How Big is the Private 5G Network Opportunity for Mobile Operators?

Which is easier: building a general purpose communications network or integrating a new computing, application or platform for an enterprise? In other words, where does value lie? 

Most of us instinctively would opt for the latter as the driver of value, rather than the former. 

Which is easier: a systems integrator bundling local private 5G networks with application, platform or computing solutions for an enterprise, or a connectivity provider integrating those functions with its connectivity solution? Most of us instinctively would opt for the former, not the latter. 

All of that suggests the private 5G networks business will be led by the same cast of characters that have led the information technology system integration business for decades, namely third-party system integrators. 

As much as mobile operators might hope for a big role in system integration or supply of private 5G services and networks, optimism must be tempered. There will be few instances where a general-purpose private 5G network is the driver of enterprise interest.

In most cases it will be some business process the private 5G network must support that drives the deployment. And that suggests system integrators with vertical market knowledge will have the upper hand. 

Qualcomm and Capgemini are partnering to create and supply private 5G networks for enterprises, with Qualcomm supplying off-the-shelf private networks while Capgemini is the system integrator. 

“System integrator” is one of several private 5G network roles that mobile operators might themselves undertake, and, as always, illustrates the obstacles to success at scale in that role. Large consulting firms have for many decades operated as system integrators for information technology projects undertaken by enterprises. Private 5G now appears as one of the latest of such efforts.

A survey of mobile operator executives conducted by GSMA suggests the areas where suppliers believe they have opportunities, ranging from security and managed services to connectivity and integration. 

source: GSMA  

 

Some might be surprised by the belief that providing security services ranks at the top of possible opportunities. This apparently is because 44 percent of operators have seen increased growth in demand for security services from their enterprise clients due to Covid-19, GSMA notes. It is not immediately clear why the 5G connectivity provider is the most-logical provider of security, which also must be embedded directly into devices, applications and platforms, for example. 

The logical implication is that 5G offers more security than does Wi-Fi, for example, which is reasonable enough. The extent of this opportunity for sale of public services is the issue. It does not seem inherently clear that a public network 5G access provides advantages over a private 5G network that is logically isolated from the public network. 

That said, 45 percent of respondents consider it extremely important to invest in security because enterprise customers want it, GSMA reports. 

Fully 77 percent of operators are planning to offer security as part of their private network solution, which makes sense. 

Professional services include consulting (technical and business), systems integration and managed services, together account for 40 percent of operators’ potential revenue upside. The opportunity is arguably greatest for larger mobile operators with IT capabilities or with a strong local channel in local systems integrators or IT value-added resellers, GSMA notes.

How Will Operators Deliver 5G Private Network Features?

At the moment, mobile operators globally are optimistic about selling a variety of new connectivity services to enterprise customers. SD-WAN is among the new varieties of WAN services offered. But beyond the expected sale of 5G mobility service to enterprises, many expect to offer internet of things connectivity as well.

Private network services of several types also are envisioned. In some cases operators might engineer, build or operate such facilities as a managed service. In other cases the public networks might work. 

Edge networking, as opposed to edge computing services, also are seen as a growth area. 

source: GSMA 

Multi-access edge computing could affect public and private networking choices, at least when enterprises look at use cases requiring high-performance, low-latency, high-availability and high device density performance. 

Traditionally, private networks have been preferred when enterprises want high isolation from public networks, plus full control and lower cost, higher-performance in-building networking. Some 50 years ago, the use case was support for personal computers and peripherals connected to servers and cabled networks were the platform. 

These days, Wi-Fi tends to dominate for general purpose computing requirements supporting devices of all types. Mobile phone support remains a mix of public network access for voice and messaging, while data access can use either Wi-Fi or the public network. 

5G plus multi-access edge computing allows new permutations. Some 66 percent of mobile operators polled by GSMA are looking to use their core assets to support private enterprise networks. Some 59 percent also believe they can use their public networks to do so, operating with service level agreements.

In all likelihood, quite a mix will emerge. 

source: GSMA 

Is RCS Part of a Long-Term Pattern for Telecom?

