Will Pervasive Computing Shift Enterprise IT Spending?

No matter how you decide to characterize “computing” these days, as it becomes more pervasive, most computing simultaneously has more of an  “enterprise” character, even as the purpose of enterprise computing increasingly has been to serve “consumers.”

Most mobile apps rely heavily on cloud computing, for example, and much cloud computing supports content delivery to consumer devices, for example. Where in past decades “computing” supported internal enterprise functions, computing these days most often support consumer content delivery.

On the other hand, in the next wave of computing, though human users will continue to be a key focus, more computing will shift back to more-traditional “enterprise” applications, where the purpose of computing is to support internal organization functioning (industrial processes, health processes, vehicle control, traffic management).

Just how enterprise IT spending might shift will be an issue as pervasive computing is enabled.

source: Morgan Stanley

Worldwide information technology spending is projected to total $3.5 trillion in 2017, a 2.4 percent increase from 2016, according to Gartner, including a one percent higher impact due to currency issues.

IDC estimates global IT spending will grow by 4.5 percent in 2017 in constant currency terms, a significant improvement on last year’s growth of 2.5%, with stronger upgrade cycles for infrastructure and mobile devices leading the improvement in the second half of the year.

But IDC also estimates total IT spending at $2.1 trillion.

Asia/Pacific (excluding Japan) will post the strongest regional growth in IT spending in 2017, set to increase by eight percent in constant currency terms, followed by the United States at four percent.

China and India are both expected to post overall IT spending growth of 10 percent in constant currency terms in 2017.

IT spending growth will be weaker but stable in Western Europe (three percent), Canada (two percent), and Japan (two percent).

Worldwide IT Spending Forecast (Billions of U.S. Dollars)

	2016 Spending	2016 Growth (%)	2017 Spending	2017 Growth (%)	2018 Spending	2018 Growth (%)
Data Center Systems	170	-0.3	171	0.3	173	1.2
Enterprise Software	326	5.3	351	7.6	381	8.6
Devices	630	-2.4	654	3.8	677	3.6
IT Services	894	3.2	922	3.1	966	4.7
Communications Services	1,374	-1.3	1,378	0.3	1,400	1.6
Overall IT	3,396	0.3	3,477	2.4	3,598	3.5

source: Gartner

Will Global Telecom, Having Moved From Monopoly to Oligopoly, Wind Up a Duopoly or Monopoly, Again?

Those of you who remember the monopoly period of telecom have by now gotten quite used to how different a competitive framework really is. But some analysts predict we might already have passed the absolute peak of competition, in most markets, globally.

What follows is not so clear. It seems unthinkable that we ever can return to the old days of sanctioned monopolies. But we have gotten comfortable (or at least resigned to the situation) with marketplace-created oligopolies.

What might come next could be equally unsettling, if present trends (lower average revenue per user, subscriber saturation, more-expensive networks) persist, could be quite a change.

Simply, if the typical “telecom services provider” can earn less revenue, at lower profit, than presently is the case (or has been possible in the past), then consolidation is inevitable, to boost both revenue mass and allow higher sustainable profit margins.

The big challenge for policymakers then will shift to frameworks that promote investment and competition where there are far fewer suppliers able to remain in business.

Though it seems unthinkable, some executives within the U.S. cable industry, for example predict and favor the emergence of a single cable company with national scale. Regulators have opposed such a situation in the past, and have taken steps to limit the total market share any single cable company, or telco, can have.

And though it might seem equally unthinkable, if that single cable operator emerges, is there room in the market for two “telecom” operators with national scale? Might the new oligopoly consist of one big cable company and one big telco? In other words, might the former duopoly (of one cable company and one telco) become an even-bigger duopoly?

Where today one cable company and one telco--but not the same telco and cable company--compete in virtually every market, could consolidation produce a situation where the same telco and cable company compete across 90 percent of the U.S. market?

