Home Broadband Strategy is Heavily Dictated by the Business Model

AT&T is targeting about 30 million homes for new fiber-to-premises availability. Verizon is emphasizing fixed wireless. As always, strategy is based on firm strengths and weaknesses. AT&T has the largest footprint of homes; Verizon’s fixed network possibly reaches 20 percent of U.S. homes. 

Of a total of 140 million homes, AT&T’s landline network passes 62 million. Comcast has (can actually sell service to) about 57 million homes passed.

The Charter Communications network passes about 50 million homes, the number of potential customer locations it can sell to.

Verizon homes passed might number 27 million. Lumen Technologies never reports its homes passed figures, but likely has 20-million or so consumer locations. 

AT&T has more fixed network share to protect, compared to Verizon, while Verizon has bigger upside in taking home broadband share outside its footprint.

The same holds for T-Mobile, which historically had zero percent share of home broadband. 

“If all I had was a wireless network, if I did not have a scaled physical infrastructure fiber thriving and growing network, I might have no other choice than to leverage my wireless spectrum portfolio to go grow my business,” said Jeff McElfresh, AT&T COO.

The point is that business strategy around home broadband platforms is not based strictly on what is "best" long term. Strategy also must be based on how fast competitive home broadband access can be enabled; at what cost and how fast.

Cable operators have differing opinions about whether to upgrade directly to FTTH now, or conduct at least one major hybrid fiber coax upgrade before doing so. Sheer technology performance is but one consideration. The payback model is more important.

In U.S., Home Broadband Competition is Increasing Fast

Here’s one way of looking at the impact of 5G fixed wireless platforms. In study looking at fixed network competition in the U.S. market, the ACA points out that the percentage of markets in which three internet service providers all operate is growing. 

The numerical test ACA uses is “three providers offering service at a minimum of 100 Mbps downstream and 20 Mbps upstream. So consider 5G. 

source: ACA

Looking only at mid-band spectrum--and ignoring areas where millimeter wave spectrum is available and activated, mobile services by AT&T, Verizon and T-Mobile would be added to the list of markets with at least three competitors. If we assume that AT&T or Verizon already are on the list of fixed network competitors, but only one operates in any area,  then at least two new competitors get added to the ACA list. 

In many areas, three new competitors get added, as neither AT&T nor Verizon are already in the market as fixed line providers. That is a huge change, and will happen faster than new fixed-network competitors enter the market. 

source: Opensignal 

Looked at that way, we would expect the percentage of markets with three to six competitors to shoot sharply upwards by December 2021. That ACA data might not reflect this, as ACA only tracks fixed network competitors. 

But functional levels of competition, using the ACA criteria, will skyrocket. We can thank mid-band mobile spectrum for that change.

FTTH the Platform of the Future, "And Always Will Be"

Fiber to the home is a better long-term solution than fixed wireless, most would agree. Of course, it all depends on whether we are looking at pure technology or practical business models.  

It frequently happens that FTTH is not a practical platform for many service providers, for all sorts of reasons. In a study by the Benton Foundation, commissioned by the Communications Workers of America, the 30-year total cost of FTTH ownership often is lower than the 30-year total cost of ownership for fixed wireless. 

source: Benton Foundation 

We can always quibble about the cost assumptions, but the comparisons seem reasonable enough, on a 30-year payback basis. In competitive markets, 30 years is not a meaningful time frame. Companies go out of business and executives are fired if their platform choices take too long to produce actual financial results. 

So the issue is not whether fiber is better on a 30-year time frame, but whether it is workable right now, and for the next decade, for most internet service providers that must make a deployment decision. 

To be sure, multi-user households remain the customers with greatest need for lots of bandwidth, as the report suggests. 

source: Benton Foundation 

But a substantial percentage of U.S. households are not that sort of multi-user case. Some 28 percent are single-person households, for example. About 30 percent are two-person households. 

In other words, even as bandwidth consumption continues to increase, the direction of change for many decades has been towards households that are not “married couples with children,” the prime example of multi-user accounts. 

source: PRB. 

Since 1960, for example, the average number of persons per household has declined. 

source: Statista 

The point is that even with increasing typical bandwidth consumption, FTTH is not the only platform capable of serving a significant percentage of households, with cable modems being the alternative that is most similar to FTTH in terms of capacity. 

source: Benton Foundation 

As always, the business decision about bandwidth is a balancing of end user demand in specific neighborhoods with the cost to upgrade platforms. Also, mobile operators can use their 5G platforms to reach a significant portion of the market that does not have the highest multi-user household requirement, especially when they cannot justify an out-of-territory FTTH build. 

