High Latent Mobile Web Demand?

If iPhone users, and a recent study of smart phone users, are any indication, there is clear and vast potential for mobile Web applications, devices and services.

And that is despite the relatively low usage of mobile Web services at the moment. "It is amazing how unaware consumers are of what is, and what is not available" in mobile, Web and other forms of communications, says Elaine Warner, Compete.com analyst.

On the other hand, there is clear potential. “We asked smart phone users what was important to them and 68 percent said Web access was really important,” says Warner. Considering that just seven percent of respondents to the Pew study say they do so on a typical day, Compete’s findings suggest there is vast untapped potential.

One of the biggest struggles the mobile industry has is getting the user experience right, though Warner says the iPhone was a breakthrough.

“We did a study about iPhone and found the two things people want is surfing the Web and checking their personal email,” says Warner. “They still feel they can't do that easily.”

Along the way, application and service providers will have to adapt the context of mobile Web use. “You don't search for the same things you do on a PC as you do from a mobile handset,” says Warner.

“You don't want a Wikipedia page to be the top listing when you enter a search term, she says. “That’s not likely to be what you want.

More typically a user will want to find a place to get to, or something to buy.

Though “voice in your pocket or purse” was the initial “killer app,” sizable demand now exists in the “email in your pocket or purse, “music in your pocket or purse” and to a lesser extent “Web in your pocket or purse” user segments.

That few people have used the mobile Web up to this point is understandable. It has been a difficult experience, for the most part. And it may turn out that early iPhone users are particularly avid users of the Web.

But if Compete’s survey findings are any indication, there is pent-up demand for mobile Web access.

How Many Lines or VoIP Accounts?

Suppliers shipped an estimated 9.8 million VoIP subscriber feature server licenses for deployment in service provider networks, according to analysts at iLocus. Those licenses generated $177.4 million in revenue, and grew
34 percent, quarter over quarter.

The growth is due to high voice over broadband activity in Europe and among cable operators in North America. In Asia-Pacific VoBB growth is still confined to Japan mostly.

Of the 9.8 million VoIP subscriber licenses sold during 4Q07, licenses for hosted business phone system (hosted PBX or hosted Centrex or key system) lines account for about 1.2 million.

The remaining 8.6 million were mainly deployed for residential VoIP or switch replacement, iLocus says.

That suggests, at least for the short term, a belief that 12 percent of overall VoIP sales by service providers are of the hosted phone system sort.

Keep in mind that such data is not so granular as we might hope. In fact, even the reported penetration of landlines is less granular than one might think. If one looks at reported landline phone penetration, for example, there is a period between 2005 and 2007 where the installed base appears to oscillate wildly.

It appears that changes in the survey instrument are partly the reason. Government researchers now ask whether "any" phone service is available, specifying that mobiles count, where they used to ask whether a phone line was available. The government now makes a distinction between phones "in the living unit" and "available in the building" as well.

So it is likely we simply have reporting error in recent data. Over time that should correct. But the point is that even the official Federal Communications Commission data now have to be interpreted.

It's just another reminder that all our survey data should be considered indicative of trends rather than firm descriptions of physical reality.

Google IS Online Advertising

Though total U.S. ad revenue at 17 public companies increased nine percent in 2007, online revenues grew 28 percent, says Henry Blodget, Silicon Alley Insider author. Total ad revenue for the 17 firms was $58 billion, while online revenues were $18 billion.

Offline revenue grew about $1 billion while online grew $4 billion. Google got $2.7 billion of that total, while online ad revenue at Yahoo, Microsoft, and AOL grew $1.3 billion. In other words, says Blodget, Google captured twice as much revenue as its closest three competitors combined.

Google.com's U.S. revenue growth was more than twice as much the growth of ad revenue in all of the 13 offline media companies Blodget tracked.

More Online Video, More Managed P2P

Online video sites have delivered promising stats recently, says Compete.com analyst Aniya Zaozerskaya. Netflix’s WatchNow, which allows subscribers to any Netflix plan to watch full-length movies and TV episodes online from their collection, had 69 percent more people using the service this quarter as compared to last quarter.

Veoh.com, which allows users to view and share short YouTube-like videos as well as stream full-length TV show episodes, has grown from just under 1.5 million unique visitors one year ago to over six million in February 2008.

Hulu.com, a newer site offering both full-length movies and TV shows, including the most recent in-season episodes, also is gaining traction, she says.

Assuming peer-to-peer applications are deemed lawful, and therefore not to be blocked--and that seems a certainty--managed P2P services would seem to be poised for growth.

One reason P2P chews up so much bandwidth on service provider backbones is the unmanaged way P2P traditionally operates. Bits of content might be fetched from long distances when the same material actually resides on a user hard drive someplace local.

So far, it appears, managing P2P streams can reduce overall backbone network traffic by 60 percent or more, executives at Pando Networks and Verizon Communications say.

Network-aware versions of P2P that can fetch data from local sources rather than reaching far across the network, can help,in that regard.

103 Million HDTV Households

Analysts at Pike & Fischer predict that up to 103 million U.S. households will be paying their multichannel video provider for some form of HD service or rental equipment by the end of 2012. And while most analyst forecasts wind up being overly-aggressive, this projection is about as grounded as a forecast could be. Since more than 90 percent of U.S. households subscribe to cable, telco or satellite-delivered television service, and since the U.S. broadcasting system is moving to an all-HDTV format in 2009, it only makes sense that most people are going to be watching at least some HDTV from the moment of transition.

