Verizon said this week that it now has 1 million customers using its FiOS fiber to the home network, and around half of these are using it for just broadband and the other half for broadband as well as FiOS TV.
Verizon FioS TV only passes 3.1 million US homes, which means that it gets revenues from one in six of these homes and raises the issue that perhaps the rest of the world got it wrong when it decided to use DSL lines for IPTV and broadband?
Specifically it gives us a good moment to contrast AT&T’s efforts to gain a triple play, with Verizon’s, and gives us a glimpse of what the future might hold for these two giants if they each remain on the current track.
Given that Verizon’s introduced FiOS TV in just September 2005, only 21 months ago, its current rate of installs over all that time is around 25,000 a month. During the last quarter it has added a further 141,000 customers, as it has opened up in more markets, which is more like 47,000 a month, or 12,000 a week. Remember the rate will rise as two things happen, the total fiber footprint goes slowly up, and as consumers visit the homes of early adopters and see just how good, or otherwise, FiOS is, currently comparing it to satellite or cable.
But already Verizon has built out fiber to pass 6.8 million homes, one of the key factors slowing it down has been the rate at which it has acquired the 760 video franchises that it holds. Potentially all of the outstanding licenses might be issued tomorrow, doubling the rate of FiOS customer acquisition and the footprint doubles overnight.
In fact those existing franchises reach around 10.1 million homes, said Verizon CEO Ivan Seidenberg speaking at the NXTcomm show in Chicago last week, so it has more permissions than it has fiber, and as it builds more fiber in areas where it already has the right to offer service, quite quicly these numbers will come together to be above 10 million potential households.
Seidenberg said that five years ago, Verizon had 1.6 million broadband customers, whereas it now has over 7 million, and that for the first time in a long time, consumer revenue is growing again at Verizon.
His key point is that around 80% of those people that do decide to go with the video FiOS package, take the triple-play package across the board. Currently Verizon is advertising a $99.99 a month triple play from its web site, and this is the deal that many of this 80% are taking.
That’s an easy sum to do. Verizon has 45 million customers today, and by the end of 2009 it should have around 20 million homes passed by the FiOS service. If a third take some form of service that’s 6 million customers by sometime in early 2010, and with half of them taking TV, that’s 3 million FiOS TV customers, 80% of which are likely to be paying $1,200 a year to the company, some $3.6 billion a year in triple play.
Now we’re not sure how much change that will generate in the company’s entire churn rate across the US, because this is just among a tiny group of the Verizon customers, but that amount of business may be worthwhile long term in order to justify the $18 billion or so that the company has said it will spend on fiber to the home. Of course there will also be other revenues from that fiber, such as broadband and VoIP service, both of which can provide innovations and new service capabilities (take for instance converged fixed mobile voice) and both of which are future proof in terms of the 50 Mbps plus that each consumer could potentially opt into.
In order to reach a number like 3 million, by say this time in 2010, FiOS would have to sign up an average of around 69,000 a week, which is entirely feasible, given that with 20 million lines and video franchise homes both in place this would represents just 1.4% of the total homes reachable being attached each month. In fact any linear extension of the rates of monthly additions would suggest that even more homes, perhaps as many as double that number of 3 million could eventually sign up for FiOS TV.
If the world’s operators look in detail at that kind of figure, it will shortly be Verizon executives in demand at IPTV conferences around the world instead those from Italy’s Fastweb or PCCW. If the investment case makes sense both in terms of return on investment over a predictable period and in terms of churn reduction in the basic telephone homes, then Verizon could prove the blueprint for the future of telecommunications. Or could it?
At present Verizon’s wireline revenues in 2006 were $50 billion, and this represented 55% of Verizon’s total revenues. Looking at the diagram below we can see that in fact this 55% gave rise to just 29% of all of Verizon’s profit, the rest coming from Verizon Wireless.
A brief look at these figures would make any operator that does not have wireless telephony spectrum go out and prepare a bid for some, rather than begin costing a case for putting fiber in the ground.
