The End of the Broadband Buffet Is Nigh
Get ready for the next generation of fiber to the home, which will deliver 10 Gbps downlink and 2.5 Gbps uplink shared across 32 homes. Verizon will announce next year that it has achieved these results in its labs, a huge improvement over the 2.5 Gbps down and 1.2 Gbps up the company is currently deploying. But get ready to dig deeper into your wallet, too, because even if the demand for broadband isn’t breaking the Internet, it’s surely forcing ISPs to rethink how they charge for such a valuable service — even Verizon.
While many ISPs are implementing caps or tiers because they have real resource constraints at points in their network – Comcast’s 250 GB-per-month cap, for example, is aimed at stopping folks from unduly clogging its shared, last-mile networks — some are eying such measures as a source of additional revenue and a way to fend off potential competition from online video. The best example of this is Time Warner Cable’s tiered broadband plan, which caused such consumer ire that the company ended up backing down from it. However, when Verizon’s CTO Dick Lynch said in September that Verizon was also in favor of some kind of consumption-based billing, it was kind of like watching your favorite indie rocker sell out. Why would Verizon, which is building out a fiber-to-the-home network, plan to eventually move to some sort of consumption model?
The answer is because it can. I spoke with Brian Whitton, executive director of access technologies at Verizon, about the FiOS network in an effort to get some clarity regarding the rumor that even it would eventually face constraints under the onslaught of video. Whitton quickly disabused me of that notion, pointing out that the network is built to be upgradeable for decades to come by replacing electronics at the ends of the pipe. He basically told me that Verizon didn’t spend $18 billion (it spent $23 billion in total but some of that would have been spent anyhow) in additional upgrades to its network only to rip it out a few years later, and explained how the fiber stretches from the customer home all the way back to the fiber-based long-haul network.
But he did acknowledge how valuable broadband has become — valuable enough that people will pay for premium access to it, especially those using up a disproportionate amount of network assets. “Ultimately this is the fairest cost recovery model, and with a tiering plan or a meter everyone is paying their fair shares to finance the network,” Whitton said. Unlike other ISPs, Verizon doesn’t view heavy bandwidth users as hogs, but it does view them as potentially high-end customers.
What’s frustrating as a consumer is that because of the fragmented nature of competition in U.S. broadband markets, providers offering these consumption-based plans don’t have real competition to keep prices in check. Most places have two providers that offer slightly different sets of services and plans, making it hard to compare prices. I don’t mind paying more for a better network (I do so for my cell phone), but most consumers don’t have that option when it comes to wired access. While Comcast (which competes against Verizon in about 12 percent of its footprint) is rolling out faster broadband to ensure its customers don’t leave the cable provider for fiber, in other areas of the country, such as here in Austin, Texas, folks get the choice between DSL (with some U-verse) and cable that hasn’t been upgraded to the faster DOCSIS 3.0 speeds. So while Verizon’s FiOS will deliver faster service to customers, as it implements consumption-based billing of some sort, there’s no real competition to keep its pricing in check for those who have a need for speed. And eventually, we’ll all have that need.
This article also appeared on BusinessWeek.com.







I am in support of consumption based billing. Make it fair and easy to identify. Downloading a Blue Ray movie costs say a $1 where as a standard DVD costs $0.25, as a consumer I will make the appropriate decision.
In order to facilitate this billing, broadband operators need to provide a good meter and have a rational plan. I suspect that providers have not come up with one yet because consumption may go down and they won’t be able to make even as much money as they do now!!!
“I am in support of consumption based billing.” – I support this too, if competition were available. So far the only real competition I have seen is in larger cities. In smaller cities, many suburbs (even around some larger cities), and rural areas have little to no competition to keep prices moderate to low and keep greed in check.
As long as this model continues, I don’t support anything other than unlimited. As time goes on technology advances and costs typically drop over time, yet just like with POTS, the major players in the game seem to think that making services that are continuing to age, advance, and lower costs deserved to be labeled “premium” (notice I DON’T capitalize the ‘p’) and charge more for it. I think the issues brought up in this article as well as that companies, noticing talk about potentially making internet a utility (which would be consumption based — but regulated as long a little competition is available) want to make more and more from it and make an aging technology suddenly sound like a Porsche that few can afford rather than the family sedan that it is quickly becoming.
The only way to “make it fair” is by not implementing it at all. If ISPs want to charge per bye (as with electricity or water) then fine, but that’s not what’s being proposed here. Instead of paying per byte (which could reduce costs for light users), they want to charge you a flat rate but tack on arbitrary and punitive overage charges, similar to over-limit fees and late charges the credit card companies levy. This isn’t about fairness or altruism, it’s a money grab pure and simple. I can guarantee you that no one’s bill will go down under this plan, and we will ALL end up paying a lot more….not just the “bandwidth hogs”.
