5G Might Truly be Different

There is one way 5G might be quite different than all prior generations of mobile, and not for reasons directly related to network performance. All prior generations of mobile service had business models based on humans using phones. But 5G is the first mobile network where most of the actual subscriptions might be used by computers and sensors.

Ericsson points out that consumer-related service revenues will have an annual growth rate of 0.75 percent through to 2030. Providers of apps and services in other parts of the information and communications technology, on the other hand, will see compound annual growth rates (CAGR) of close to 12 percent over the same time frame. 

Ericsson believes connectivity service providers could address as much as 18 percent of total information and communication technology revenues, representing about $700 billion. 

By definition, those information and communication technology opportunities will come from enterprise customers, not consumers. And that could be the biggest strategic change caused by 5G. 

Unlike all prior mobile generations--might feature incremental revenue sources and growth mostly coming from enterprise services, not consumer mobility that has driven growth since the 2G era (the first generation of mobile arguably having been adopted mostly by businesses). 

As Syniverse sees matters, for example, “5G is not dependent on consumer subscription of services like the wireless generations before it.”

“With decreasing average revenues from consumers in developed markets, U.S. mobile operators (60 percent of survey respondents) are shifting the focus to enterprise use cases to make money on the evolution to 5G,” says Syniverse. 

Nearly 60 percent of poll respondents say that 5G will swing their company’s focus to enterprise ecosystems, while 77 percent of respondents believe services such as network slicing supplied by 5G core networks. 

Mobile industry executives believe most of the new revenues from 5G will come from enterprise services including network slicing. Some service providers expect revenue lift from enterprise customers of perhaps five percent to 10 percent within just a few years. 

In fact, "In the 5G era, telcos will earn 70 percent of their 70 percent of their net revenue from enterprises” says Sanjay Kaul, Cisco head of Asia Pacific and Japan service provider business.

U.S. Streaming Video Providers Debt-to-Cash Flow Leverage

With the caveat that debt loads have been acquired for diverse reasons, firms operating in the streaming video subscription business have debt-to-cash flow leverage around the 3.3 times to 6.7 times  range (AT&T the former, Netflix the latter). 

At least some of that debt relates to investments in content. 

source: Moody's

Churn Rates are Non-Linear

Nearly 45 percent of respondents of 2,500 people surveyed by EFTM about service provider switching behavior said they had not switched telcos in more than five years. That is not as unusual a figure as you might think. 

source: EFTM

In the U.S. market, churn rates have been steadily dropping. For at least three of the four largest U.S. mobile service providers, churn typically is perhaps 1.5 percent monthly, sometimes less. 

Roughly speaking, a churn rate of 1.5 percent a month works out to be about 18 percent annual churn. The caveat is that churn is non-linear: high rates in the early months of an engagement, then lower churn over time. 

source: Profitwell

The key concept, when evaluating churn, is to recognize that most customer churn happens relatively soon in a supplier relationship. Note that churn rates are very high early in the relationship, but becoming relatively stable after about six months. That is the reason the Australian data collected by EFTM, which might lead you to assume a 20-percent rate effectively means all the customers acquired in a particular month are gone in five years, turns out to be incorrect. 

source: evergage

In each new customer cohort, churn rates drop dramatically after perhaps two months, so the average churn rate is not reflective of the lifecycle churn rate. 

After five years, at  20 percent annual churn, about 65 percent of the cohort have left, but 35 percent remain. More significantly, after 10 years, perhaps 10 percent of the original cohort remain  customers. 

source: Chaotic Flow

So the results of the EFTM survey are not unusual. After five years, at 20 percent annual churn, or about 1.7 percent monthly, one would expect 35 percent of customers to remain, where the EFTM survey found 45 percent remaining. 

In my own case I have had a continuous relationship with one mobile service provider for more than 20 years. Generally speaking, higher average revenue accounts have less churn than lower ARPU accounts. Customers on contracts churn less than customers who can leave after any particular billing period. 

And the customer on-boarding process can be a key enabler of lower churn for new customers, who are not only learning how to use a service or a device offered by a particular provider, but might also be testing whether a particular plan, coverage or device  is right for them.