That would require a historic change in in thinking about permissible market share. Many observers would worry about the reduced competition. But it also is possible that competition migrates elsewhere in the ecosystem.

Some might argue that already is happening, as app and device providers take more share of the “telecom spend.”

source: Deloitte University Press

Such a development would not preclude the existence of many smaller specialists, even some with facilities. Firms serving rural areas and metro specialists might continue to exist. But the “access” function might become much more consolidated.

How much that could matter remains to be seen. If surviving service providers have moved “up the stack,” then perhaps much of the competition will occur there, and not at the level of access. Most executives at service provider organizations might be quite happy, were that to be the case, as it would means that their organizations indeed have moved “up the stack,” and earn much of their revenue from applications, platforms and services, not access.

Apparently Nobody Likes Dumb Pipe, ISPs or Customers

Apparently, nobody--customers or providers--is too happy with dumb pipe. “Customers rank internet as the line of service with the lowest return on their investment,” a study by InMoment finds. Internet access also “is the line of service with which they are least satisfied and least likely to recommend.”

North American customers also say internet is the line of service with which they are least satisfied and least likely to recommend. Streaming services sit at the opposite end of the spectrum, ranking highest among customers among the categories of internet access, subscription TV, telephone service, mobility and streaming video services.

In fact, InMoment says consumers tend to view their service providers as “the enemy,” the customer experience study by InMoment found.  

Beyond, that, “ the industry has become an unsympathetic target everyone loves to hate,” InMoment’s survey of 11,000 North American consumers suggests.

In fact, service  providers themselves “are still viewed as a necessary evil rather than lauded for providing much-needed services,”  InMoment says.

Generally speaking, one would expect satisfaction scores to rise over time, as unhappy customers desert, while relatively more-satisfied customers remain. That does seem to borne out by the InMoment finding. After two years, scores generally rise, even if. after a year, most consumers are less happy.

“Satisfaction decreases universally for the first time at the one year mark, no matter the line of service,” InMoment says. “The same pattern occurs with a customer’s likelihood to recommend.”

“Customers who have switched providers in the last year rate key areas of the end-user experience lower than those who have not switched providers,” the study finds.

source: InMoment

Mobile Ecosystem and Market are Changing

It is hard to make good long-term decisions when industries are changing in fundamental ways, and that could be an issue for the U.S. mobile industry, which is expected to change quite a lot over the next decade.

One reason for believing that any merger of Sprint and T-Mobile US would lead to less competition is that this is precisely what equity analysts say are the advantages of such a merger: it would lead to less competition.

To be sure, there is an argument to be made (and will be made by Sprint and T-Mobile US if such a merger is attempted) that the combination would create a stronger competitor to AT&T and Verizon, which likely is quite true. But “stronger” for what purpose is a reasonable question.

It might be reasonable to argue that Sprint is in danger of business failure, or to argue that T-Mobile US might be able to grow itself enough organically to become a stronger competitor, but would get there right away by merging with Sprint.

But no equity analyst seems to make the argument that a Sprint merger with T-Mobile US will increase competition. They all seem to argue such a merger will lessen competition, and lead to higher prices.

That might be considered a good thing if what one wants is sustainable levels of competition beyond a duopoly. Three nearly evenly sized competitors might mean more competition is possible, long term, since that new firm would have greater scale and financial resources to compete with AT&T and Verizon.

Beyond that, the structure of the market might also be changing. With Charter Communications and Comcast entering the market, and others with niche business plans (internet of things networks using unlicensed spectrum; possibly app providers with e-commerce or advertising models), it is not clear that even three or four traditional mobile operators are the only contenders.

At least in principle, one can envision a market that blurs significantly, becoming more porous and harder to define. If Comcast and Charter cooperate, effectively operating as one national firm, it remains possible that no matter what happens, there will still be four or more leading national operators.

In fact, some would argue that already is happening, with the “mobile ecosystem” already changing, with new competition coming from the device supplier and app supplier parts of the value chain.

