Internet of Things Builds Toward Peak of Hype

The “Internet of Things” is building towards a hype cycle peak. In 2013, Gartner positioned the Internet of Things just shy of the peak of expectations. If the typical pattern holds, that means we might see and hear quite a lot about how big a market the Internet of Things represents.

Machina Research forecasts $100 billion in revenue by 2020 for applications such as toll-taking and congestion penalties. Smart parking-space management, expected to drive $30 billion in revenue.

Smart electricity grids that adjust rates for peak energy usage will represent savings of $200 billion to $500 billion per year by 2025, according to the McKinsey Global Institute.

And it is possible to see IoT upside in many consumer appliances, traffic management, waste management, Internet-managed assembly lines, connected factories and warehouses, for example.

But that illustrates a problem, namely that IoT represents hundreds of potential applications, industry segments or business models.

Also, IoT might in many cases represent a feature of some other existing product, such as kitchen appliances, not necessarily a discrete new revenue source.

Of course, if the hype cycle has validity, it also means those wild revenue hopes are certain to be dashed, in the near term.

After a new technology reaches the peak of the hype cycle, there is a longish following period where observers become disillusioned by the market’s failure to generate the promised revenues and growth.

Only after that period is over does a new market actually develop around a new technology that has proven its value.

Internet of Things will hit the peak of hype soon. After that will come a period where practical and real-world applications seem not to materialize. Eventually, that will be overcome. But IoT reminds me a lot of the mobile payment or mobile wallet opportunity.

After the peak of the hype cycle, the industry has settled down into a period where expectations are modest. And some observers are using language such as the “death of” near field communications, failure to gain traction and “no business model.”

IoT will likely experience a similar deflation of expectations. And observers will finally understand that IoT is, in fact, not a single market.

Netflix is Responsible for Experience Degradation, Verizon Says, After Study

Netflix itself is responsible for congestion experienced by its customers, a study by Verizon engineers suggests.

A FiOS customer in Los Angeles wanted to know why the Netflix experience was not optimal, so the Verizon network operations team studied the network connection for this customer for the week preceding the date that he emailed us.

They measured the utilization at every link in the Verizon network to determine where, if at all, congestion was occurring.

The review confirmed here was no congestion anywhere within the Verizon network, Verizon says.

There was, however, congestion at the interconnection link to the edge of Verizon’s network (the border router) used by the transit providers chosen by Netflix to deliver video traffic to Verizon’s network, Verizon notes.

While the links chosen by Netflix were congested, the links from other transit providers to Verizon’s network did not experience congestion and were performing fine.

The maximum amount of capacity used on those links ranged from 10 percent to 80 percent,  with an average peak utilization of 44 percent.

“For whatever reason (perhaps to cut costs and improve its profitability), Netflix did not make arrangements to deliver this massive amount of traffic through connections that can handle it,” Verizon says. “Instead, Netflix chose to attempt to deliver that traffic to Verizon through a few third-party transit providers with limited capacity over connections specifically to be used only for balanced traffic flows.”

“Netflix knew better,” says David Young, Verizon VP, Federal Regulatory Affairs

U.S. Mobile ARPU is Growing

Consumer demand for mobile Internet access has reversed a trend of declining average revenue per subscriber in the U.S. mobile market, the latest analysis by the Telecommunications Industry Association shows.

The average subscriber paid $50.35 per month in 2013, a five percent increase from 2012 and the largest gain over the past three years, says the TIA.  Prior to 2011, mobile average revenue per user revenues were generally declining.

In 2009, average revenue per user stood at $45.96. In 2013 ARPU rose to $50.35. By 2017, ARPU should stand at about $59.75, TIA predicts.

The implications for industry revenue are direct. Where in 2010 U.S. mobile service providers earned $152.6 billion, by 2013 the industry was earning $201.2 billion. By 2017, industry revenue will stand at about $263.1 billion.

Globally, telecom spending (both mobile and fixed) will grow at about a 4.3 percent compound annual rate.

Aside from the strategic role mobile Internet access now plays, the U.S. mobile industry has mostly moved from a “scale” business model (get more customers) to a “scope” model (sell more things to existing customers).

