"Winner Take All" or "Winner Take Most" for Future Telecom Markets?

Referrals to content sites provide a clear illustration of the “winner take all” nature so often seen in application markets. On the Parse.ly network, for example, more than 80 percent of referrals are generated either by Facebook or Google sites.

source: parsely.com

So far, telecom markets have developed since dergulation as "winner takes most" structures. Unlike the structure of the Parse.ly referral market, most mobile markets (and mobile markets are most of the telecom business, these days) feature three to four dominant providers.

That is not as concentrated as the Parse.ly referral market, but still concentrated. The big issue now is whether a stable long-term pattern requires a reduction of leading providers to just three, from the four or five pattern still prevalent in most markets. 

A few countries have the opposite problem, suggesting that there is room for serious debate about whether three contestants really is enough to encourage robust competition on a sustainable basis. In South Korea and Japan, for example, policymakers seem to want to create room for a fourth mobile provider, though all efforts to do so, so far, have failed. 

The point is that it does not seem telecom markets are that different from most others. Over time, "digital" product markets seem to establish "winner take all" structures. And telecommunications now is a "digital product."

Are OTT Video Service Churn Rates High or Low?

How one defines “churn” radically affects one’s calculation of customer churn rates. And Parks Associates counts churn at least two different ways. One method suggests low churn, the other suggests high churn. Take your pick.

By one methodology, which compares churn as a percentage of lost customers across the whole base of U.S. broadband homes, churn rates are low for a consumer service, amounting to annual losses of about 20 percent, or monthly rates of less than two percent.

Using the other methodology, one more common for churn measurements--the percentage of the current subscriber base who drop service in a month or year--OTT video service churn is low for Netflix, relatively low for Amazon Prime and high for Hulu Plus.

In 2015, Netflix lost about nine percent annually; Hulu Plus about 50 percent for the year; Amazon Prime about 19 percent over 12 months.

On a monthly basis, that suggests Netflix churn of about three quarters of one percent a month--quite low for a consumer service of any type. Hulu Plus appears to be about four percent a month, high by consumer service standards.

Amazon Prime is about 1.5 percent a month, acceptably low for a consumer service.

Mature access services, especially mobile and triple play services, can have churn of less than one percent a month, by way of comparison.

source: Parks Associates

source: Parks Associates

Parks Associates' OTT Video Market Tracker shows 33 new OTT services entering the U.S. market in 2015. Among all U.S. broadband households, 64 percent of U.S. broadband households subscribe to an OTT video service, up from 59 percent in 2015.

Google, Facebook Drive More than 80% of Referrals to News Sites

Referrals to content sites provide a clear illustration of the “winner take all” nature so often seen in application markets. On the Parse.ly network, for example, more than 80 percent of referrals are generated either by Facebook or Google sites.

source: parsely.com

Fixed Network Now Drives Just 7% of Verizon Operating Income

If you wanted a one-sentence description of how the U.S. fixed network business has been transformed over the last 15 years, here it is: “Wireline now accounts for less than 30 percent of Verizon’s total operating revenues, down from 60 percent in 2000, and less than seven percent of our operating income,” noted Verizon Communications CEO Lowell McAdam.

In the fiscal year ended December 31, 2014, Verizon generated $127.1 billion of total revenues.

Fully $87.6 billion revenues, 69 percent of total, came from the mobile segment.

Verizon generated $38.4 billion revenues, 30 percent of the total, from fixed networks. Verizon generated $18.0 billion (47 percent) from mass markets, $13.7 billion (36 percent)  from global enterprise, $6.2 billion (16 percent) from global wholesale, and $0.5 billion (one percent) from other operations.

source: Revenue and Profits

As often is the case, revenue contribution and profit contribution can vary. Fixed network operating income margin was 4.3 percent in first-quarter 2015, Verizon reported. In first-quarter 2015, mobile segment operating income margin was 35 percent.

source: Market Realist

Typical Uber Ride Generates 19 Cents of Earnings Before Interest, Taxes, Employee Compensation

You might have heard more than a few Uber drivers grumble about the 25 percent share of gross revenue Uber takes from each ride sold to a passenger.

In February 2016 Uber earned 19 cents per ride in the United States,  according to company documents.

Of the 25 percent share of a typical fare, most goes to antifraud efforts, credit-card processing, customer support, marketing, and software development.

That 19 cents estimate does not include such matters as interest, taxes, or equity-based compensation for employees.

Google Fiber Unveils New 25 Mbps Symmetrical Service in Kansas City, for $15 a Month

How much should 25 Mbps symmetrical service cost? $15 a month, Google Fiber essentially says, as it introduces a new “budget” plan for residents in Kansas City, Mo. and Kansas City, Kan. Google Fiber neighborhoods with low rates of Internet connectivity.

In some U.S. markets, 20 Mbps to 25 Mbps from a telco or cable TV company can cost $60 a month. 