That is what Deloitte consultants now argue. In the mobile industry, handset and other ecosystem players are capturing an increasing share of revenue.

App providers also are effectively removing demand from the wide area data transport business, and operating their own captive networks. At the same time, value is migrating up to the app layer of the business.

And, in some cases, app providers are entering the access and device businesses as direct providers.

As a result, though total mobile ecosystem revenue is growing, more of the revenues now accrue to app and device partners, not access providers. Where in 2000 U.S. mobile operators might have retained about 85 percent of billed revenues, by 2013 access providers were retaining only about 63 percent of billed revenues.

Where we will be in a decade is not so clear.

source: Deloitte University Press

Foghorn Systems Intros New Industrial IoT Edge Computing System

FogHorn Systems today announced the availability of Lightning ML, the newest version of its Lightning  edge intelligence software platform for the Industrial Internet of Things (IIoT). Lightning ML is now the industry’s first IIoT software platform with integrated machine learning capabilities and universal compatibility across all major IIoT edge systems.

“The money and time required to move massive amounts of machine data to the cloud for analysis, only to send the results back to the edge, often makes little sense,” said Mike Guilfoyle, Director of Research and Senior Analyst at ARC Advisory Group. “In many instances cloud computing won’t be practical, necessary, or desirable.”

Lightning ML leverages existing models and algorithms, using live data streams produced by existing physical assets and industrial control systems, with tools directly accessible by non-technical personnel.

Lightning ML enables complex machine learning models to run on highly-constrained compute devices such as PLCs, Raspberry Pi systems, tiny ruggedized IIoT gateways, as well as more powerful Industrial PCs and servers, FogHorn Systems says.

Even with the addition of advanced machine learning capabilities, the complete Micro edition of the Lightning ML platform requires less than 256MB of memory footprint, Foghorn Systems notes.

Can Easier Make-Ready Change the FTTH Business Case?

"Make ready" costs (the cost of readying an aerial facilities pole for a new set of communication cables) might represent $4,000 to $35,000 per mile of cost for a new distribution network. That represents about a low of two percent and perhaps a high of eight percent of total distribution network costs.

So it stands to reason that rules that lead to "make ready" costs that are lower, with execution faster, should improve the business model for either fiber to home or 5G small cell access networks.

Removal of barriers to investment in next-generation mobile and fixed broadband networks proposed by the U.S. Federal Communications Commission could lead to deployment of fiber-to-customer  facilities to 26.7 million premises that would not have gotten such investment under the older rules, according to an analysis by Dr. Hal Singer, Economists Incorporated principal, Ed Naef and Alex King,  partners at CMA Strategy Consulting.

Those moves would make it faster and less costly to deploy next-generation networks, including measures such as reducing pole-attachment costs, the time and cost of make-ready and barriers to copper retirement. Those moves potentially are significant because construction represents most of the cost cost of a fiber-to-home network.

The new rule also would accelerate legacy time-division multiplexing (“TDM”) product discontinuance and reduce barriers to locating and deploying wireless infrastructure.

If the rules are changed, Economists Incorporated estimates an incremental 26.7 million U.S. premises would be passed by fiber-to-home facilities, as the better business model would make the investment worthwhile.

The analysis also suggests that the change in rules would enable about 15 million new fixed wireless 5G locations to be added, about 66 percent of those being in rural areas.

The new line of thinking assumes that faster and lower-cost network construction includes “fiber deep” distribution networks that create radio small cell sites covering areas of 1800-meters radius, or 3600-meters diameter, using 3.5-GHz frequencies. That implies coverage of roughly four square miles from each such small cell.

In urban and suburban markets, that could cover significant homes per small cell In U.S. suburban areas, that might mean one small cell, using 3.5-GHz spectrum, might cover 200 to 600 homes.

Since connecting new customers represents such a huge part of the total "cost to serve a customer" ($830 per customer up to $1870 per customer, it is possible the ability to use small cells to supply a fixed wireless drop, not a cabled connection, would allow a huge improvement in business model.

