How Does AT&T’s Public Cloud Strategy Fare against that of Verizon?

On June 30, Microsoft announced that it intends to acquire AT&T’s intellectual property and human capital for AT&T’s Network Cloud Platform. That platform serves as the foundation for AT&T’s 5G core network. AT&T continues to manage its network and customer relationships, but Microsoft is obtaining cellular capabilities to control, design, and build core network functionality. In other words, the intelligence—and the service logic—resides within Microsoft’s public cloud infrastructure and is under its control. With this decision, AT&T takes the plumbing and inner workings of the core network to Azure for Operators. This is enabling Microsoft to accelerate deployment of its public cloud in telecoms as discussed in ABI Research’s Insight (IN-5693), “Microsoft Brings the Power of IT into Telecoms.”

By contrast, Verizon announced recently that it will use its public cloud selectively. For example, Verizon deploys its own cloud infrastructure to host its 5G core network. Given that the core network constitutes a core strategic asset for Communication Service Providers (CSPs), this may well be a prudent move by Verizon. At a very fundamental level, what Verizon obtains from commanding the workings of its core network is the monetization potential that comes from control and integration. Verizon remains committed to using public cloud innovation, but there are strategic points where what matters is not technological innovation; what matters is value chains and the point of integration on which a company’s sustainable differentiation is built. In the cellular world, the core network has been, and continues to be, a strategic point of differentiation.

A discussion on commercial dynamics between hyperscalers and CSPs should be framed around two strands: (1) the realities of telecoms and information technology/hyperscale industries and (2) the differentiating attributes that constitute a core competency. In telecoms, growth revolves around technical excellence and integration in cellular. The location and control of the intelligence or service logic resides within the network itself and is under the CSP’s control, producing positive feedback loops and outsize profits. Further, like most other industries, telecom is defined and bounded by economic models, explicit customer expectations, and competitive structures that are known to all. Hence, it is impossible to change strategy in a short period of time, so the growth strategy for CSPs has been universal—as, for example, in control and integration of cellular networks to provide mass market and country- and region-specific subscriber telephony services.

Hyperscalers leverage a homogeneous tech ecosystem to monetize cloud innovation. A hyperscaler’s strategy is to invest a significant amount of capital up front. They make up this capital on volume because of superior cost structure and infrastructure commoditization that comes from a (close to) zero-margin nature of software in which they have become experts. Hyperscalers place their core cloud offering at the center and modularize/commoditize their complement to extract the most value. With the core network shifting to the public cloud—as it is the case with AT&T—the key point of differentiation for CSPs is modularized and driven to commodity competition. A commoditized core network is traded as if it is fungible—one “unit” is indistinguishable from another. A commoditized 5G core network has no (engineering) differentiation, but Verizon is maintaining that differentiation with its decision to host the core network in its private cloud.

Further, a lack of control (integration) in the core network without some point of lock-in may not be conducive to capturing the value that stands to be created in cellular. With intelligence (service logic) shifting from being centrally controlled by CSPs to residing in end devices/infrastructure (e.g., PCs, servers, etc.), customers can choose different providers or can, in effect, self-provide. This may potentially have implications in terms of control over the development of new services. But it may also serve as a foundation for new value creation. If there is value eroded in a specific part of a value chain, there will be new value created in another part. However, profits in new value chains will flow to those entities that compete on distinctive core competencies. This will very likely require integration of different components of newly created value chain spots.

CSPs can never be as focused as application providers or cloud players that are in business for cloud computing and software and that do it better than anyone else. CSPs can benefit from public cloud efficiency because they can consolidate much more functionality with less equipment. So better economics, business agility, and innovation are all key drivers for the public cloud to potentially propel innovation in the industry. Technological innovation comes from hyperscalers as discussed in ABI Research’s Insight (IN-6150) “How Public Cloud Can Spur Innovation in Telecoms.” But equally important is the point of integration on which a CSP’s core competency will be built. And that must come from the CSP community as they reshape existing business models to compete with distinctive, sustainable competencies.

CSPs must come to grips with the fact that,  with announcements like that of AT&T, the integration that drives their business will disintegrate. Disintegrating CSPs’ integration allows hyperscalers and other new entrants to create new integrations and profit elsewhere in the value chain. And this is a lesson that can be applied broadly. If some part of the value chain becomes loose, it will constitute not only an opportunity but also a warning that the entire cellular value chain is going to be transformed. The challenge that the industry faces is that value chain transformation is a very dynamic process. Where new opportunities materialize and who best commercializes them remains to be seen. Succeeding and profit extraction requires differentiation that CSPs must turn into an enduring economic moat, much like the cellular core network.

Lastly, CSPs should understand that, at a very fundamental level, optionality, monetization potential, and sustainable competencies come from control and that, when it comes to infinitely copyable software such as 5G core networks, control comes from integration. For now, CSPs’ strategic high ground is the core network, especially with 5G Next Generation Core that promises to deliver enterprise applications, including network slicing and low latency applications. Within the core network specifically but in cellular generally, the industry’s competitive structures remain the same. So, broadly speaking, winners will be those that not only keep an eye on the present but also look out to the future as discussed in ABI Research’s Insight (IN-5939), “How Can NEPs Satisfy Today’s Market Demands While Pursuing Tomorrow’s Growth Opportunities?” Verizon’s strategy preserves the present, and that is a balanced decision. By contrast, AT&T’s partnership is arguably aimed at the future by pursuing a novel strategy that hedges its bets. This matters. But whether it is sufficient is a function of the companies that AT&T—and the broader CSP community—competes with—now and in the future.