In other words, the industry as a whole cannot grow too much by adding new phone subscribers. Instead, subscription growth largely hinges on sales of connections for other devices such as tablets.

Voice and text messaging revenues no longer are growing, making the revenue growth agenda a matter of increasing both mobile Internet access revenue (selling new connections for tablets, selling more data plans) and new services based on Internet access.

For the first time in 2013, increased spending per subscriber contributed more to overall revenue growth than subscriber growth, TIA notes. Over the next four years, TIA expects higher ARPU will drive revenue growth, with increases at a compound annual rate of 4.4 percent.

While subscriptions grow at a compound annual rate of 2.5 percent (tablets will be key), TIA also predicts 6.9 percent compound annual revenue growth to 2017.

Carrier "Next Generation Network" Initiatives Rarely are Easy, Often are Not Successful

It wouldn’t be unfair to note that telecom industry “next generation network” initiatives often have failed to get traction. Integrated Services Digital Network and Broadband-ISDN (now called Asynchronous Transfer Mode) provide examples.

More recently, IP Multimedia Subsystem and Rich Communications Service have been proposed as next generation network platforms. In similar fashion, network functions virtualization represents the latest carrier effort to overall the physical layer of public networks.

Although voice over LTE (VoLTE) is the largest driver for IP Multimedia Subsystem, it also presents new challenges such as a limited set of compatible handsets and a weak business case.

“It’ll be tough to justify spending capital on a new network for just voice that represents flat-to-declining revenue, requiring providers to look well beyond running voice over IMS,” says Diane Myers, Infonetics Research principal analyst for VoIP, UC, and IMS at Infonetics Research.

“VoLTE rollouts are taking off, but “There is no business case,” at least yet, for voice over Long Term Evolution, says Stéphane Téral, Infonetics Research principal analyst.

To be sure, that could change over time, and some might argue that high-definition voice services already operating on over 100 global GSM networks do have a business case, mainly related to boosting the value of carrier voice services, and thereby stemming churn.

Myers also notes that “most providers in this space are making very little money per user, an unsustainable business model for many independent companies”

Some 83 percent of carriers surveyed by Infonetics say they will deploy voice over LTE (VoLTE) by 2016, up from three percent today. Supporters say the value will come as VoLTE will enable converged services over multiple access networks and devices.

How much that might happen, and whether the benefits are largely found in new revenue, avoided churn or some other value, is as yet a bit unclear.

Fixed-line voice over IP service continues to represent the majority of IMS deployments, with 100 percent of survey respondents planning to have business and/or residential VoIP services running over IMS by 2016.

Will Bundles Slow Voice Line Abandonment?

How soon could half of U.S. homes be mobile-only for voice services? If trends tracked by the National Center for Health Statistics should remain in place, that could happen by about 2020.

Some 41 percent of U.S. households did not buy or use landline phone service in the second half of 2013, according to the latest study by the National Center for Health Statistics.

That represented an increase of 1.6 percentage points since the first half of 2013 and 2.8 percentage points since the second half of 2012.

Should the roughly three percent abandonment rate continue, half of U.S. homes will not have a fixed network voice connection by 2020.

To be sure, if bundling remains as popular as it is at present, it is possible that the rate of abandonment will slow. On the other hand, it also is possible that many of those accounts will essentially be inactive.

At least some customers already have taken that route, buying a triple play service to get the bundle prices, then not attaching an actual phone to the fixed network connection.

Others might attach a phone to the line, but rarely use it.

Among households with both landline and mobile phones, 34 percent received all or almost all calls on mobile devices in the second half of 2013, the NCHS reports.

These mobile-mostly households make up 16 percent of all U.S. households. During the second half of 2013, about 44 million adults (18 percent) lived in mobile-mostly households, not too different from the 17 percent who lived in such homes in 2010.

That is a reasonable way of putting some boundaries on future potential triple-play or quadruple-play driven buy rates for fixed network voice services that might not be used too much, if at all.

