Dropbox and Uber: Worth Billions, But Still Inches From Disaster

Dropbox went dark over the weekend.

According to the company, the widespread outage was the result of a bug it introduced while updating the hundreds of computer servers that drive its massively popular file-sharing service. But the problem was bigger than that. The San Francisco-based startup not only faced countless complaints from users across the net, it was forced to deflect rumors that the service was hacked, something that turned out to be a hoax.

On one level, a dust-up like this is just part of life as a startup. Things go wrong, people get upset, problems are solved, lessons are learned. But the stakes are higher when you're Dropbox -- or any other tech startup that has ascended to the misty heights of the billion-dollar club. This weekend's Dropbox outage, along with recent problems for Uber and Snapchat, show just how close such companies skate to complete disaster -- not because of anything they necessarily did wrong, but because of the very nature of their businesses.

In those tender days between two-scrappy-founders-in-an-apartment and established business, these burgeoning outfits have hundreds of millions of dollars invested in their future, and that future is far from certain. In an age when people can so easily abandon one web service for another, a single screw-up is all it can take to bring things crashing down for good. And the best of these companies know it.

The most successful tech giants -- think Google and Facebook -- have been able to insulate themselves from the big SNAFU by performing well for long enough that we become inescapably dependent on them. For many of us, Gmail would have to delete our entire accounts before switching became even plausible anymore. But even for billion-dollar companies still in a period of massive growth, such cushions aren't always there to catch them. If they fall, the landing could still be hard.

Do One Thing, Do It Best

In an interview with WIRED this past fall, Dropbox co-founder Drew Houston acknowledged that his company has almost no margin for error. If Dropbox accidentally destroyed just one person's file, he said, it could erode the trust of all its users. "This is like the same sort of genre of problem as the code that you use to fly an airplane. Even if it's a little bug, it's a big problem."

The risk for Dropbox is that at its core, it essentially does only one thing: It syncs your files across all your devices. On the one hand, this single-mindedness has brought Dropbox its tremendous success. Founded in 2007, the company concentrates on doing thing and doing it well. But that strength is also its greatest vulnerability -- Dropbox is not diversified. Many other companies large and small now offer much the same service. If Dropbox loses your trust by messing up the one thing you thought it did best, you could easily switch your allegiance to another company.

In my several years using the service, I have never had a file lost or corrupted. Its tool for uploading photos from mobile devices is a breeze. And like the best-designed products, it works so effectively that it fades into the background.

This weekend, it didn't work. But files weren't lost or destroyed, the company says. It's just that people couldn't reach them. "Your files were never at risk during the outage," Dropbox engineer Akhil Gupta wrote. The databases affected, he said, "do not contain file data."

Rather than diversifying, Houston has worked to hire some of the smartest talent in tech, including the inventor of the Python programming language, to find the best answers and buffer against problems like this weekend's outage. So far, that seems to have paid off.

This weekend's blip won't put many people off Dropbox. If they're like me, users have come to rely heavily on Dropbox for quickly storing and sharing files and generally getting work done. And Dropbox has created a strong well of trust from which it can draw. If it had branched into other services at the expense of its core syncing service, you can bet that trust wouldn't be there.

How Uber Goes Under

Uber, the ride-sharing startup, is facing its own moment of crisis. Over the past months, many people -- and many news stories -- have complained about the company's "surge pricing," where it raises fares during times when lots of people want a ride, such as during snowstorms and over holidays. Now, a different kind of anger has surfaced: Paris-based blog Rude Baguette reports that, in France, protestors are attacking Uber cars, throwing eggs, slashing tires, and breaking windows. Uber confirmed the attacks.

The violence comes as the French government attempts to address complaints that app-based car services like Uber are undermining the traditional taxi industry. The U.S. taxi industry feels similar ill-will toward Uber for undermining its business model, but it has turned to the courts and city councils to protect its interests.

Surge pricing and conflict with taxi services might seem like separate issues. But both reflect the consequences of Uber's choice to stake its success, like Dropbox, to an unbending vision of doing one thing exceptionally well. It too is not diversified. Uber could make the recent complaints go away fairly quickly. It could drop surge pricing. And it could acquiesce and change its service in cities where government and industry have come out against it. But Uber doesn't do either of these things, because in the eyes of its outspoken CEO Travis Kalanick, backing down would compromise the foundation of his business: to provide a great ride.

Surge pricing, according to Uber, is intended to stimulate supply and curb demand to ensure the two match. Otherwise, the logic goes, would-be riders are left stranded without a car. Last month, during the height of the backlash against Uber over fares reported at seven times the usual during a New York snowstorm, Kalanick told WIRED that the bad publicity his company faced over surge pricing would pale compared to the impact of Uber not being able to offer a ride at all.

"If you are unreliable, customers just disappear," he said. "The thing is that nowhere in any of the press are you hearing about us being unreliable."

If rides don't come through, or are slow to arrive, the reason for Uber's existence disappears. Kalanick is a professed admirer of Amazon founder and CEO Jeff Bezos, an Uber investor. Amazon now does a lot of things well, but if it stopped delivering the products people ordered from the site quickly and accurately, it would fold. Uber sees itself as offering a similar level of service for rides. If it backed down, if the rides stopped showing up, in Uber's eyes, it would be like the Amazon box you ordered not arriving on your doorstep.

This is also why Uber believes it can't compromise once it begins to offer its services in a new city, or a new country, like France. It strives to work with regulators to accommodate its existing service rather than changing how it works. As a company, Uber is as much a designer of algorithms as a provider of rides. On the streets, Uber guides cars to certain places at certain times. Back at its San Francisco headquarters, teams of mathematicians and data scientists are figuring out how to guide them. And the math is hard enough without additional constraints.

Yes, the constraints still come. A new French law that requires drivers to wait 15 minutes between taking a reservation and picking up a passenger. But the company has shown a defiant reluctance to put constraints on itself.

Such stubbornness is often seen as arrogance: the hotshot, elitist startup that believes it's above the rules. But Uber has made the choice that getting bashed on Twitter -- or by City Hall -- isn't as bad as customers opening up the app and seeing no rides on the map. The first threat is manageable. The second is existential -- customers just open up another ride-sharing app to see if an Uber competitor has cars instead.

Snapchat Disappears

Snapchat's moment of weakness was more acute, and the company appeared to deal with it the worst. Recently, a group of hackers exploited a known security hole in the private messaging service, leaking the phone numbers of millions of users. Snapchat co-founder and CEO Evan Spiegel faced a barrage of withering criticism for failing to apologize quickly for the incident and the sometimes peevish posture the company has taken in the wake of the leak.

For a company whose whole business is based on privacy, such a breach is a serious threat to its livelihood -- not least because so many others are now angling to offer similar services.

The cautionary tale here is Friendster. As the startup that was Facebook before Facebook, Friendster was the one of the first companies to gain genuine traction in what was then widely called "Web 2.0." But then it stopped working the way people wanted it to. It didn't even screw up that badly -- accounts of its demise describe fairly typical management and technical issues. But even this can bring down a startup in a digital world that moves so quickly. Facebook arrived and also did the one thing Friendster did -- connecting people -- but better. And for Friendster, it was too late.

For Snapchat, a meaningful apology isn't about good manners. It's about showing it appreciates the gravity of its violation of trust. As is making sure it doesn't happen again. As with Dropbox and Uber, Snapchat has earned the love of its users by doing one thing they love really well. Take that thing away, and the love goes with it.