Rich Communications Services was intended to be the multimedia successor to SMS (short message service). It still seems likely to do so, though the way it may do so illustrates some key changes in communications value and control, ultimately reflected in business models. 

The GSMA has promoted the use of RCS since 2008, primarily as a way of making carrier text messaging behave as do chat apps. But application-based messaging such as iMessage, WeChat, Slack, Skype, Viber, Android Messages, also are built on SMS. 

So RCS was envisioned as a “telco” messaging platform to rival app messaging. As with a few other salient telco initiatives, it has not quite worked out that way. 

In the U.S. market, AT&T and T-Mobile have opted to use the Google Messages platform, based on RCS. So instead of a “telco alternative” to app messaging services, RCS--as developed by Google Messages--becomes the AT&T and T-Mobile default for advanced messaging, not a carrier version. 

Telcos once hoped to become significant platforms for mobile app stores, data center hosting services and cloud computing as well. Many have moved into roles as video entertainment providers or content owners. They still hope to create roles in edge computing and internet of things value chains. 

History suggests it will be an uphill battle. Already, many major tier-one telcos are opting to become partners with hyperscale computing-as-a-service suppliers, rather than compete. Some telcos already are outsourcing their compute facilities to hyperscalers. 

AT&T will move its 5G core network to the Microsoft cloud. The switch means Microsoft’s Azure Cloud provides a path for all of AT&T’s mobile network traffic to be managed using Microsoft Azure technologies. 

As is typical for many outsourcing deals, the provider of the outsourced functionality also will absorb AT&T employees who presently operate the function. 

As part of the deal, Microsoft will gain access to AT&T’s intellectual property and technical expertise to grow its telecom flagship offering, Azure for Operators.

Microsoft also is acquiring AT&T’s carrier-grade Network Cloud platform technology, which AT&T’s 5G core network runs on. AT&T’s Network Cloud platform has been running AT&T’s 5G core at scale since the company launched 5G in 2018. 

Unlike some outsourcing deals, AT&T will continue to operate its core network, using Microsoft as the hardware platform. AT&T expects the deal will reduce its engineering and software development costs. 

Microsoft will assume responsibility for both software development and deployment of AT&T’s Network Cloud immediately and bring AT&T’s existing network cloud to Azure over the next three years, AT&T says. 

The point is that for all the wisdom of telcos “moving up the stack” and “adding more value,” that has proven exceptionally difficult. If that cannot be done, at scale, telcos are likely facing the alternative strategy of sharply reducing costs and consolidating. 

On the scale of potential strategies, ranging from a low-cost “bit pipe” optimized for cost to the “full featured application provider” roles, history suggests the former, not the latter, will be the ultimate fate for most service providers. 

Many note the centrality of connectivity for modern life. All true. But Wi-Fi also is central, and that is not a major revenue driver--if a revenue source at all--for service providers. It will not be a pleasant thought, but perhaps that ultimately is the fate of the global public networks business. 

Telcos used to drive global bandwidth needs and supply. Now hyperscale app providers do so, primarily relying on their own facilities. Wide area networks now are computer networks, albeit with some highlighted apps. 

Increasingly, WANs are not public networks, though an important role remains for public networks (voice as a last resort; public safety communications; military communications; internet access for consumers). 

Consolidation is definitely on the table in coming years. Beyond that, in many markets, common carrier operation might be the last resort. So a 50-year move towards privatization and deregulation might ultimately, in many markets, drastically revert towards common carrier market structures, at least for fixed networks. 

Wireless and mobility networks, with different cost structures, might remain the place where facilities-based competition remains more common.

"DX" Can Mean Almost Anything

Without a workable definition, it will be hard for any telcos or enterprises to figure out whether their “digital transformation” efforts and spending have had any positive effect. 

“Digital transformation” frequently is assumed to be an enterprise or telco objective, without specifying what that means. Colloquially, many assume it simply means converting manual processes to automated processes, or substituting digital technology for existing processes.  Quite often, that is a meaningful practical definition. 

