WHEN A TECH company tells you something is unlimited, don’t believe ’em.
Last week Microsoft nixed the unlimited storage option from its OneDrive service. Meanwhile, Comcast started billing users extra in some cities if they gobble more than 300GB of bandwidth per month. Last month Sprint followed the lead of most of its competitors and began throttling download speeds of its “unlimited” data plan for customers who exceed 23GB per month of data usage.
The message is clear: if you want to download or store lots of data, you’re going to have to pay more for it. But why isn’t it possible to offer an unlimited service that’s actually, y’know, unlimited?
Some companies that promise unlimited data storage or bandwidth may simply be making cynical marketing claims they know they can’t back up. But telecommunications industry analyst Jan Dawson argues that in most cases, these companies are usually more naive than cynical.
“Service providers often guess wrong and find themselves losing money either because they underestimated average usage or because a small number of people abuse the system and use way more than the provider anticipated,” he says.
The funny thing is, that abuse can ultimately double as market research. If people want more data than you thought, then you know you’ve got something you can charge more money for.
Setting Limits
Last year Microsoft unveiled its unlimited plan for OneDrive, which cost $6.99 a month. This week it nixed that plan entirely, placing a limit of 1 TB for $9.99 a month. According to a Microsoft blog post, some users began backing up multiple computers and storing DVR recordings, a usecase Microsoft didn’t anticipate. “In some instances, this exceeded 75 TB per user or 14,000 times the average,” Microsoft says.
Setting specific limits also makes it easier for companies to predict costs. Even when companies have plenty of capacity to meet demand, usage patterns can change rapidly. Back before 2010, all major mobile carriers offered unlimited data plans. That made good business sense back then when most smart phone users probably only used their phones to check their email and do some light web surfing. But then people started watching crazy amounts of YouTube on their phones. One by one, the major carriers have either stopped selling new unlimited plans or begun throttling speeds for heavy users.
The big question, though, is whether these data-hungry users actually cause companies to lose money and whether they actually adversely effect other users. In some cases, they might actually help companies make more money in the long run.
Are the Outliers Really That Bad?
To state the obvious, there’s no such thing as a hard drive with infinite storage capacity. That means that Dropbox, Microsoft, and the like need to have enough drives on hand to meet their customers’ needs. Apart from drives, a cloud storage provider also needs to have data centers to house all those servers, technicians to keep them running, and enough bandwidth to carry all those bits back and forth between customers’ laptops and the company’s data centers. Even if these companies have plenty of money to pay for these things, it can time to add more capacity if they underestimate their needs, especially if they need to hire new staff or expand a data center.
Likewise, bandwidth largely depends on the quality capacity of the networking hardware. The more data users download and upload, the more networking gear the telcos need. Netflix customers learned that the hard way last year when disputes between the company and network providers like Comcast and AT&T over who should foot the bill for equipment upgrades led to a downgraded experience for some viewers. Wireless providers might even need to build new cell phone towers to keep up with growing demand.
So it’s possible that a small minority of users ate up all of Microsoft’s disk space, leaving it unable to meet the requirements of its other paying users, even though their needs were more modest. And it’s possible that a few bandwidth hogs were slowing down internet speeds for everyone else. That’s what Sprint implied when it announced it last month that it will customers’ traffic if they used more than 23GB of data in a given month. The announcement was headlined “Protecting the 97%.”
But it’s not clear at all that excessive users cost companies all that much money or actually adversely affect other users. Microsoft only claimed a small number of users were using the service for “extreme backup,” and there are plenty of companies that offer more space than OneDrive for less money. Sprint figures that only 3 percent of users are routinely exceeding 23 gigabytes of data. Are these outliers enough to ruin it for everyone?[related-posts]
Maybe not. Comcast, for example, doesn’t claim that heavy users adversely affect other users. In fact, the company has published a paper detailing how it avoids congestion without limiting bandwidth. Rather, the company frames the higher prices for heavier usage as a matter of fairness. Comcast spokesman Charlie Douglas says that about 10 percent of its customers nationwide account for 50 percent of all data usage amongs its customers. “The idea is those who use more pay more,” he says. And, increasingly, they do. In 15 trial cities, customers who use more than 300 gigabytes of data are charged an extra $30 per month. The company also now offers a lower tier service in some cities that includes a limit of only 5 gigabytes per month for about $5 less than they would be paying for “unlimited” service.
Critics of Comcast’s new fees have argued that the company is trying to crack down on users streaming video in order to protect its own television and video streaming services. But 300 gigabytes is enough to stream an awful lot of digital video. What it comes down to is that Comcast and others have recognized that heavy users will generally pay more for their service than average users. A few people using far more than average are just outliers. But if enough people go over the average, that’s a market. “Unlimited” is really just a placeholder for new tiers of service that companies haven’t invented yet.
[Wired]