The rates of fixed network voice service abandonment are higher in certain groups. Some 66 percent of adults aged 25 to 29 lived in households with mobile service only.

Among households of 18 to 24 year old adults, 53 percent were mobile only. Among households containing people 30 to 34, 60 percent were mobile-only.

The percentage of adults living in households with mobile service and no landline voice service  

decreased for those in households older than 35.

Mobile-only voice was characteristic of households 48 percent for those aged 35 to 44, 31 percent for those aged 45 to 64 and 14 percent for those aged 65 and over.

That pattern of faster adoption by younger households has been characteristic of many new technologies and services.

And that pattern seems to be true for adults living in mobile-only homes as well. Among all mobile-only adults, the proportion aged 35 and over has increased steadily, the National Center for Health Statistics says.

In the second half of 2013, more than half of mobile-only adults (55 percent) were aged 35 and over, up from 48 percent in the second half of 2010.

In households inhabited by non-related people, 76 percent are mobile-only.

About 47 of adults living alone used only mobile phones for voice. Adults living only with spouses or other adult family members were mobile-only at a 31 percent rate.

Some 62 percent of households occupied by adults in rented homes were mobile-only. In homes occupied by owners, about 29 percent were mobile-only.

Among all mobile-only adults, the proportion living in homes owned by a household member increased. In the second half of 2013, 49 percent of mobile-only adults were living in homes owned by a household member, up from 43 percent in the second half of 2010.

source: Pew Research Center

How Important is VoLTE?

It is perhaps contentious to ask how important voice Over Long Term Evolution (VoLTE) will be, as similar questions might be raised about high-definition voice or Rich Communications Service.

Telekom Austria CTO Gunther Ottendorfer has said "We shouldn't overhype and over-promise."

There are operational issues of some importance, such as how to handle voice if 3G bandwidth is repurposed as 4G bandwidth, for example.

And RCS, high-definition voice and VoLTE are believed to be important to protect the future voice revenue stream.

Just how important any of those developments might be sort of depends on the strategy any service provider adopts toward voice revenue streams overall.

Some believe the best way forward is to enhance value, to maintain differentiation with other independent and over-the-top alternatives.

Others might argue that harvesting revenue makes more sense, while concentrating development efforts and capital on creating brand-new revenue sources.

It is possible nobody will know for sure for some time, as VoLTE deployments are only beginning. But one issue remains: most carrier voice efforts try to make carrier voice "more like" over the top alternatives.

Whether that will pay off is the issue.

"All Apps Treated Equally" Will Slow Internet Adoption in Emerging Countries

Pandora and Facebook have at least one thing in common: both apps have been bundled as "no incremental cost to use" apps in markets such as the Philippines. In other words, those apps are "zero rated." People can use Facebook, in some cases, even if they do not buy a mobile Internet access plan.



In other cases, Pandora can be used as a zero-rated app when mobile Internet access is purchased. 



It is clear enough why both practices raise the value of a smartphone or mobile Internet access. But both practices would seem clearly to fly in the face of the notion that "all apps be treated the same."



Zero-rated app access--to one app--does not treat all apps equally. 



Mark Zuckerberg argues that, "in the future, everyone should have access to basic Internet services as well, even if they haven't paid for a data plan."



That will encourage more people to get a mobile Internet access plan. Such practices also allow people to experience the value of mobile Internet access. 



But make note: such practices do not treat all apps equally. They treat one or a few apps quite unequally. 



How fast mobile Internet is adopted in many countries could hinge on such practices. That is an illustration of why some think network neutrality is not such a great idea. Antitrust laws exist to prevent unfair business practices.



But giving people access to valuable apps, without charge, is not an unfair business practice. 



Some might call it an important marketing tactic. Others might say it is one way to provide some basic value of the Internet to people who do not yet use the Internet. 



As already has been discovered in developed nation Internet markets, at some point, the barrier to adoption is not "access," the physical ability to get connected to the Internet.



The late adopters typically are people who do not see the value of using the Internet. Zero-rated apps are one way of demonstrating value. 



People can disagree about whether quality of service mechanisms ought to be lawful. It is harder to justify blocking zero-rated apps.