Bandwidth Caps Not Evil If Done Right

According to a number of readers, in my post on Tuesday I gave the impression that AT&T; and the cablecos are big, evil corporations for having begun implementing bandwidth caps. But that's not necessarily the case.

My overall stance on caps or metered bandwidth is that they're simply a natural evolution of the broadband business model. In fact, to some degree I see them as a good thing as they're indicative of the fact that at least some people are consuming a ton of bandwidth and therefore I'm assuming are more reliant on broadband than ever.

For many, though, caps are a sign of not enough bandwidth and that the solution isn't metering but deploying more capacity. But there's a problem with that thinking: Private enterprise as a rule of thumb only builds enough supply to meet demand, and the percentage of people using a ton of bandwidth today is still small. It's hard to build 100Mbps to everyone when only 1% of your users need or even want it.

But this post isn't about deployment of big bandwidth, it's about the implementation of bandwidth caps, and the most important thing to remember about all this is that by and large the caps that have been implemented to date have been trials. Companies like AT&T; will admit that they don't have everything figured out for how this is going to work; that's what these trials are for.

While things could always be better, I'm reasonably impressed with some of the accommodations that are already being made.

Take AT&T; for example. The big thing they're doing is protecting customers from getting hit right away from these changes. They're letting all impacted customers know about the trial specifics at least 60 days ahead of any overage charges. They're waiving any charges the first time a customer exceeds their cap. They're sending them notifications any time they reach 80% of their cap. They're providing an online tool for customers to track their usage. They're also beginning to educate their customers as to what kinds of usage will add up to exceeding these caps.

And in talking with them I get the sense that there are a lot of other possibilities on the table for consideration as to how they can further develop and evolve these new business models. For example, not counting bits transferred during off-peak times like at night the same as those sent and received during peak daytime hours.

From their perspective what this is driven by is preserving the user experience for all their customers. Bandwidth caps are a way of maintaining fairness in the use of a shared resource, which all broadband networks are to one degree or another, so that the heaviest users don't negatively impact average users.

That said, as I argued earlier this week, this may be a challenging argument to get across without seeming like they're covering for their networks not having enough capacity, especially with Verizon seemingly unaffected by the same challenges due to their investment in FiOS. But in the end I do believe that what's primarily driving this are the interests of consumers.

At the same time, I can't deny that it's troubling when I read that the cost of delivering additional gigabytes is a fraction of the $1 per gigabyte they're charging for overage fees. In my mind it suggests one of three things: there are additional costs that I and others aren't aware of, the price is set artificially high in order to dissuade the heaviest of users from overconsumption, or these caps are intended to be a new profit center.

Yet regardless of the intent this is an area where it'll be interesting to see how (or if) the market works. As more and more broadband providers adopt a metered model across their footprints eventually the cost of these additional gigabytes will start having an impact on consumer decision making. At that point, assuming there's not any form of collusion going on between providers to keep overage charges high, they should start trending downwards as consumers pick providers with the lowest rates.

At the end of the day the one really good thing that could come out of the introduction of bandwidth caps is the end of the practice of some broadband providers, most commonly cable companies, whereby the heaviest of users were simply cut off from their service with no immediate way of getting it restored, regardless of if they're willing to pay more for it. Even if you don't agree with what the terms and costs are, at least hopefully we can agree that any system of caps is better than no system if it means power users no longer have to live in fear of losing service for exceeding some undefined cap.

Long story short, I think my last post on this subject was guilty of getting overexcited about an area of initiatives that are still in their nascence. We all need to respect that these things take time to shake out, and that we shouldn't assume that what's driving this is the greed of big, evil corporations. There really is an aspect of preserving the user experience to this. And I'm not alone in believing this to be true as respected thought leaders like Gigi Sohn and Tim Wu have said similar things.

This doesn't mean though that we should not be vigilant in monitoring what's happening and vocal in expressing our displeasure when good intentions turn into harmful actions. But let's take a breath and let things pan out before rushing to judgment.