The new Google Fiber plan is offered without any data caps, no application process or contracts and no equipment rental and no construction or installation fees.

Broadband plan coverage areas are determined using publicly available data from the U.S. Census and Federal Communications Commission (FCC).

The new service will be available in Kansas City starting May 19, 2016.

Ingenu Launches IoT Network in Dallas

Ingenu, a supplier of a machine-to-machine network platform, has launched in Dallas. The Machine Network is intended to provide Internet of Things (IoT) connectivity to the region and will cover approximately 2,116 square miles, serving a population of more than 4.4 million people.

The Machine Network is powered by Ingenu’s RPMA (Random Phase Multiple Access) technology, designed to provide robust, reliable connectivity for M2M apps.

Application development for the Machine Network is currently underway with partners such as Dallas-based, Plasma, said to be a leader in enterprise digital transformation and IoT.

Plasma is partnering with Ingenu to support various Smart Cities initiatives.

Plasma supplies an enterprise-grade IoT platform that allows rapid prototyping and deployment of IoT and mobile solutions, Ingenu says.

If you have been in the communications or computing business long enough, you know that not all proposed technology platforms and standards succeed in the market. But it always is hard to pick winners and losers early on in the development of any coming technology and business.

In fact, as sometimes is the case, the general shift from proprietary or special-purpose networks to general purpose networks also occurs at the same time as special-purpose networks also are proposed.

It is too early to have complete assurance of how the overall IoT market and business will develop, which industries will deploy first, which deployments will prove to add the most value, which firms will emerge as leaders in the various markets or which access networks will be significant.

source: The Connectivist

Facebook's 10-Year Roadmap Includes "Connectivity" as Among Top 3 Technology Directions

Here’s another illustration of growing changes in the Internet ecosystem that pose a huge challenge to incumbent access providers.

Facebook has released some details about its 10-year roadmap. On the agenda: virtual reality, artificial intelligence and “connectivity.”

Connectivity includes drones, satellites, terrestrial solutions, telco infrastructure.

A couple of key points: Internet access no longer is the exclusive province of telcos, cable TV, fixed wireless or satellite providers of Internet access. Many other entities, including ISP businesses set up by local government, independent ISPs and application provders, are going to bundle Internet access with their other products.

That is not so unusual. Broadcast TV, broadcast radio, linear video entertainment, fixed and mobile voice and text messaging are apps or services that bundle access as part of some other product.

When we say that content is going to a growing part of the access provider business, that is only a reflection of the broader historic reality for many managed services.

Over time, it is highly likely that content and apps will be bundled with access for consumption by consumer users and customers, as connectivity, storage, computing or other services often are bundled as part of specialized business customer networks.

Source: Facebook

The other observations are that participants within the Internet ecosystem often expand into adjacent roles within the ecosystem, while it also is proving to be easier to move “down the stack” than “up the stack.”

When access providers add managed services or Internet apps, or app providers add access, those are examples of firms taking on additional roles within the ecosystem.

When app or service providers become device suppliers, or device suppliers become access providers, those are additional examples of ecosystem participants moving into adjacencies.

When app or service providers build their own devices, those are examples of moves into adjacencies.

Sometimes the moves are driven by strategic reasons, such as maintaining or gaining dominance in a market segment.  In other cases the moves might simply reflect the ability to grow gross revenue or reduce costs.

Ericsson has built an interesting business operating service provider networks, for example, essentially allowing access service providers to outsource operations.

The long term implications for access providers are clear enough. Staying within the one assigned ecosystem role (access) is going to be difficult, or dangerous, as time passes and more suppliers in other parts of the ecosystem decide that access is something they should be providing.

Half of Free Basics Users Buy Mobile Internet Access Within 30 Days

Half of all “Free Basics” users buy a data plan from their mobile service provider within 30 days of trying Facebook's free service, said Emeka Afigbo, Facebook's manager for product partnerships for Middle East and Africa.

Mobile operators receive no compensation from Facebook or Internet.org, and give away mobile Internet access to Free Basics users.

The perceived--and apparently real--upside is the chance to acquaint new users to the value of mobile Internet access and key Internet apps themselves.

Free Basics is offered in 37 countries, though Egypt and India have banned the service.

Some 25 million people have used the service, with six million users added since January of 2016.

If the conversion rates everywhere were to reach levels in the Middle East and Africa, potentially 12 million new mobile Internet customers would have been added in developing nations since Free Basics launched.

But there is far to go. Free Basics now is available to 1.67 percent of the potential population of the target nations representing around 1.4 billion people.

Other ISPs might be interested in a graphic Facebook has released about its 10-year roadmap. On the agenda: virtual reality, artificial intelligence and “connectivity.”

Connectivity includes drones, satellites, terrestrial solutions, telco infrastructure.