In other cases, incumbent fixed network providers might have to carefully consider the payback when facing cable operators with lower near-term bandwidth upgrade capabilities and strong market share positions. 

As one wag said in the late 1980s: “fiber is the technology of the future, and always will be.” That is a bit of an exaggeration, but still germane.

Right Now, Metaverse is an Idea, as Was the Browser in 1992

The character-based internet evolved in the mid-1990s with the advent of the web browser https://en.wikipedia.org/wiki/Mosaic_(web_browser) and the subsequent development of the World Wide Web, an audio and video capable internet. The web also enabled new revenue models, especially advertising and e-commerce, with new payment models. 

Many would also say the web also ushered in an era of user-generated content (social media). It is the difference between a read-only internet and a read-write internet. 

source: internethistory.org 

In other words, the history of the internet suggests an evolution towards higher realism, greater enhancement of the “real world” and higher amounts of user-created content, with revenue models developing accordingly. 

Basically, those are also changes many believe will happen in the next iteration of the internet, the shift to more-immersive experiences collectively known as the “metaverse.” Keep in mind that the term was first used about three decades ago, about the time that the first browser was created. 

The point is that it might take decades for this next iteration of the internet to be commercially used on a wide basis. As always, the gestation time is far greater than many hope for or expect. 

source: McKinsey 

Key innovations such as packet switching, transmission control protocol/internet protocol, the domain name system, hypertext markup language (HTML) and the uniform resource locator (URL) were among the innovations that propelled the character-based and then visually-based internet. 

Most observers believe artificial intelligence, virtual and enhanced reality, blockchain and cryptocurrencies are important enablers for the coming transformation of experience, along with digital infrastructure changes such as edge computing, lower latency and higher-performance networks. 

The possibility exists that the “metaverse” will have less impact than many expect, in the nearer term. It might, for example, succeed in gaming without changing much existing e-commerce, education, communications or advertising. 

We are about at the place in the past when the idea of a browser was reality. We really could not foresee how the browser would lead to the range of activities, experiences and business models of the web.

40% Annual Data Consumption Increases Produce Exponential Impact on Capacity Needs

If consumer internet data consumption increases 40 percent per year, usage nearly doubles every two years and grows by 500 percent every five years. If a household consumes 435 Gbytes per month, and increases consumption at 40 percent annually, what appears to be linear growth eventually becomes exponential. 

Source: IP Carrier calculation

Even markets with relatively fixed potential, such as mobility service, which is essentially bounded by the number of living human beings, often feature such exponential growth in the early days of adoption of a new mobile next-generation network, for example. 

source: CCS Insight 

When AT&T executives say they expect data consumption to increase by five times in about five years, that is because they expect a 40 percent annual increase in consumption.

How Much More Can Service Provider Strategies Diverge?

Once upon a time, connectivity provider strategies were quite homogenous. Then came privatization, mobility, deregulation, competition and the internet. These days, service provider streategies continue to diverge.

Some things do not change: connectivity providers are in a business that is capital intensive, slow growing and subject to lots of regulation and competition. Connectivity is a “utility” type business that can have defensive moats and predictable cash flows, but carries lower price-equity ratios than many other businesses, based on the low growth rate. 

So, fundamentally, every connectivity provider has to decide to make the best of possibilities in a slow-growing business, or attempt to boost growth in some way, inside the current business or by moving outside it. 

source: McKinsey 

Many have looked at, and will move to, some form of structural separation, either voluntarily or by government policy action. Co-investment schemes are growing and some of the ownership is shifting away from public to various private forms, including institutional ownership. 

All strategy hinges on those choices. In some markets, organic growth might be possible, especially where gross revenues and profit margins are higher than average. In other markets growth by acquisition is the only feasible path. 

In yet other markets, movement into new business adjacencies might be possible. The net result will be more diversity of business models globally and between industry segments.

Metaverse Building Blocks Will be Commercially Deployed Earlier than Full Environments

As with all other applications, metaverses and metaverse-similar use cases  will require connectivity, data centers and cloud computing providers in the value chain. 

From a digital infrastructure perspective, internet of things, blockchain, edge computing, cloud computing, artificial intelligence and networks are part of the metaverse value chain. 

