company's biggest hits was Tweetdeck, a Twitter app that was sold to Twitter itself for $40 million in 2011. That's obviously not a trivial amount of money, but it's modest as Silicon Valley acquisitions go.

Instead of making a lot of long-shot bets in hopes of getting the occasional home run, Betaworks has figured out how to generate a consistent stream of singles and doubles: solid, profitable companies that aren't transforming whole industries but are helping millions of people and making a tidy profit in the process.

This model is aided by the falling cost of computing resources. Remember, the original Digg cost $250,000 per month to run. Digg needed to raise significant capital to cover the costs as the site grew. And the company felt pressure to generate a lot of revenue to recoup the value of that investment.

Modern cloud computing services make web hosting a lot cheaper and a lot more flexible. The new Digg costs tens, not hundreds, of thousands of dollars per month to run. And other things have gotten easier, too. For example, when Digg was founded, there was still a lot of uncertainty about the viability of advertising-supported websites. Today, the tools for monetizing popular websites are a lot more sophisticated.

The combination of lower costs, more powerful tools, and greater certainty about business models means that Betaworks can afford to make a lot of bets on sites that have the potential to be small-scale successes. It can build a search engine for GIFs or a site that generates personalized weather reports and count it as a success even if it's never going to be worth billions of dollars.

Web innovation is shifting

The Betaworks analogy between the web and the movie industry is apt. At the dawn of the 20th century, most of the excitement came from technological and business innovations — people were still figuring out what moving pictures could do and how to make money selling them. But by the 1930s, the feature-length motion picture was well-established. Innovation became more artistic than technological — people had figured out how to capture and distribute moving images, but they were still learning how to use that medium to tell stories effectively.

Today the web is going through a similar transition. The technology and business model of the web has become well understood. Anyone can use cloud computing services and ad networks to create websites and make money from them. The hard part today is figuring out how to use these tools to create websites people want to read. That's about good design and good editorial decisions more than about technology.

Betaworks is positioned to take advantage of this shift. It's not likely to produce the next Facebook or Google. But it has assembled web talent with a knack for creating websites that make our lives a little bit more comfortable, productive, or interesting.