Mobile Product Life Cycle in India

Here’s one way of visualizing a product life cycle at work. Over a seven-year period in India, mobile data customers will nearly completely switch product choices from either 2G or 3G to 4G. 

source: KPMG

Telcos Must Replace 1/2 of Legacy Revenue Every Decade

By definition, a mobile access platform has a definite product lifecycle, as the next-generation network is introduced about every decade, with a useful life of about 20 years. 

But that is true for virtually any product. The implications for connectivity services executives is quite clear: they must plan to replace every new product they introduce. My own rule of thumb is that service providers must plan on replacing about half of current legacy revenues every decade.

source: University of Michigan Information and Technology Services

Illustrated on this chart, for example, are a number of products including:

Bring Your Own Device (BYOD) Resources for Networks & Communications Systems

Cable Television

Cellular Femtocell

Core Network

DAS (University Owned & Managed)

DAS (Vendor Owned & Managed)

GSM, CDMA, & LTE

In-Building WiFi Networks

In-Building Wired Networks

InfiniBand

IPTV

IPv4 and IPv6

Live Event Streaming

Local Server Rooms

Modular Data Center (Custom Data Centers)

Personal Cellular Use for University Business

Smartphones & Cellular on Campus

Software Definable Networks

Software Definable Radios

Streaming Media Players

Telepresence Robots

Traditional Telephone Systems

U-M Data Centers

Videoconferencing (BlueJeans)

VoIP Telephone Systems

VPN/Remote Access

White Spaces

WiFi Calling

source: Slide Team

Can Telcos Become Platforms?

In many emerging industry segments, when executives are asked what role their firm might aspire to, they often have said they were aiming to become a platform. What has been less clear is whether established firms such as connectivity providers might likewise aspire to become platforms. 

Perhaps it matters how a “platform” is defined, and how it is different from marketplaces and retailers in general. 

Some define a platform as a business model that creates value by facilitating exchanges between two or more interdependent groups, usually consumers and producers. Using that definition, a shopping mall is a platform. 

Using that same definition, Amazon is a platform. So are Google, Facebook, and other apps of that sort. Irrespective of what the specific value is for end users, the platform brings together advertisers and merchants with potential customers. 

source: Applico

Using a different definition, Apple might also be a platform, as many considered Wintel to be in the 1980s and 1990s. 

Another way of looking at matters is that platforms feature “two-sided” business models. Such models involve an intermediary that enables exchanges between at least two distinct sets of actors. 

Examples include credit cards (cardholders and merchants); health maintenance organizations (patients and doctors); operating systems (end-users and developers); yellow pages (advertisers and consumers); video-game consoles (gamers and game developers); recruitment sites (job seekers and recruiters); search engines (advertisers and users); and the internet. 

But that definition involves a subtlety: all market transactions involve two actors, one buying, one selling; one producing, one consuming. The notion of platform impliciting involves the notion that some economic value is produced by the platform that enables such transactions at greater scale. An advertising network or social media app might provide a good example. 

Facebook enables advertisers to reach an audience, for example, at scale. But network effects, in and of themselves, are not what defines platforms. Telecom networks have network effects, but arguably are not historically platforms. 

True, traditional communication networks bring users together--as does Facebook--but not users with advertisers, or customers with retailers. That might be less true in the future. Telcos who sell video entertainment services might become platforms to the extent that they connect advertisers with audiences, while also providing services to subscribers. 

That is another twist on two-sided markets: sometimes revenue is earned from serving the interests of both actors. In the case of video entertainment, suppliers earn subscription revenues from subscribers, and also advertising revenues from retailers. In some other cases the same might happen when telcos also become support app providers trying to reach their subscribers. 

The point is that the shift from a traditional one-sided (revenue from subscribers) telco model and a platform or two-sided business model is a big change.

Google Stadia Will Stress U.K. Usage Caps

If Google Stadia, the streaming gaming service, actually does consume as much as 15.75 GB per hour when used at the highest settings, then a significant number of home internet customers in the United Kingdom are going to blow through their data allowances, Broadband Now believes.

That estimate is based on statistics from the NPD Group suggesting that the 34 million gamers play 22 hours per week on average. “If these individuals switched to using Stadia as their primary gaming platform, they would eat through even the highest data caps (usually around 1 TB, or 1,000 GB), coming in at roughly 1,386 GB monthly,” says Broadband Now. 

 source" Broadband Now

“We estimate that approximately six million out of the 34 million daily gamers would eat through their  data caps if Google Stadia becomes their primary gaming destination,” Broadband Now says.