In other cases digital transformation applies mostly to the customer-facing operations (sales, activation, support, fulfillment). Others emphasize the creation of new services and revenues beyond data transport. In that sense, it is a euphemism for “moving up the stack” or “taking on new roles within the value chain.”

The point is that DX is such a broad term as to be almost devoid of meaning, unless we speak concretely about the particular process, function or outcome we are trying to change. 

source: Ericsson 

With the caveat that a business model embraces all operations that any enterprise or organization must support to make money, digital transformation in its widest sense means new revenue models. In itself, that is about as broad a concept as that of the “business model,” which is all the processes any entity has to undertake to identify a customer problem, create its solution, deliver the solution and create a sustainable revenue model.  

source: TM Forum 

“Digital transformation is the process of using digital technologies to create new--or modify existing--business processes, culture, and customer experiences to meet changing business and market requirements, says Salesforce. As you would expect, Salesforce focuses on all customer operations. 

Clothes also tend to put the focus on customer-facing operations. “the realignment of, or new investment in, technology and business models to more effectively engage digital customers at every touchpoint in the customer experience lifecycle,” says Altimeter Group. 

In principle, according to the TMForum, DX can apply to almost any process conducted by an organization. So it might be best to envision digital transformation as a layer cake or continuum, where processes are changed little by little, starting with customer-facing operations in many cases, but ending up with creation of wholly-new or different products and revenue streams. 

That means DX will happen in stages, over time, from easiest to hardest processes, as would be the case for most information technology projects, changing operational processes one at a time.

"The End of Communications Services as We Know Them?"

“The end of communications services as we know them”  is the way analysts at the IBM Institute for Business Value talk about 5G and edge computing. 

One of the realities of the internet era is that although end user data consumption keeps climbing, monetization of that usage by communications service providers is problematic. Higher usage might lead to incremental revenue growth, but at a rate far less than the consumption rate of growth.

That is the opposite of the relationship between consumption and revenue in the voice era, when linear consumption growth automatically entailed linear revenue growth. Though there was some flat-rate charging, most of the revenue was generated by usage of long-distance calling services. 

“On the surface, exponential increases in scale would seem like a good thing for CSPs, but only if pricing keeps pace with the rate of expansion,” the institute says. “History and data suggest it will not.”

“Ericsson predicts that by 2030, ICTs (app, platform, content and service providers) will net a staggering $31 trillion in 5G-related consumer revenues,” the institute argues. “But, only 12 percent of this market is expected to be addressable by CSPs.”

“CSP consumer revenue growth is projected at less than an anemic one percent annually,”s ays the institute. “Despite optimism that consumers may be willing to pay a premium for at least some 5G services—for example, low-latency gaming—initial attempts by operators to charge one have failed,” the institute says. 

Analysis by Nokia and Bell Labs Consulting projects that only 13 percent of the $4.5 trillion spent by enterprises with ICTs in 2030 will come from basic connectivity. 

In substantial part, all that is a result of deliberate choices. When the global connectivity industry decided on use of TCP/IP as its next-generation platform, it also agreed to a formal separation of apps and content from connectivity. 

In other words, where app availability and network ownership were unified, in the IP era, there is a fundamental separation of apps and connectivity. Any lawful app can be used by any customer on any compliant network. 

The networks are open, not closed. That being the case, no connectivity services provider can bar lawful use of the access network to use any public IP platform, app or content provider. 

Nothing bars a connectivity provider from owning and supplying applications, content or platforms. But neither is a connectivity supplier logically required to approve use of the access network by any lawful IP app, content or platform supplier. 

So there was some early frustration about the fact that “anybody can use my network without paying me.” 

But that is precisely what TCP/IP networks allow and support. It is the fundamental logical architecture of the whole ecosystem. 

It is realistic and correct to state that “communications service providers (CSPs) are projected to miss out on most growth unless they adapt to add differentiated value into cloud native digital services, applications, and solutions,” as the institute says. 

