I’ve said many times that broadband connectivity services have commoditized, and I believe the truth of that is fairly self-evident. I’ve not really gone into the “Why?” of the matter, though, and if it’s true that there’s a fiber war raging, as some tech sites have claimed, then we could benefit in any assessment of the potential of fiber, or 5/6G, by exploring the commoditizing forces to see if anything could overcome them.
Networking these days is really all about Internet access, and the Internet is a network service cooperative, not something a single operator can control. The stuff you “access” can be (rarely) truly free, ad-sponsored, or paid for, but it’s usually provided by parties other than the access operator. In fact, what users want is the access stuff, and broadband is just what provides it. You see your bathroom as the pretty fixtures, not the plastic-pipe-plumbing. So on the surface, you can see why commoditization could be seen as inevitable.
If I’m an operator who wants to change this situation, I have two broad choices. First, I can get into the stuff-you-access business myself. Go, as they say, up the value chain. Second, I can promote features that can somehow improve my access service. Network operators have been reluctant to embrace the first option, probably due in part to historic failures in their attempts, partly because of cultural inertia, and partly because you can’t be a network operator who relies on competitive OTT services to overcome unprofitable base services. It’s door number two they want.
There are really only two differentiating features to broadband access, price and speed. Persistent service problems and poor customer support can drive churn, but even business users will tolerate occasional issues for better price per bit. Lack of addressable differentiators is what drives commoditization.
There are two possible avenues to adding access differentiators. One is to improve one or more of the QoS metrics, and the other to improve security. Both have their own unique challenges, but the issue they share may be the most problematic. That issue? It goes back to the Internet as a cooperative.
Even for a single OTT site, it’s rare that the access ISP controls the entire traffic path. Unless the access provider is the OTT as well (even more rare), the access ISP can’t control the content or application properties. They can’t deliver either security or enhanced QoS end to end, and so if they promise it they render themselves vulnerable, perhaps even legally so. Even if they dodge legal issues, they inherit support burden they can’t carry because users with any problem with the Internet will expect their promising access ISP to solve it. The same is true with security. Can an access ISP guarantee end-to-end security, and since that’s clearly impossible, dare they promise anything?
The unique issues? For QoS, the latency issue has been tried by 5G proponents, but the same end-to-end point applies to it. Can you reduce latency enough in access to be meaningful overall? Hardly. And for security, there are things an access provider could do, in regards to things like denial of service or perhaps protection from dangerous sites, but either of these are probably easier to address fully in other places in the flow. But these do show something, which is that there may be a kind of application that would validate additional service features. The problem is that we don’t have that kind of application today, and getting to it would take the network operator at least some distance beyond simple connectivity.
One example of this is offered in edge applications. Today, enterprises tend to use on-premises edge hosting where the length of the control loop, meaning end-to-end latency, is a critical issue. Edge computing as a service could not only provide some OTT service revenue where there is no major direct competitor, it could also validate low-latency access premium services. Or, perhaps, you forego the hosting and offer a low-latency mesh back to a customer point so that edge computing resources could be pooled for efficiency. Some enterprises have actually looked at this but have not gotten much interest from operators, or have had the operator approach involve something like getting all their industrial IoT elements their own 5G service subscription.
I think you can begin to see my point here. What operators want is not only connection services with premium elements that can justify a significant premium, but elements that can keep operators in their comfortable roles and wait for the progress of glacial-scale standards to boot. This just doesn’t seem likely to me.
Operators have two clear paths to increased broadband revenue. One is to sell more customer subscriptions, and the other to generate more ARPU. Put aside any OTT services or premium connection features, and the only thing that you can get ARPU gains from is selling higher speeds. That’s been tried too, and the truth is that it’s sort of worked, but it’s limited in its impact by the fact that there’s little or nothing an operator can do to push the value of speed. Streaming video has been the big driver there, but content providers and content producers drive that space.
Most streaming plans already charge more for 4k video, and most users don’t pay for it. Could you push 8k out? To a few for sure, but some retail TV sellers tell me that in a showroom it’s hard for users to tell the difference between a 4k TV that’s the standard today, and an 8k model, because there’s no material produced. The streaming providers say most users won’t pay for even 4k, and that 8k is useless because there’s no 8k TV to view it. You can see that there’s a whole ecosystem, like our Internet cooperative, that has to go along with a change before anyone can see the benefit, or spend the money. Who moves first, making not return until everyone does? Few, very few.
I think that “fiber wars” are delaying the inevitable at best, and at worst are diverting resources that the operators could use to actually advance new services and new service features. I also think that the delay benefits that operators seem eager to seize is running out of room to have favorable impact, which means that a reckoning on telco/MSO profits is likely coming. Not in the next three years, probably, but in the next five.