That, in general, illustrates the business problem internet service providers face: data consumption keeps going up, but ability to pay is relatively fixed. That is why performance goes up, but average monthly bills tend to stay flat. 

Prices per gigabyte are highest in lesser developed countries, as you might guess, adjusted for purchasing power parity, but really low in developed countries, looking at cost as a percentage of gross national income per person. But the clear trend over time is for internet access costs to fall. 

source: ITU

Enterprise Revenue Trends

Why Connectivity Cannot be the Only Driver of the Connectivity Business

Ericsson offers a fairly simple argument for why big service providers have to consider moving into other areas of the information ecosystem: growth will not be found in the consumer connectivity business. 

Simply put, growth rates in the consumer communications business are forecast to grow only about 0.75 percent per year to 2030, while the broader information technology business grows at a compound annual growth rate of 12 percent per year.  

source: Ericsson

The service enabler role shows the biggest growth opportunity for service providers and includes providing digital platforms on which businesses can configure and integrate value-enhancing digital capabilities into their processes. That is a huge challenge, but offers high rewards, if connectivity  providers can create service platforms, system integration and content management roles. 

Service enablement also logically includes becoming a supplier of edge computing facilities, managing devices, software and data. 

source: Arthur D. Little

source: Arthur D. Little, Ericsson

5G: Less Impact than You Expect Now, More than You Expect Later

New technologies almost always have less impact than expected at first, and important new technologies almost always have greater impact than expected later in their adoption cycle. Get ready for that to be true of 5G as well. 

According to Gartner, after a period of building hype, 5G is about to enter a possibly-inevitable period of disillusionment that might last for a few years or more. 

Perhaps some of that explains the productivity paradox, which sometimes includes the observation that the introduction of advanced technology can lead to lower productivity for a period.

For that reason, one would be right to remain skeptical that 5G, in and of itself, will dramatically boost productivity beyond the benefits of fast 4G. The hoped-for advantages of 5G-related edge computing and internet of things use cases will require rethinking and retooling the way organizations and people work. So we might not see clear advantages from those technologies until 6G is well underway. 

That said, productivity often eventually does show up, after a decade or more. The productivity paradox was seen very prominently in the United States in the 1970s and 1980s when there was a big uptake of information technology, but productivity growth slowed down over the same period.

Labor productivity growth came down from about three percent in the 1960s to about one percent in the 1990s despite the increase in computing and information technology investment. 

source: Gartner

One possible explanation is that productivity is increasing, but we simply cannot conveniently measure it. It is difficult to quantify the value of better computers that cost the same, but increase in performance, for example.

Another possible explanation is that it takes organizations and people a while to adjust to much-better technology. In that view, organizations have to retool the ways they work before the IT investments actually can help. 

U.S. Consumers Might Reap as Much as $32,000 Each Year in Economic Value from Internet Apps, Services

The actual amount of consumer welfare from new digital products and services might be quite high, even if measures of gross domestic product value them at zero. 

Products with zero price are difficult to value. Measures of gross domestic product measure goods with prices, so any products with a zero price are not reflected in our GDP or productivity statistics. 

Although information goods have become increasingly ubiquitous and important in our daily lives, the official share of the information sector as a fraction of the total nominal GDP (about four percent to five percent) was the same in 2016 as it was 35 years earlier. That might strike you as odd, and that is precisely the problem. 

Facebook users spent 50 minutes per day on Facebook and Instagram, up from zero in 2005. The same might be noted for any number of other digital apps. They either created completely new goods that did not exist before,  or replaced and significantly improved previously existing non-digital goods. 

The average American spent about  22.5 hours each week online as of 2018, according to Erik Brynjolfsson, Avinash Collis, and Felix Eggers in an article in the Proceedings of the National Academy of Sciences, but much of the value of that time cannot be captured by normal GDP metrics. 

In many areas, such as music, media, and encyclopedias, people substitute zero-price online services such as Spotify, YouTube, and Wikipedia for goods with a positive price such as CDs, DVDs, and Encyclopedia Britannica. 

As a result, the total revenue contribution of these sectors to GDP figures can fall even while consumers get access to better quality and more variety of digital goods, the authors note.