1) It’s not about fairness. The cost for the ISP is in establishing and maintaining the connection, not the amount of bits “used” or “consumed” 2) ISPs don’t need the money. They’ve seen steadily increasing revenues and profits with broadband and decreasing hardware and backhaul costs. Just look at their financials. 3) It “conveniently” allows multiple system operators (MSOs) who offer TV and phone service a way to artificially protect those business which are organically giving way to Internet services
Keep it “fair” by keeping the same business model we have today. Offer tiered pricing based on speed of the network connection provided. Keep it “easy to identify” by not forcing users to watch a meter or count bits, and allow users to not worry about impending overage charges and punitive fees.
“Get ready for the next generation of fiber to the home, which will deliver 10 Gbps downlink and 2.5 Gbps uplink shared across 32 homes. Verizon will announce next year that it has achieved these results in its labs, a huge improvement over the 2.5 Gbps down and 1.2 Gbps up the company is currently deploying.”
What’s your source for this assertion? What technology is it?
Verison FiOS.
That is not a technology, but a marketing term. What specific PON standard is supposedly behind this, if there is one?
Tim, source is Brian Whitton. I would love to know which tech they trialing (current is GPON), but when asked for clarification a spokeman told me the standards for the new stuff haven’t been nailed. Since they aren’t announcing it until 2010, I think what I wrote is all I’m going to get :(
Thanks for the reply Stacey. Given the current state of 10 Gbps GPON, my impression is that the reason Verizon is leaking plans to announce lab trial results (!) next year (!!) is to prevent other vendors from following TellLabs’ and Nokia’s lead in abandoning GPON, which is far behind EPON in terms of the upgrade path to 10 Gbps.
This article, dated last month, at http://www.cable360.net/ct/voice/10G-PON-Moves-Ahead_37879.html claims, “The NG PON (G.987) standard, both 10/2.5Gpbs asymmetric and 10/10Gbps symmetric phases, is expected to be in “consent” phase until June 2010.” The consent phase is the first step of the ITU standardization process. Compare that with the recently ratified IEEE 802.3av standard for 10GEPON, for which implementations from multiple vendors have already been tested at interoperability trials.
Tim has a very good point in here. 10G-EPON came much earlier, is already stable and given interest of China and Japan in mass deployments in H1 2010, I think we are looking at a very firece competition and rapid cost erosion which will drive the interest in this technology among small and large operators. Put that on top of 30 million existing 1G-EPON lines, and the picture is clear.
Looking from a perspective of an SP, this is a solution which is available now, will become only better with time (as the cost goes down) and provides an SP with bandwidth now instead of 2-5 years in the future, when the folks from 10G GPON team work out all the kinks out of their systems (if that happens at all, given how long it took to set up basic interoperability for GPON products). Even when they get there eventually, it will be just another ‘me too ‘ kind of system, providing nothing revolutionary, different or better for that matter. What counts at the end of the day is how much a Mbps of bandwidth costs and there is nothing that beats a fat Ethernet pipe.
Considering the news on Motorola getting rid of their access department (http://nerdtwilight.wordpress.com/2009/11/11/motorola-division-up-for-sale-wsj/), a question remains whether GPON has any future at all … I’ll say 2010 will be an interesting year to watch how the PON story unravels
I disagree. Get ready to pay more? I don’t think so. I am getting ready to dump any provider that can’t deliver a fat pipe for a reasonable price.
I have had Verizon FIOS for just over a year. It is a fantastic broadband pipe that delivers 100% of my paid for speed 24/7. Never an issue. It delivers beautiful HD television on more channels than I’ll ever use. The quality of both broadband and TV services puts my old Time Warner premium package to shame.
While I would gladly pay a bit more for even fatter pipes, bottom line is that they’re not going to get much more out of me for the combined services. Up the speed and price of my broadband? No problem, I’ll switch to iTunes and Hulu and torrents and drop my premium cable channels. Even premium subscribers have limits to what they’ll pay.
It’s interesting that Whitton appears to acknowledge a distinction between tiered and metered broadband.
Right now, I am wondering what the FCC’s national broadband plan will add to this discussion. More competition among providers and greater availability might put and end to any metered or tiered plans.
On another topic, Stacy mentions that she pays more for her cell phone to get a better network. Months ago she posted about making the switch to prepaid because it seemed a better value: http://gigaom.com/2009/08/05/im-taking-the-prepaid-plunge/
Is the experiment with prepaid over? Can we look forward to a post about your experiences?
Jake, thanks for the reminder. I’m still on Verizon and will try to write up my prepaid experience before Thanksgiving.
So, Verizon’s slowest tier of FiOS will be, what?, 50/20? for $99.95? (plus any usage charges, of course) Sorry, don’t really need that… few do. As we know, it’s not about paying for the network itself; it’s about paying for all those new toys (boats, cars, mansions, etc.) wanted by corporate officers and investors.
Everything I actually “need” Internet access for I can get free down at the library or at the coffee shop down the road. For those other things I don’t need–video streaming, for example… where’s my 5/2 for $30 (or even less)?
“Muni” is not a four-letter word. :) Maybe it’s time to nationalize those cables running through OUR public lands… hmmm?
“Unlike other ISPs, Verizon doesn’t view heavy bandwidth users as hogs, but it does view them as potentially high-end customers.” – Wow, what a breath of fresh air!