And while analog tiers of service will be offered for several years after the broadcast transition, most viewers are going to switch, fairly quickly. Cable operators, of course, now are saying they will voluntarily continue to simulcast analog local station feeds until 2012.

Keep in mind that the total number of television households in the U.S. market, including Alaska and Hawaii, is 111.4 million, according to Nielsen Media Research. So in predicting that 103 million U.S. homes will be paying for some form of HDTV, three years after the transition date, Pike & Fischer is making a simple observation that 93 percent of households will be on an HDTV-capable tier of some sort by the time U.S. cable operators will have switched off their off-air analog feeds in favor of the HDTV feeds provided by local broadcasters.

So there may still be some hold-outs in 2012. But, by and large, Pike & Fischer simply makes the point that most people will continue to buy a multichannel video service, and that by 2012, virtually all those providers will be selling HDTV programming widely available on the basic service tier.

Again, most analyst projections err on the side of excessive optimism. This isn't one of those cases.

FTTH is inevitable

No matter what posturing now occurs, cable operators and at&t someday will switch access platforms and adopt fiber-to-home as the standard wired access approach. For the sake of pleasing investors, who seem to hate investments in FTTH that are the only long-term hope for any wired access provider, lots of companies insist they do not presently need to do so, and they arguably are correct.

Other small independent providers in very-rural areas likewise will insist they cannot afford FTTH. That ultimately will be resolved either by new forms of rural or high-cost area subsidies, or by some new hybrid delivery platform using fixed wireless as the tail circuit.

None of that is relevant. Demand continues to increase, and at some point, the only sane choice for a fixed network that has to deliver a minimum of 100 Mbps worth of data bandwidth, not to mention video, is FTTH.

We might be four to eight years away from that point. The precise timing, though, isn't so important. No matter what executives may now believe, they ultimately will have to scrap hybrid fiber coax and fiber to the node, for competitive reasons. When wireless broadband starts to offer anything close to that sort of bandwidth, no wired network is going to be able to avoid upgrading.

That doesn't mean it is sound business practice to deploy platforms of such bandwidth today, in the mass market. The ramp up frankly is best handled on a gradual basis, as local competitive conditions dictate, to conserve capital for a time when the move is unavoidable, under conditions where there is little incremental revenue to be gotten.

But that won't always be the case. One way or another, service providers are going to discover and then create funding mechanisms that make FTTH a rational choice. Just because we can't predict in precise detail what those mechanisms will be is not the issue. Neither could cable industry executives have rationally explained in detail what all the new demand for video choices would be if capacity were upgraded.

Nor could wireless executives, 10 years ago, have presented a clear and compelling line of argument about why text messaging, email or ringtones or music would be generating significant or growing amounts of revenue.

Though there now is an investor revulsion to financing "build it and they will come schemes," in fact that precisely is the history of innovation in the communications and entertainment business. When given choices, developers have responded and consumers have bought.

That doesn't mean every new application, or even most, are going to succeed in the mass market. The point is that we never are very good at figuring out what developers will dream up, and what consumers will flock to.

It is clear that supply creates its own demand, ultimately.

Nearly-Ubiquitous Broadband

Nearly 33 million US households will have broadband services with speeds of 10 Mbps or higher by 2012, up from 5.7 million at the end of 2007, according to Parks Associates. That's not the most important prediction, though.

Parks Associates also forecasts that 32.5 million U.S. consumers will have at least 10 Mbps broadband access service by 2012. Even that is not the most important prediction Parks makes.

No, the most significant prediction is that 75 percent of U.S. households will subscribe to broadband services by 2012. And that is significant because it will make broadband a fundamental service purchased by U.S. households, on the order of cable TV or the place once held by wired voice services.

It is worth noting that very few services ever have reached that level of penetration. Cable TV, mobile phones and wired phones alone could have claimed such distinction. By 2012, if Parks Associates is correct, only for the fourth time in history will any service have achieved such near-ubiquitous penetration.

Some day, and probably not by 2012, we might be able to make a similar claim about wireless broadband as well. For the moment, though, it is noteworthy that broadband seems destined to reach such broad penetration. Lots of services exist, or have existed, without ever getting nearly universal acceptance.

Compared to that level of acceptance, the subsidiary question--how fast is fast enough--while not trivial, is not fundamental. Access speeds have been increasing on a fairly steady basis, much as storage capacity on PCs, mobile devices and other devices has been increasing, and much as cable TV or satellite networks steadily have increased capacity and channels over time.

The average download speed of a US broadband connection is currently 3.8 Mbps, while the average upload speed is 980 Kbps, according to In-Stat researchers. But there was a time when a typical cable TV network delivered just three channels. Then capacity went to 12 channels; then to 25; 40; 60; then 66; then more than 80; then 115; then, with digital, hundreds of channels. Ad-free formats, then pay-per-view, then on-demand programming developed. Music services also were introduced in the 1980s, though not to notable success.

Over time, mobile services have added text, Web access, email, audio and video services. And there have been continual improvements on the value side as well, as costs for calling have dropped dramatically. Even legacy wired voice services have been upgraded in important ways.

Party lines were replaced by private lines, and enhanced services expanded particularly when digital switches were substituted for analog switches. Now, using VoIP, all sorts of enhancements beyond what are known as CLASS features are possible.

The point is that penetration matters. Given high penetration, continual evolution of features and value almost are inevitable.