The authors estimate that digital products actually add as much as $32,000 in actual economic value per person. 

source: PNAS

SDN is an Architecture, Not a Product

Every now and then, a promising technology either is subsumed by other platforms and technologies, or simply fails to gain traction. Gartner believes software defined networking has reached that stage, in part because its fundamental premise--separating the data plane from the control plane--now is simply the foundation of network design, not a “product.”

Network functions virtualization in the communications networking space provides an example of how that basic principle--separation of control and data planes--simply is an architectural principle for modern networking. But some had hoped SDN would abstract hardware from software in ways that would foster an awful lot more innovation in software. That arguably has not happened so much, some would argue. 

At least in the communications networking space, SDN influence has lead to NFV, where the ability to separate data and control planes is allowing service providers to operate with lower cost, using generic hardware in some cases, and control software that is more centralized than before, meaning less-complex network elements can be deployed. 

source: Gartner

Can Telcos Be Platforms?

Every fast-growing or large company, in any industry, might well aspire to become a platform, it seems. How widely platforms might be adopted, or “who can become a platform” seems less clear. Perhaps the clearest examples are provided by software or application companies such as Facebook.

Increasingly, many of the biggest businesses are software companies with platform business models. In a platform business, creators and suppliers add content and services to the platform, which draws users, which creates various monetization models.

It is not so clear that any connectivity provider has an easy road, in that regard. Operating as a platform means becoming the gravitational center of a broad ecosystem of consumers, suppliers and developers. 

The problem for any connectivity services provider is that, in the internet protocol ecosystem, which now virtually underpins all digital services and products, there is a logical separation between physical networks and devices and the apps or services people and businesses want to use. 

Under such conditions, by definition, ecosystems are built in the disarticulated apps sphere, and are not inherently dependent on the physical networks that supply connectivity. The best analogy I can think of is electricity. Think of all the businesses and revenue streams and products that are built on the assumption that electricity is available. 

Then ask yourself whether a direct business relationship must exist between any supplier of a product and the supplier of electricity. The answer, of course, is “no.” Electricity has to be provided, but there is no essential business relationship required by any others in the ecosystem who supply products using electricity.

And that, fundamentally, is the problem connectivity providers face when looking to become platforms. They simply have no actual advantages in the device, applications or value-creation roles that are not directly related to the core business of supplying connectivity. 

To be sure, that is not an inherent problem for some suppliers of infrastructure, including roads, electricity, waste water or drinking water or natural gas. Growth rates might be nil, but there is modest, if any competition, which means profits, if not high, are steady. 

Telecom, in contrast, is in what might be termed the worst of all possible worlds. Once a formal monopoly with low but guaranteed rates of return, it now is It a competitive  business with high capital intensity, significant regulation and changing consumer preferences. 

It is as if an electrical energy supplier discovered that its customers were, in large numbers, disconnecting from the grid and creating their own energy. Think of voice or text messaging services and you’ll get the analogy. 

That would create incentives to “find something else to sell.” And that is where the obstacles begin. To be sure, many platform suppliers created themselves from nothing. In principle, any firm can hope too become a major platform, early on. But that is key: a firm has to move “early on.”

Established platforms beat others to market. A decade headstart often is insurmountable, once 

network effects are obtained. Once a firm becomes a platform, aggregating value, the scale advantages become moats.

So is 5G a platform? Mobile operators can hope that will become true, to some extent, perhaps as Amazon Web Services might be considered a platform. Most might tend to prefer the term “enabler,” even if some platform characteristics have developed. 

Skeptics might well conclude that connectivity providers selling services directly to retail end users have little chance to become major platforms on their own. Additionally, connectivity providers might have precious little ability even to leverage a growth path others have employed, namely, working with other key app providers.

Perhaps a good example is the way Uber and Lyft leverage other existing platforms such as iOS, Android, and Google Maps. It is not so clear how a connectivity provider can create a platform role when, by definition, other potential partners can simply assume connectivity exists, with no business relationships required. 

The trick is finding use cases where a direct business relationship, though not formally required, adds value and speeds market adoption by the other partners. Among the advantages large connectivity providers always tend to cite are scale, customer relationships and brand awareness (perhaps trust, as well). 

Those assets might lend themselves to marketing and distribution roles. That is why many firms believe they can be suppliers and creators of linear and over the top video entertainment services; home security; banking or payment services. 

Detailed Voice Guidance for Google Maps

Google Maps for sight-impaired, in U.S. and Japan, for the moment.