Friday, November 30, 2018

What AT&T Revenue Segments Suggest About its Strategic Challenges

One way of gauging the strategic value of various AT&T revenue segments is to examine either revenue or earnings contributions.

The Mobility business unit represents about 50 percent of AT&T’s adjusted cash flow (EBITDA). WarnerMedia represents about 17 percent of the company’s revenue and adjusted EBITDA.

Business Wireline represents about 17 percent of the company’s adjusted EBITDA. Entertainment Group represents about 15 percent of the company’s adjusted EBITDA.

In terms of revenues, mobility represents 40 percent, entertainment group 26 percent and business wireline about 15 percent of total quarterly revenue of $45.7 billion.

So 99 percent of cash flow comes from those four revenue segments. But what might really stand out is the 15 percent contribution from AT&T’s landline voice, video distribution and internet access products (the triple play suite). These days, consumer internet access, voice and entertainment video (recall that AT&T is the largest U.S. subscription video provider), contribute relatively small amounts of cash flow.

Also, keep in mind that DirecTV, for the moment, is delivered primarily by satellite, and likely represents $8.5 billion in revenue. So it is possible that consumer landline services now contribute only about seven percent of AT&T revenue.

That suggests one important strategic implication. Unless you believe AT&T can materially grow its landline voice and internet access revenue by investing heavily in optical access, and thereby taking significant share in the access business, the capital is arguably better spent elsewhere. But there is a caveat.

Some of us would argue that the ultimate fate of the present DirecTV business is its transition from satellite to over-the-top streaming delivery. If so, then AT&T has to ensure that its fixed network access lines can handle the data load of streaming DirecTV.

It is a delicate matter, as many believe total revenue from OTT delivery is going to be less than linear delivery. So investments in next generation networks will be made to support potentially less revenue than presently is earned.

But that is not a new strategic issue. Many fixed network executives have essentially operated on the assumption that such investments essentially must be made so “you keep your business,” and not because revenue upside is so great.

Such challenges often are downplayed by suppliers, service providers, financial analysts, consultants and others with business interests in the industry. These days, it is becoming more common to hear frank discussion, however.

“Operators have a choice,” Phil Twist, Nokia VP said. “Do you want safety and security, or the risk of new growth?” Twist characterized “safety” as a prison; growth as entailing “uncertainty.”

Others characterize the state of the industry as “unraveling” and “desperate.”

To be sure, virtually nobody but AT&T’s key competitors are likely happy about the huge debt load AT&T has taken on to diversify its revenue sources. But it is a rational, if controversial argument, that AT&T is hemmed in--in terms of revenue growth potential--in both its fixed and mobile business segments.

Committed to a financial strategy built on ever-increasing dividend payments, AT&T essentially has no choice but to risk expansion beyond connectivity services, as its opportunities to take significant market share in mobile or fixed network connectivity services in its core markets are slim to none.

Wednesday, November 28, 2018

Streaming Services With Biggest Customer Bases Have Lowest Churn

Customer churn is a big concern for any supplier of a consumer application or service, and it generally is recognized that much churn happens in the early days of any engagement with a new customer. Conversely, churn tends to be quite a bit lower for accounts with some longevity.

That is fairly easy to explain. New customers without prior experience with any given service or service provider, appliance or app have a learning curve to climb before most of the value of the product is grasped. For many software as a service products, most of the churn happens in the first 90 days.


So it is probably not surprising that U.S. customers of streaming video services seem to churn least on services that have the longest tenure and the greatest number of accounts.

Basically, most of those customers have had time to decide whether they value the services enough to keep paying for them, have concluded that the value-price relationship is satisfactory and have learned how to navigate the services.

Many of the newer or smaller services arguably have a higher percentage of brand-new users, with the obvious risk of higher abandonment early in the relationship.
source: Juniper Research

Tuesday, November 27, 2018

Double Down on Connectivity or Diversify?