If we re-calculate this, taking out tax and other provisions and eliminating the Vodafone 45% holding in Verizon Wireless, there’s still a huge swing in favor of the wireless business, with 35% of net income due to the wireline business and 65% to the wireless part of Verizon.
Funnily enough that’s not in keeping with AT&T’s figures, where wireless revenues is only 37% of last year’s $63 billion total and net income from wireless is just 21% of the $7.3 billion total for 2006. This is likely to change dramatically the next time AT&T releases annual numbers and in this last quarter they were already showing the acquisition of Bellsouth where revenues were on track for over $100 billion, with Wireless set to be around 40% of total revenues, and deliver around 33% of income. The whole thing of course is skewed by the consolidation with the old AT&T which adds weight in the enterprise and makes it look like mobile is not so important, but it is. New CEO Randall Stephenson was making that very point at NXTcomm, telling the world that mobile was the future and that everyone should remember that not only was AT&T Mobility larger than Verizon Wireless, but it was 100% owned by AT&T.
All of this would underline our earlier question, why invest in wireline at all, especially at Verizon, where it brings so little profit. But funnily enough, every mobile operator around the world right now is trying to cover the wireline base, and where they do not already own fixed lines in the ground, such as at Vodafone they are trying to acquire them. Why’s that exactly?
Part of this is because of the trend for voice convergence whereby cellular can take over the entire voice world as long as it is wedded to Wi-Fi voice traffic which can bring relief from the high charges of cellular voice, when in range of the home or workplace. Convergence means one voice mail, one device, even if it has multiple identities and phone numbers, and single billing for all voice.
And also because with the emergence of Pivot, the Sprint run mobile service for cable operator customers, there is an inherent threat from cable on the most profitable parts of the RBOCs’ business, mobile.
But this isn’t true outside of the US. Cable is weak, less technically advanced, operating under differing legislation and regulatory environments and nowhere near ready to take on cellular through anything other than limp wristed MVNOs.
So bereft of the threat of resurgent cable, does that mean that elsewhere in the world the fiber route is not worth spending the considerable sums of money on, at a rate close to $1,000 per connected home?
Well first we must look at the AT&T figures for IPTV.
The Verizon figures compare pretty favorably with AT&T’s U-Verse roll-out, and while we have to concede that much of the AT&T delay might be down to its own caution, rather than a reluctance of consumers to take on its service, its own numbers are just 600 sign ups each week, around one twentieth of the current sign up rate of Verizon. AT&T says that by year end it will be signing up customers at a rate of 10,000 a week, which will remain around two thirds of the eventual rate for Verizon installs, and this may create, in the minds of its investors, the idea that its previous management, in the form of Ed Whitacre, had it all wrong.
But there are differences between the two systems. The Verizon service is not an integrated IP based system. It runs on Microsoft TV Foundation software that was designed for cable operators, where each TV channels sit in their own spectrum instead of being part of an IP multicast. That software has been much re-written by Verizon who took control of the project early on, and began writing code in abundance and unlike AT&T, made its displeasure widely understood, rating the Microsoft code poorly in reports and leaks to newspapers.
Regardless of who wrote the more advanced features, the system can offer an IP channel in and out of the home, on a single dedicated pipe, so high speed internet and VoIP voice are not a problem to deliver, but the data involved in video is not inherently IP in nature and much of the plant and equipment that Verizon had to buy was openly available on the market, low priced and proven. AT&T was far more pioneering and had to wait for equipment to arrive with no guarantees that it would work with consequent delays, frustration and a lack of confidence in the market place.
So what looks to the outside world like superior execution from Verizon may in fact turn out to be little more than just the choice of existing technology. But what can new, all IP technology do, which Verizon’s system might not be able to?
Well let’s take the example that has been kicking around for years of a VoIP call coming in, and it being displayed on the TV screen. Opting to take the call pauses your program and runs it out onto the DVR for later viewing. Cut off the call and you can carry on watching TV. With an all IP work, such an application is far easier to execute. A set of APIs can be created and expose various services for new applications to discover. There is no transcoding, no file transposing and less set top based processing.
It can be done, sure it can be done, but it will almost certainly mean that there is a subsequent system penalty.
