If you're in the market for wireless data, I'd humbly suggest that Karma is a good deal. It's data that works with any of your devices, never expires, and includes none of the complexity of subscriptions, cancellations, or artificial constructs like "phone lines." Compared to paying $50 a month for an LTE hotspot, Karma Go gets you all the connectivity with a truly pay-as-you-go pricing model. Anecdotally, paying for the data I need from Karma is much cheaper than paying monthly for data I might need from a traditional carrier.

All that said... $14 per GB? Is that really the best we can do?

I assure you, we're not marking this stuff up very much. We make a profit when we sell data, but none of us have yachts. So, what is this "stuff" that we sell, and why is it so pricey?

The cost of transporting data is what makes the cost of wireless data different

In a previous post, I explained what 1GB (one Gigabyte) is. In short, a "Gigabyte" refers to an amount of digital data. Your computer's hard drive probably holds 100s of Gigabytes, your phone might hold 8 or 16GBs. A Gigabyte is enough information to stash 1,000 ebooks, or half an HD movie, or a dozen hours of music. When you pay Karma $14 for 1GB of data, you're basically paying us to convey 1GB of information wirelessly, from the internet to you. The cost of transporting data is what makes the cost of wireless data different from the cost of, say, a computer hard drive.

So what's so hard about transporting data?

I don't want you to think I'm dodging the question or trying to make wireless data sound like magic... but it's basically magic. More precisely, wireless data is the end result of thousands of absurdly smart people designing technology to encode, transmit, and decode a stream of 1s and 0s over invisible radio waves.

In a very simplified model of the world, all those 1s and 0s are floating freely around the internet, available for access by anyone hooked in to the network. What a wireless provider does is buy a high bandwidth pipe to the internet, run that pipe up to a tower, and beam that information to its customers over the specific portions of radio spectrum it owns.

To reflect further on what an accomplishment this is, recall FM radio. Across all of the FM spectrum, there were only about twenty or so "stations" available in a given area. If crowded too close together, those stations would interfere with one another. If too underpowered, or too far away, or behind a hill, a station would be hard to tune in to, static-ey, or fade altogether. Wireless data networks work within the same physics as FM, except imagine a "station" for every one of the thousands (or millions!) of data-hungry wireless devices in an area trying to use the internet.

It all starts with spectrum, the piece of radio frequency real estate a wireless carrier like Verizon or Sprint buys for its exclusive use. A recent FCC auction for a relatively small chunk of spectrum broke $40 billion. The intense demand for wireless data is driving these prices up, and upcoming technologies for faster wireless broadband are even more spectrum hungry.

Wireless carriers use all sorts of amazing technology to fit as many simultaneous connections as they can within a slice of spectrum, but they can't beat physics. If more devices attempt to connect to a tower than that tower can handle, the excess devices will simply be out of luck. This isn't a problem in most areas of the country, but it's been a big issue in places like New York (where I live), where you can have thousands of people packed into a tiny space. You might have a great "signal," but you can't get any internet because there's no "station" reserved for you. Adding more towers doesn't always help, either, because overlapping radio waves in the same frequency interfere.

They can't beat physics

Still, for most parts of the country, putting in more towers, putting towers in good locations, and upgrading existing towers, all serves to improve wireless data for customers. Wireless carriers spend billions of dollars a year upgrading and maintaining their networks. Paying for high quality spots to put their towers, and high speed internet (a huge bottleneck) for that tower to transmit. And if they score big in a wireless auction? That usually means upgrading towers once again to handle the new spectrum.

I'm not an economist, and I can't compute exactly how all these network-building costs result in the exact amount carriers charge for data. $14 per GB still sounds like a lot, but "expensive" doesn't necessarily mean extortionate, especially in an arena as competitive as wireless data in the US — if one of the big four carriers could cut costs in half, they're hugely motivated to pass those savings on to their customers in order to steal subscribers away from their competitors.

The good news / bad news is that the technology keeps marching on. Now that wireless data is nearly as fast, or in some cases faster, than traditional wired internet connections, wireless carriers can compete with home internet providers like Comcast and Time Warner Cable. That means they'll need to add more capacity, which is of course expensive, but it also means they'll need to get costs down so that users can watch Netflix without going broke.

Everyone at Karma wants wireless data to be competitive with wired data, if for no other reason that it's just simpler: why pay a phone bill, a hotspot bill, and a home internet bill when you could simply pay one price for all your data? When wireless data gets cheaper we'll be the first to celebrate, and we'll be more than happy to pass the savings on to you.


About Paul Miller

That guy who left the internet for a year