What tier-one communications service providers should do about revenue growth is a matter of huge disagreement. Some favor a “stick to your knitting” approach. Others believe connectivity growth prospects are so limited that movement into new lines of business, to create new revenue sources, is imperative.

It is not hard to find critics of the Time Warner or DirecTV acquisitions that argue AT&T should have invested in its fiber to home capabilities or its 5G network. Many prefer the “pure play” approach taken by Verizon or T-Mobile US, for example.


That might or might not be strategically wise. Globally, revenue growth in the mobility and fixed network segments is flat or slowing. Significantly,  revenues in developed markets are declining, and have been since perhaps 2011.


Nor has AT&T or Verizon been able to budge market share too much since about 2012. To be sure, where AT&T had 17 percent share in 2000, it had 31 percent share in 2012. Where Verizon had 25 percent share in 2000, it had moved up to 35 percent share in 2012, as both service providers made big acquisitions to bulk up.


Verizon’s buys included the portion of Verizon Wireless purchased from Vodafone, but also Cellco and Alltel. AT&T grew by acquiring Cingular and Dobson.




Since 2011, AT&T and Verizon market share, as measured by accounts, has been flat. So there are reasons to believe that even alternative investment in 5G would not move the revenue needle very much.




The other argument is that there are very few ways incremental cash flow or revenue from investing in the fixed network (more fiber to home) or investment in faster 5G.


In 2018, U.S. cable companies had 65 percent internet access market share, More importantly, cable companies have been getting virtually all the growth for several years.




To be sure, it is rational for any equity analyst to doubt the wisdom of the additional debt AT&T took on to buy DirecTV and Time Warner.

But it is not clear how much revenue upside exists for AT&T in its fixed networks business, especially given cable operator market leadership.  

Access Network Traffic Gets More Asymmetrical

Though an argument can be made that a shift to symmetric internet access will lead to applications being created to take advantage of bandwidth symmetry, present traffic trends do not favor heavy internet service provider investments in symmetrical bandwidth for consumer connections.

With the caveat that a big shift to peer-to-peer distribution of consumer entertainment would have a big effect, growing internet distribution of content has intensified the asymmetric pattern of internet access traffic. That matters since video dominates core, metro and access network traffic load.

Globally, IP video traffic will be 82 percent of all IP traffic (both business and consumer) by 2022, up from 75 percent in 2017.

Upstream traffic has been slightly declining as a percentage of total for several years, Cisco says. It appears likely that residential Internet traffic will remain asymmetric for the next few years, at the very least.

Global IP video traffic will grow four-fold from 2017 to 2022, a CAGR of 29 percent. Internet video traffic will grow fourfold from 2017 to 2022, a CAGR of 33 percent.

The sum of all forms of IP video, which includes Internet video, IP VoD, video files exchanged through file sharing, video-streamed gaming, and video conferencing, will continue to be in the range of 80 to 90 percent of total IP traffic, Cisco says. Globally, IP video traffic will account for 82 percent of traffic by 2022.

User-generated content and security camera video have not affected the asymmetric traffic pattern, downstream and upstream.

Global IP traffic by application category
white-paper-c11-741490_12.jpg

Internet of Things (Machine-to-Machine) Drives Connected Device Growth

Internet of things (machine-to-machine) devices will underpin global connected device growth through 2022, Cisco says.

Machine-to-machine (M2M) connections (internet of things sensors, cameras, monitors) will be more than half of the global connected devices and connections by 2022, Cisco says.

The share of M2M connections will grow from 34 percent in 2017 to 51 percent by 2022. There will be 14.6 billion M2M connections by 2022.

source: Cisco

The number of devices connected to IP networks will be more than three times the global population by 2022, Cisco now predicts.  There will be 3.6 networked devices per person by 2022, up from 2.4 networked devices per capita in 2017.

Globally, there will be 28.5 billion networked devices by 2022, up from 18 billion in 2017, Cisco forecasts.