Take another example such as talking on a cellphone, walking in the door and the service transferring to the local broadband line, and adding video to the call, perhaps IM, on the TV screen. Same principle applies, codec issues, file format issues, real time transcoding, it’s kinda horrible in a non-IP world.
What about advertising. You want to insert an advert live into a video stream, you send the advert from an internet based server. The ad might have different versions for gaming consoles, DVRs, live TV, and it will need to discover the device dynamically, do some demographic lookups and insert the advert. Such an application will only come together once there are standards, and multiple links in a logistics and demographic chain, and these all need to talk to each other through IP.
It sure as hell isn’t going to be easy to retrofit this to the Verizon set top box without putting in something approaching the power of a supercomputer inside. To date we are led to believe that these types of features are tough to fit into U-Verse too, but that is certainly execution, and not due to the fact of the pure IP nature of the system.
But does this adherence to IP only technology actually help when keeping big bad cable at bay? Surely if Verizon can drive up TV and triple play subscribers now, it can deliver these types of applications itself in the future, even if it needs to work a little harder to do it. It has already shown it executes will on this type of thing.
Well we can already see the effect of this on cable itself. Cable set tops and its multi mode way of working internally has already put it at a disadvantage in adding things like advertising insertion networks, and they have been far slower to develop than genuine IPTV.
Let’s take the recent idea that is circulating regarding AT&T’s partnership with Apple, not just on the iPhone but on the Apple TV. Could AT&T add Apple TV alongside its existing set top boxes and give AT&T subscribers access to the iTunes library of music, TV shows, movies and podcasts, as well as all that YouTube content that Apple has just signed for? Well yes of course it could if it thinks that’s a good strategy.
Can Verizon do the same? Now that’s tricky, it’s got to re-engineer lots of steps in the process. It can do it, but not as fast, not so transparently and not with ESG integration, done cheaply and swiftly. But AT&T could.
Okay we realize that with Apple having been rejected by Verizon for the iPhone, it won’t be doing a deal with on Apple TV with Verizon anytime soon. But you take our point, these types opportunities come up and being fully standard and IP based make them easier to take advantage off.
So what’s likely to happen here. AT&T is under constant scrutiny by investors, but not for spending the $4.5 billion it plans to spend on network upgrades during the next few years for U-Verse TV. Instead it has to make a success out of integrating its acquisitions. If its new CEO manages that well, no-one on the investment side will be able to complain about a little money spent on IPTV.
Verizon on the other hand, may have a TV system that brings it little relief on churn across its entire customers base, because it won’t be available to even 50% of its customers in the near horizon. And that TV system may run into development difficulties and in turn become a millstone around its neck that ends up dating the system and giving the company developmental headaches. Deployments could, and in fact are likely to, slow down.
Europe and Asia Pacific will not follow fiber to the home, and even where it does it will do it in pure IP, and lead to little in the way of system commonality. Solutions to problems like integration of new systems will not be shared then with other tier one operators.
So the articles that ran this week crowing about Verizon’s success for FiOS and taking that as a slam dunk against the AT&T strategy, should perhaps look a little further out into the future. Each operator makes the decision that appears right for them, at a given point in time. Verizon for instance had to cope with Cablevision taking away customers to its triple play and VoIP offers a full year before the rest of cable had their systems in place. It could be that Verizon took the right decision for its circumstances and that AT&T also did the same.
Perhaps the bigger question is whether or not either one of these telcos, and the whole host of global telcos that have already entered the IPTV space, needed to go up against cable operators at all with wired systems.
Hybrid satellite IP systems, would have been every bit as good as hybrid fiber IP systems, but they would have cost virtually nothing and would have taken care of the churn issue that is driving 6% to 8% of wireline customers away each year to either VoIP, cable or through fixed wireless substitution. And then the telcos could have spent all their time where their largest margins are almost certainly to be made, investing in more advanced wireless services, voice convergence, mobile TV, and bringing the web to the handset, all services which are likely to have a much bigger effect on profitability, but which until now have been virtually ignored.