Source: Facebook

What 4G Tells Us About Future of 5G

If our experience with 5G is anything like 4G, business model assumptions will include:

• Build it and they will come
• Cost to deliver a retail gigabyte to customers will drop
• Revenue for delivering a retail gigabyte will not increase
• Any new “killer app” will be discovered: we do not know what it is, yet
• “Dumb pipe” still will drive most of the growth, but managed services will drive most of the gross revenue (globally).

The mobile industry has tended to replace its network by a “next generation” network about every decade and nobody seems to think that will end.

Though the phrase “build it and they will come” (hope, in other words) is a discredited business model in many quarters in the wake of the Internet bubble around the turn of the century, it rather neatly captures actual reality in the mobile business.

Every next generation network has, in fact, created its own demand. So, in truth, a change in demand by consumers almost never is the actual business rationale for shifting to a next generation network. More often, next generation networks are preferred because they help mobile operators run networks that are more efficient, and therefore potentially more profitable.

The formal justification for building the next generation network often hinges on new revenue sources or ability to raise prices. Our experience with 3G and 4G is that this expectation was largely unfulfilled.

As a practical matter, whatever is possible as a matter of technology platform, prices have been falling, largely due to increased competition. So the expectation that a next generation network leads to an increase in retail prices per gigabyte is going to be dashed, if in fact anybody actually thinks that can happen. So revenue per delivered gigabyte will remain flat, or possibly even drop a bit.

But the supplier cost to deliver a gigabyte will drop with 5G, perhaps not by an order of magnitude, but by hundreds of percent. One might well argue the actual driver for 5G is benefits for suppliers, not end users, even if end users are going to experience more value.

That is what has happened for any number of earlier mobile technology innovations, including signaling systems, the switch from analog to digital switching, better air interfaces, better radios, faster backhaul, signal compression, processing and storage costs.

Many now say 5G will enable many new Internet of Things applications, though the emergence of alternate networks focusing on low-bandwidth, low power consumption networks designed to support sensors suggests IoT might be a significant new market, if not necessarily an application unique to the 5G network.

Other big new applications might yet emerge from 5G. But that was said about 3G, and did not happen until relatively late in the platform lifecycle. Internet access, and especially use of video has emerged as the “killer app” for 4G.

We cannot really predict what might emerge from 5G (even if hopes are pinned on IoT).

And even if it will make sense, and likely happen, that mobile operators actually create some new managed services delivered over 5G and 4G networks, most of the revenue, and the foundation of the business model--in terms of revenue--will still come from “dumb pipe” Internet access.

Managed services such as voice and text messaging still will be significant, and still drive most of the gross revenue in many markets.

But revenue growth will hinge on Internet access revenues. And that, by definition (because of current interpretations of network neutrality rules), means “dumb pipe.”

source: ITU

Global Voice Revenue is Dropping, As Total Revenue Grows

Any way one looks at the matter, global telecom service revenues derived from voice are shrinking, even as global service revenue grows, the result of more subscribers and new revenue sources, especially mobile data and Internet access.

source: ITU

Huge! Verizon Communications to Build FiOS in Boston

This is huge: Verizon Communications, which had halted its FiOS program in 2010, is going to rebuild Boston with FiOS. Boston was among a notable number of major Verizon markets that Verizon concluded it would not upgrade for fiber to the home services. In addition to Boston, Buffalo, N.Y. and Baltimore are some of the larger cities that did not get FiOS.

But Verizon now says it will build FiOS citywide, spending more than $300 million over six years to do so.

The network will offer speeds up to 500 Mbps.

Initially, the project will begin in Dorchester, West Roxbury and the Dudley Square neighborhood of Roxbury in 2016, followed by Hyde Park, Mattapan, and other areas of Roxbury and Jamaica Plain.

At least in part, the change might be the result of an expedited permitting process that has allowed other Internet service providers, such as Google Fiber, to build faster and at lower cost.

It is possible that Verizon has concluded the business case for FiOS now has changed for additional reasons beyond the lower make-ready costs.

The strategic rationale might now also be more significant, irrespective of any improvements in the payback model. Use of the fixed network to offload mobile data traffic arguably is more important.

Use of the fixed network to support mobile backhaul operations from small cells might be a new driver on operating cost front.

Also, by now it is clear that high speed Internet access is the anchor service for a fixed network, and that Verizon must be competitive with cable TV offers and those of other ISPs if the economics of a fixed network are to work at all.

Stranded Assets Affect All Key Parts of the Telco Business Case

Stranded assets are a huge business model problem whenever a monopoly communications business becomes competitive. Stranded assets--network facilities that drive zero revenue--also directly related to market shares, gross revenue, profit margins, marketing, capital investment and operating costs.

Which is to say, the key drivers of the business. Though well understood now, the fundamental change in access network economics was not immediately and universally understood.