Monday, November 26, 2018

Rural Internet Access is Tough

Any reasonable observer would say that a 1 Mbps internet access speed costing $80 a month is less than adequate, as it costs in Fort Yukon, Alaska.

The problem is that current solutions to boost access speed are not commercially viable, owing to geographic isolation and the small number of potential users.

Fort Yukon has a total population of 547, and is served by three internet service providers (one using satellite, one using fixed wireless, one using digital subscriber line).

There are an average of 2.6 persons per home, implying a total of 210 homes, of which 38 percent are occupied seasonally. Some estimates suggest full-year occupancy is 40 percent.

So assume full-year occupancy is about 62 homes to 84 homes are the universe of home broadband services.

Assume 62 percent of homes buy internet access. That implies a universe of 38 to 52 potential customers. Recall that Fort Yukon is isolated. Fort Yukon, located northeast of Fairbanks, is not served by the GCI TERRA network.



Terrestrial for Every Rural Region in Alaska (TERRA) project, a 3,300-mile rural high-speed broadband network delivering high-speed broadband access to 45,000 Alaskans in 84 communities that are spread across an area the size of Texas.

That means the only backhaul is by satellite, with all that implies for cost and capacity, in a region with very low population density. The estimated cost of building a mile of optical fiber middle mile facilities is $120,000. Microwave middle mile might cost $26,000 per mile. While fiber middle mile might support 10 Gbps, microwave might support less than 1 Gbps.

It is at least 138 miles from Fairbanks to Fort Yukon. Low earth orbit satellite constellations eventually will be a viable solution for middle mile, but not using terrestrial facilities. A microwave connection could cost $3.6 million. A fiber line might cost $16.6 million.

There is no payback model, using terrestrial facilities.

Some Industry Professionals Believe Connectivity Will be Less Important in 5 Years

One always should be on the lookout for anomalous data. So it is with a survey of 550 industry executives from 11 countries, conducted by Osborne Clarke for The Economist Intelligence Unit. Note that executives in the digital business, energy and utility and financial services verticals say they believe “connectivity” will be less important in five years.

Conversely, professionals in real estate/infrastructure and transport/automotive industries believe connectivity will be more important in five years. So what to make of those results? Some argue that real estate and transport traditionally have been slower adopters of information technology.

Also, financial services and digital businesses have been early adopters.


The results might also indicate that, in some industries, connectivity itself is less an issue, while other elements of business success are viewed as more significant.

NASA Spacecraft Lands on Mars

Never get bored with a Mars landing!



Is 5G as Big as the Internet?

Is 5G going to be as big an innovation as was the internet? It is too early to tell. But here’s an exercise, taking a range of claimed Internet benefits and simply substituting the word “5G.”


“Unlike its predecessors though, Internet is a technological paradigm shift, akin to the shift from typewriter to computer. And it isn’t just a network. Internet will become the underlying fabric of an entire ecosystem of fully connected intelligent sensors and devices, capable of overhauling economic and business policies, and further blurring geographical and cultural borders. It will be capable of delivering at every rung of the ecosystem’s ladder, and will provide seamless, continuous connectivity for business applications.”

Image result for internet economic impact


  • “5G will be a fundamental shift in the way communications networks are managed and operate; a marked distinction from previous mobile generations
  • 5G is no longer just about communications, but connectivity solutions. Data rate and capacity have traditionally been key metrics, but Internet significantly improves other metrics such as latency and coverage. This allows enhanced connectivity for a much wider range of applications across all sectors and industries.
  • Major mobile operators will have to focus on an expanded value chain to sustain high growth, while key players from other sectors will find themselves integrated into the wider telco industry.
  • 5G will become a universal connectivity platform, and will have a profound impact on the way devices and people communicate and interact.”