But that was always an understood outcome of using IP and the internet as the fundamental architecture for customer and user access to platforms, apps, content and services. The whole point of TCP/IP is to allow “anybody to connect to anybody” no matter which network they use. 

That is a logical outcome of an architecture that separates the physical “access” function from all apps, content, marketplaces or services using such networks. There are logical conclusions for growth as well. 

Connectivity spending tends to grow in line with gross domestic product, especially when take rates approach 100 percent. At that point, every potential customer already is one. 

To be sure, new connectivity products are created from time to time, and some of them have mass market implications for connectivity providers. But most of those innovations are replacements for existing services. 

5G is an example, even if it mostly replaces existing 4G accounts in mass markets. Secure access service edge is an enterprise app that replaces virtual private networks. Software-defined wide area networks replace MPLS (multiprotocol label service). 

Relatively rarely does an entirely new category of connectivity services arise which does not cannibalize an existing service. Of course, there are other growth strategies. 

“If CSPs are to thrive, most will need to develop new competencies and assert themselves in new roles in value chains,” the institute argues. “CSPs should seek new ways to make money,

beyond metering connectivity and access to data, as these traditional mainstays of CSP business models are likely to commoditize.”

It is reasonable advice. Still, the observations and advice are well within the bounds of what has  been talked about for decades: namely  move beyond connectivity and into platforms, applications, solutions. 

In other words, take on new roles in the value chain. It is reasonable but difficult advice, not because executives are unwilling to do so, but because the obstacles to such strategies are so daunting. As the old adage goes, “telcos are lousy at innovation.”

“Strategic and operational decisions being made by CSPs today are critical to their ability to compete long-term against cloud native competitors,” the institute argues. Many would argue that the fight was lost long ago. 

It is reasonable enough to argue that connectivity providers should, if they can, become platforms. The stark reality is that this is going to be exceedingly difficult. 

Huge amounts of new value have to be discovered and created, no doubt. That value creation leads to revenue generation. It is reasonable to argue that much of that value will be found in new roles within the ecosystem. All are sound observations. Execution has been the issue, historically.

Telco Core Competence and Outsourcing

If, 20 or 30 years ago, you had asked any telecom executive what a service provider’s core competence was, you’d invariably get an answer related to “we operate quality networks” or “we understand networking” or something of that sort. 

Today, you might often get an answer that continues to emphasize the ability to build and operate communications networks. The odd thing is that most service providers are “building and operating” less these days. 

Few have gone as far as Sprint, which at one point outsourced the operation of its core network. But it is quite common for service providers to outsource towers, for example. And compared to the sourcing of core technology today, as compared to the monopoly era before 1990, telcos outsource almost all their core technology to third parties. 

To illustrate, where AT&T once developed and built all its core technology using Western Electric and Bell Laboratories, it now buys almost all its core technology from third party suppliers. 

That includes operating systems, edge computing, cloud computing, billing and operations support systems and even core networks, according to a survey of 500 telecom industry executives undertaken by the IBM Institute for Business Value and Oxford Economics.

source: IBM Institute for Business Value 

Though the IBM institute was looking at service provider willingness to use third-party cloud computing suppliers, the general principle--more outsourcing--seems to  apply to all core network technology. 

Mobile operators now are open to partnering with hyperscale cloud providers to supply the computational power needed by virtualized 5G networks, as well as to supply hooks for customer edge computing. 

Still, one can see clear signs of outsourcing of operations support systems, billing support systems, IP multimedia subsystems for voice, video and text messaging, as well as 4G and 5G core networks.

The point: one might wonder whether the telco "core competence" is something other than building and operating wide area communications networks.

Console Connect by PCCW Global: "We Don't Even Have to Know Who the Customer Is"

Console Connect, the PCCW Global automated ordering and interconnect system, is based on a community of partners including cloud services providers, networks and enterprise private clouds. 

The system is so automated--including settlements--that "we don't even have to know who the customer is," says Marc Halbfinger, PCCW Global CEO. Over the past year, PCCW Global has added 800 customers without any sales personnel contact.