That means connectivity providers might play parts of roles in IoT, blockchain, edge computing or cloud, while obviously functioning most directly as connectivity providers. Data centers and computing-as-a-service suppliers will have a wider range of roles in edge computing and hosting, 

source: Sketch Bubble 

Of course, it is often difficult to define what we are talking about when we discuss “metaverse.” 

Most attempted definitions for metaverse include the idea of virtual worlds where real people interact in real time. 

Beyond virtual worlds, metaverse technologies typically include use of avatars, three-dimensional representation; bots; virtual reality; cryptocurrency, blockchain; non-fungible tokens; social networks; mobile and other devices. 

But some of those technologies also will be used to support more-realistic experiences that are “less than” full immersion in virtual worlds such as gaming. Digital twins and conferencing provide examples. It is at least conceivable that such uses might initially be more important than full metaverse worlds. 

Value chains and layers are related concepts, in that regard. Since layers are fundamental to modern computing and software, it will come as no surprise that “metaverse” also might be defined in layers. 

source: Innovius 

In principle, it is possible to use many supporting capabilities for all sorts of apps that are not intended to be full virtual worlds. And since it is easier to introduce radically-new technology in a confined manner, rather than as a wholesale “rip and replace” operation, we are likely to see many building block technologies supporting higher degrees of realism before we see successful metaverses in commercial use.

Why Most Firms Applying New Technology Do Not See Outperformance

If business, nature and life have a standard distribution, then the percentage of firms able to turn artificial intelligence into measurable financial results should also be a standard distribution. 

source: Accenture 

Lots of firms, organizations and individuals make a living giving advice about how to boost performance of all sorts. But performance in any competitive arena is a standard distribution. In any market or endeavor, nearly 70 percent of actors will cluster around a median point. You would expect about 15 percent to notably outperform, while 15 percent significantly underperform. 

source: Cate Bakos 

The point is that even if all firms applied artificial intelligence, distributed computing, private neworks, internet of things or any other innovation you can think of, most would still perform at about the same levels as most peers. 

Up to 15 percent will outperform. But those firms were likely already the top performers. They are likely to be the firms already prepared to take advantage of new technologies. 

That is not to deny the value of advice about improving performance. It is to point out that, even with available to all, new technology will result in differential results. Underachievers rarely, if ever, become overachievers because of a single applied technology. 

` `1 -*/*++Perhaps some “average” entities can elevate their performance and become high achievers. More will simply keep pace with their peers and a few might actually underperform after embracing a particular innovation. 

Silver bullets rarely exist, or work. 

Diversify or Not? Sometimes it Works; Sometimes Not

Analysts and advisors often disagree sharply about what telcos ought to do about their “growth” initiatives. Some favor “sticking to core connectivity” while others emphasize “diversifying beyond connectivity.” Service providers have tried both approaches, sometimes alternating between them, as competitive opportunities and threats come and go. 

Some service providers are fortunate to operate in markets with high profit margins. In such markets the advice to “stick to connectivity” can make sense. Others have fewer chances to grow if they stick to connectivity. In those cases diversification makes sense. 

But almost every service provider explores some growth oportunities outside the core connectivity business. How to do so remains the challenge. Growth initiatives are risky, expensive and often do not move the revenue needle very much. That applies as much to edge computing as to internet of things or private networks, for example.     

Back in 2011, KT said it hoped to generate as much as 45 percent of its revenues from non-telecom sources by about 2015. It did not reach that goal, but all the South Korean mobile operators have significant non-telecom revenues, in the 25-percent range. 

source: Korea Herald 

But KT is still investing to diversify its revenue, as are rivals  SKT and U+.

At one point, AT&T earned as much as 40 percent of total revenues from non-telco sources, before reversing course and shedding its content operations to reduce debt. 

source: GSMA 

But most connectivity providers seem interested in growing non-connectivity revenues in some way. 

source: Twimbit 

As always, strategies that work for some service providers in some markets will not work for all service providers in all markets. Still, long term, if revenue growth in core connectivity services remains anemic (flat to negative growth) it is hard to see how most service providers will survive, much less prosper, without getting into new businesses of some kind.