BT is better
The 10 Gbps tech Verizon will upgrade to will probably be 10GPON or possibly 10GEPON. Either of which can handle 300+ Mbps of download throughput and 75+ Mbps of upload throughput PER HOME across 32 homes simultaneously. This is just a crzy amount of bandwidth.
Since at that point everyone can have oodles of non-blocking network capacity, there’s no need for speed tiers; Verizon could give everyone a gigabit symmetric if 10G symmetric infrastructure was used and still be able to provide service to everyone without peak time slowdowns, since no one person would be using more than 10% of the link. Contention ratios of 3.2:1 on the last mile are unheard-of except in fiber situations; Comcast has maybe 200 customers per node sharing 190 Mbps of downstream bandwidth and 30 Mbps of upstream bandwidth across five downstream and one upstream DOCSIS 2.0 channels (my DOCSIS 3 modem bonds four channels currently and I know there’s another channel on my node).. With every customer n at least a 12 Mbps package that’s 20:1 or higher oversubscription.
So now, with all last-mile congestion issues removed, Verizon doesn’t have to dfferentiate its tiers by speed. There’s no extra cost to them for provisioning a gigabit connection to the home because they have scads of capacity to work with. That said, I believe Verizon’s network backbone is “only” a few tens of gigabits in some places (though tey’re working on hundred-gigabit-per-wave links) and thus the burden is sifted back to the backbone.
Anyway, since Verizon doesn’t have to differentiate their tiers based on speed, they need to have some other way of differentiating customers by price to get the premium users paying premium prices. The easy way to do this is caps: heavy internet users data-wise get more value out of their internet connection than light users, and are thus willing to pay more. If Verizon had FiOS here I might well be on their $100 50/20 tier (bought via DSLExtreme). Currently I’m willing to pay Comcast $75 per month for 2/5 access, though I might later decide to change my mind and switch to 12/2 service for $55. In no event am I going to hobble myself with what Qwest has there: $35 4 Mbps down, 700 kbps up DSL, after you figure in overhead from ATM. Of course those tiers are based on speed but I’ve seen the pattern that the folks who pull down more bits pay for higher-tier plans.
So back on topic. I’d expect Verizon to, upon switching to 10GPON/10GEPON, up al tiers to the same (high) speed (100/50?) and start tiering by usage cap rather than speed. So a “Light” plan would be $40 for 100GB, a midgrade plan would be $60 for 300GB, a higher-end tier would be $90 for 750GB and an “Extreme speed” tier would be available, uncapped, for $140, less in competitive markets. At that point I’d probably pay for the 300GB tier most months, switching to 500GB or 100GB tiers depending on how much data I was thinking of pulling down, or whether I was going to be out of town for awhile.
When providers start talking about consumption-based (read: capped) billing people have good reason to be nervous: Time Warner Cable’s capping plan was atrocious. However if uncapped tiers are available to power users at prices lower than a typical triple play, everything’s cool. That’s why I’m not parading around denouncing Comcast for soft-capping my connection at 250GB; for $25 more per month I could get the same tier in business class and push a few TB of data over it if I wanted.
[...] Stacey Higginbotham has a good post at GigaOM today where she raised the point of tiered services with Brian Whitton, executive director of access technologies at Verizon. Why would Verizon, which is building out a fiber-to-the-home network, plan to eventually move to some sort of consumption model? [...]
[...] GigaOm’s Stacey Higginbotham found out, the answer is – because they can: Brian Whitton, executive director of access technologies [...]
[...] Stacey Higginbotham has a good post at GigaOM today where she raised the point of tiered services with Brian Whitton, executive director of access technologies at Verizon. Why would Verizon, which is building out a fiber-to-the-home network, plan to eventually move to some sort of consumption model? [...]
[...] Higginbotham (no relation to Michael Hickenbottom) wrote an interesting article for GigaOm about the end of all-you-can-eat broadband access. Most Internet providers are working towards tiered plans with set bandwidth caps. Cable companies [...]
[...] 18, 2009 | 5:00 PM PT | 0 comments | 0 tweets retweet » Even though we’re inching ever-closer towards consumption-based broadband, not all ISPs are implementing metered or tiered plans as a way to punish users who clog their [...]
[...] service providers are looking at consumption-based billing, as indicated in recent posts by GigaOm’s Stacey Higginbotham and BroadbandReports’ Karl Bode. (Oh, and who could forget the StoptheCap! guy, who’s [...]
[...] you read our blog in mid-November, you already knew about this, but we didn’t have details on the standard Verizon was using. The XG-PON standard [...]
[...] you read our blog in mid-November, you already knew about this, but we didn’t have details on the standard Verizon was using. The XG-PON standard won’t [...]
[...] billing”) for broadband services is far from over. But while as Stacey has pointed out, some broadband execs believe metered billing is inevitable, existing and proposed implementations contain significant shortcomings. So if metered billing is [...]
[...] The End of the Broadband Buffet Is Nigh (gigaom.com) [...]