Some still argue FTTH costs less to deploy than 5G fixed wireless. Others think 5G fixed wireless might well be more affordable than deploying FTTH.

source: Economists Incorporated

source: Economists Incorporated

Big Opportunities and Risk for 5G

Not since the advent of 3G has a next-generation mobile network promised so much, and carried such risk. Recall that 3G was the first to have as a key premise the creation of many new apps, services and revenue models. The 2G network did add text messaging as an important new application, but 2G mostly was about efficiency for traditional apps.

Whatever the opportunities and risks, 5G is coming fast. GSA researchers have identified at least 36 operators from 23 countries that have demonstrated 5G technologies, or announced 5G tests, or trials. More significant are the early commercial deployments.

MTS now expects to have a 5G network in Moscow at 2018 Football World Cup. SK Telecom plans to launch a pre-5G network by the end of 2017, and a 5G network by 2020. Telefonica is likely to launch 5G first the U.K. market. Japan will have commercial deployments by 2020, with targeted deployments before then. U.S. carriers likely will deploy pre-5G for fixed wireless in 2018, with more-limited commercial deployments in 2017.

Widespread commercial launches likely will not happen until 2020, in most countries that say they will deploy “soon.”

But 5G is likely to be quite different from 4G. The big new use cases and revenue streams will come from enterprise customers, not consumers. And, to an extent we have not seen since 3G, use cases must be created; they are not obvious extensions of current demand.

The coming 5G era is likely going to prove even tougher than 2G and 4G, in terms of business model, for one key reason.

Most “human user” requirements will be satisfied by 4G. The big new revenue opportunities for 5G are going to come from “non-human” users (internet of things sensors and applications) and enterprises. Mobile operators in markets with fewer enterprise customers are going to find the business model more challenging.

That might mean 5G is more akin to 3G than to 2G or 4G, in terms of adoption. Arguably, 3G, which was supposed to lead to a wave of app and service innovation, did not immediately do so. Instead, many would argue, those benefits were deferred to 4G.

In similar manner, 5G, it is hoped, will lead to a wave of app, service, product and revenue innovation. The similarities are that both 3G and 5G assume a shift of demand.

The 5G network’s real upside is “non-human” use cases, as 3G’s real upside was “new digital apps and services.” But there was an innovation lag with 3G, and some might say that is the danger with 5G as well.

Still, 5G is coming. A reasonable rule of thumb for mobile network generations is that the next-generation network (in the post-2G networks era) tends to be initially deployed when adoption of the prior network has not even reached 40 percent of customers, globally. That is likely to be the case for 5G as well.

The point: 5G is coming fast, but will pose huge business model challenges, beyond those typically seen when amortization of the prior generations is not complete.

source: A.D. Little

Mobile executives in developing markets have faced a key challenge in the post-2G eras: deployment of the next-generation networks always occurs when adoption of the prior generation remains rather moderate.

That means there is a potential business model problem: amortization of each network generation is tough. The problem arguably has gotten worse. The first digital platform--2G--had a very long life cycle. The succeeding 3G platform has had a shorter life cycle, with sluggish adoption rates, initially,  while 4G might have even a shorter life cycle.

There are a number of reasons amortization is difficult. Technology cycles are dictated by market demand in the developed markets, where a new network tends to be introduced at least once every decade.

On a global level, for example, 4G was commercialized when 3G adoption had not reached 10 percent. In North America, 3G service was launched about 2005. The LTE launch occurred only five years later.

In 2017, in North America, 4G adoption is above 85 percent. Keep in mind that LTE service was launched in the U.S. market in 2010. So 85-percent adoption took seven years.

That means lead users in developing markets will want access to that latest platform as well, even as suppliers struggle to get costs to price points lower-income users can afford.

source: Ovum

source: Telegeography

source: GSMA

source: GSA