Why Web 3.0 is Likely to Fail in Some Ways, Succeed in Others

The whole point of Web 3.0 is a change in the architecture of the World Wide Web, where decentralization of applications is founded on user control and ownership of their data. The extent to which decentralization succeeds is based in part on the actions advertising ecosystem participants take, since the whole “centralization” of the present web is driven by the revenues made possible by centralization and scale. 

source: TBD 

Also unclear is why any businesses that profit from “centralized” architectures will voluntarily give all that up. “They will be forced to do” is the retort. Value destruction, to be sure, has been part of most internet disintermediation of the past. 

One possible outcome is that value simply is destroyed at many points of the ecosystem, reducing the value of investing. Recall that past hopes for “decentralized value creation” often have failed. Some entities have made a business out of user-generated content, to be sure. But relatively few have done so. 

About the only participants that will prosper from Web 3.0 are the venture capitalists funding startups in the space. 

Some of us would argue that decentralization in the form of disintermediation is likely to happen, but without the more-futuristic advantages of “users owning their own data.” Blockchain will be foundational. But that is likely to fuel disintermediation of value chains, not a complete change of web business architecture.  

10-Gbps Home Broadband is Coming Within 3 Years

Faster home broadband is about as inevitable as Moore’s Law would predict. Having reached the point where top speeds of 1 Gbps are the current standard, we are heading to 10 Gbps over the next half decade or so. 

Which is one reason we are going to be hearing more about 20 Gbps internet access, and why firms such as AT&T already sell commercial service at 2 Gbps and 5 Gbps in lead markets. 

 

’

source: Wik Consult 

Though the demand increase will mostly make sense for multi-user households, the historic increase in top of market speeds is quite linear. That does not mean most users will buy the top-rated tier of service. The general rule is that most consumers will buy the mid-tier level of service. 

source: Commscope 

source: Wik Consult 

source: Fiber Broadband Association. 

60% of Home Broadband Non-Buyers Don't Want It

The latest data from the U.S. National Telecommunications and Information Administration continues to show why the “digital divide,” measured as use of broadband internet access, has not closed faster. 

Nationally, 81 percent of respondents report using the internet. About 71 percent say they use the internet on their smartphones. About 49 percent say they connect their laptops, while 28 percent report connecting desktop computers. 

About 76 percent say they use the internet at home. As recently as 1998, 76 percent of respondents said they did not use the internet at home. About four percent claim the internet is not available where they live. 

Most users report using both mobile and fixed networks. Some 74 percent of respondents have a mobile data plan and 71 say they buy fixed network broadband. 

“When respondents were asked why they don’t use the Internet at home, nearly 60 percent said the main reason is that they don't need it or not interested,” says George Ford, Phoenix Center for Advanced Legal and Economic Public Policy Studies chief economist. 

That finding has been consistent since at least 2015, NTIA data shows. At the same time, “cost” has declined as a reason for not buying broadband access services. Some 18 percent of “non-using” respondents said using the internet was “too expensive.” 

source: Phoenix Center 

Half of U.S. Home Broadband Customers Buy Service at 200 Mbps to 400 Mbps

About half of U.S. internet access customers buy services running between 200 Mbps and 400 Mbps. That is a shift. Until recently, about half of the customers purchased services running between 100 Mbps and 200 Mbps. 

Roughly 70 percent of fixed network broadband customers purchase service at speeds of 200 Mbps or higher. Customers who buy gigabit or faster service have reached 13 percent, while customers of services operating between 500 Mbps and 900 Mbps are six percent of total. 

source: Openvault 

ACSI Rankings Still Show Americans Unhappy with Their ISPs

The latest industry rankings of customer satisfaction produced by the American Customer Satisfaction Index show that internet service providers continue to rank dead last in customer satisfaction. That is not unusual. 

I cannot remember a time since at least the early 1980s when network-based services such as subscription TV services did not rank last to near the bottom in ACSI rankings. As usual, mobile service is the highest-ranked of the connectivity services. 

7

source: ACSI 

Perhaps those rankings have something to do with the recurring nature of the charges. Most other subscription services also rank in the lower third of the industry indexes. Most of  the industries in the top half of the rankings sell products purchased episodically. 

As always, some ISPs get higher satisfaction rankings than others. 

source: ACSI 

Some industries that once were low ranked have improved. Airlines provide an example. In 2007 the airline industry had a ranking of 63. Today airlines have a rank of 75. That is very close to an all-time high for airline ACSI scores. 

Over the last four decades, few connectivity industries have improved much, though mobility service has to be the segment that climbed the most. Now scoring about 73, mobile phone service has improved since ACSI began tracking the industry in 2004.