Consider any one-product service provided by a single provider with 80 percent installed base or take rate. Add one competent new supplier, and theoretical “maximum” share for each provider drops to 40 percent. Add a third competent supplier and theoretical maximum share drops to 33 percent.

That change alone would destroy the original economics and business model. Triple play now is foundational because it directly addresses competitive market dynamics: three products can be sold to a smaller number of potential customers.

If each unit sold (for example) is about $40, then per-account revenue could be $120, not just $40 if a single product were the only supplier option.

But stranded assets remain a problem, which is why Google Fiber and so many other gigabit providers now pre-qualify areas for network upgrade or build based on expected demand. That avoids expensive investment where returns are likely to be low or negative.

In that sense, telco access network ubiquity might actually be a negative, not a positive.

“The ILECs’ low cash flows reflect the continuously increasing cost of sustaining a ubiquitous network that is now serving roughly a third of the lines for which it was engineered,” argues Anna-Maria Kovacs. In other words, 66 percent of the deployed network does not actually generate revenue.

In other words, when the percentage of customers on the network is about one out of three locations, most of the network access investment is stranded. Dead. Inert. Not producing revenue.

“At the 2015 ILEC penetration level of 35 percent of peak, total network cost per remaining subscriber has essentially doubled and the networks passed the inflection point beyond which penetration losses result in catastrophic cost increases,” says industry analyst Anna-Maria Kovacs,  visiting senior policy scholar at the Georgetown Center for Business and Public Policy.

By way of contrast, cable TV networks still are operating in the flat part of the cost curve. That is a function of having paying customers on perhaps 80 percent of passed locations.

Metro access providers and business communications specialists such as Zayo and Level 3 have more ability to target investment. So, less risk of asset stranding.

Perhaps you can imagine a scenario where telco stranded asset problems lessen. In an optimistic scenario, telcos create or discover some new unique application that drives take rates back over 50 percent.

In a negative scenario, stranded assets are not an issue in the case of market exit. Verizon, for example, with a relatively small fixed network footprint among tier one providers, might welcome a chance to exit the fixed network business entirely, as a facilities-based provider.

That doesn’t mean Verizon would have no chance to sell services to fixed network customers of its choosing. It would simply lease those assets, probably at favored rates as part of an asset divestiture.

Lack of willing and able buyers, plus regulatory opposition, make that an unrealistic possibility for the moment.

The stranded asset problem is not going to disappear, or lessen, in terms of intensity. The only issue is what ultimately is done to “make the problem go away.”

If Low Cost Provider Wins, Cable TV Beats Telcos

In any competitive market, including consumer, small and medium business and parts of the enterprise communications markets, the low-cost provider tends to win.

Telcos are anything but the low cost providers.

Consider free cash flow margin of telcos and cable TV companies in the U.S. market, for example. Where Verizon fixed network margins are about eight percent, and AT&T margins about 12 percent, Level 3 Communications gets 16 percent, Cogent Communications 20 percent and Zayo about 22 percent cash flow margins.

Charter Communications has cash flow margins of about 13 percent, Cablevision Systems Corp. and Time Warner Cable get 18 percent, while Comcast earns 27 percent.

You might argue that Level 3, Cogent and Zayo have an advantage, and they do. They can pick and choose their markets and serve enterprises and other carriers, rather than consumers.  That tends to contain cost and boost gross revenue per account.

It also seems to lead to higher cash flow margins, though not as high as the cable TV providers are able to produce.

source: Anna-Maria Kovacs

The issue is how market dynamics change over time. A safe prediction is that cable TV operators, now turning  their attention to the enterprise customer segment, will gain share, likely at the expense of telcos.

That should, all other things being equal, lead to even more pressure on AT&T and Verizon fixed network profit margins.

Is T-Mobile US Disruptor Role Right for Netherlands?

Necessity might be the mother of invention, but having a role model helps. So it is that Deutsche Telekom, which tried and failed to sell its Netherlands mobile business, possibily is  looking at replicating the T-Mobile US approach in the Netherlands.

As in the U.S. market, DT wanted to sell its business, and could not do so. But DT apparently now believes it can follow the U.S. playbook and grow its Netherlands mobile business by being

Disruptive.

Aggressive pricing and unique promotions would be key strategies. Deutsche Telekom might ultimately decide not to pursue that tactic. But something dramatic is required, even if all DT wants to do is eventually create value so the asset can be sold, something that also would follow  its U.S. strategy.

Deutsche Telekom lost more than a million mobile subscribers in the Netherlands in the past three years, but is no late entrant, having operated in the Netherlands for 16 years.

Perhaps the key difference with the U.S. market is that T-Mobile US itself become the disruptor. In the Netherlands, that role already has been usurped by Tele2 AB, as Iliad’s Free Mobile has taken that role in the French market.

source: Tele2

source: Bloomberg