  1. “5G will turn connectivity into a platform. With Internet, wireless access networks will go beyond pipe, providing seamless, ubiquitous, and limitless connectivity for all people and all things.
  2. Everything will go 5G. Right now, most things are offline by default, and most electronic devices are not connected. With 5G, being online and connected will become the default for everything.
  3. The world will go all cloud. Supercharged with 5G, the cloud will provide massive computing power with instant transfer speeds and near-zero lag. This will make intelligence on demand available for everyone, everywhere. New business models like Cloud X – where devices are boosted by inexhaustible cloud-based resources – will begin to emerge.
  4. Devices will be redefined. With AI support across devices, network, and the cloud, devices will go from plug and play to plug and think. They will understand users better – able to actively predict our needs, not just passively respond to commands – and interact with us in more natural ways.
  5. Experience will flow seamlessly. With existing networks, our online experience is fragmented from one scenario to another. When all things are 5G and cloud-based, experience and content will flow seamlessly through time, space, and devices for a truly holistic experience across all scenarios.”


The point is that many potentially huge innovations, such as internet of things, artificial intelligence, virtual reality and autonomous vehicles are enabled by a complex of technologies and platforms. Internet is part of that complex, but ultimately just a part.


That is not to say 5G will fail to play a role in creation of a big new platform for apps, services, use cases and revenue models and sources. But separating the impact of 5G specifically, from all the other important changes, is a perilous exercise.


The internet itself, AI, VR, semiconductor advances, use of millimeter wave spectrum, device advances or other technologies might play key roles. Or not.


The big point is that all the incoming innovations likely are not solely driven by 5G. If successful innovations happen, they might not even require 5G. It just is impossible to say, right now.

Sunday, November 25, 2018

Are Telecom Prices Too High?

One argument we always hear about any communications service is that it is “too expensive.” That clearly is true in developing or lesser-developed nations, where mobile internet access can cost 14 percent of gross national income per person. In developing markets, mobile internet access can cost about six percent of GNI, per person.

Consider mobile internet access. In developing countries, mobile internet access costs about 0.7 percent of gross national income, per person, according to International Telecommunications Union data.

Much the same story holds for fixed network internet access as well. Fixed network internet access in developed nations costs less than one percent of GNI per person, perhaps 11 percent of GNI per person in developing nations and as much as 30 percent of GNI per person in lesser-developed nations.

One rather impressionistic bit of evidence that prices are not excessive might be gleaned from industry profits. On a global basis, revenues dropped by about four percent between 2014 and 2015, for example.

Developed market revenues have been dropping since about 2012, according to ITU statistics.
 


Half of Global Fixed Lines Support Internet Access

In 2017, about half of all fixed network lines globally supported voice. The other half (the growing half) consisted of internet access lines, according to International Telecommunications Union data.

Fixed Network Lines, Voice and Broadband, Lines in Millions

2014
2015
2016
2017*
Fixed-telephone subscriptions
Developed
503
490
479
471
Developing
592
556
524
500
World
1,095
1,046
1,004
972
LDCs
9
8
9
9
Fixed broadband subscriptions





Developed
354
370
382
392
Developing
377
472
535
588
World
731
842
917
979
LDCs
5
7
8
10





Total Lines




Developed
857
860
861
863
Developing
969
1028
1059
1088
World
1,826
1,888
1,921
1951
LDCs
14
15
17
19





Percentage Broadband Lines
Developed
41%
43%
44%
45%
Developing
39%
46%
51%
54%
World
40%
45%
48%
50%
LDCs
36%
47%
47%
53%


That illustrates one key facet of fixed network business models: increasingly, the fixed network is a platform for internet access, less so voice. The other trend is that more of the value of the fixed network is being generated by a range of data services including wholesale and retail lines, as well as services for enterprises.



"Tokens" are the New "FLOPS," "MIPS" or "Gbps"

Modern computing has some virtually-universal reference metrics. For Gemini 1.5 and other large language models, tokens are a basic measure...