The infrastructure of the digital economy is becoming increasingly important to the way businesses operate, compete and allocate capital. Much of the investment debate has concentrated on computing capacity, semiconductor supply and the expenditure required to support artificial intelligence. These are significant developments, but they represent only part of a much broader economic transformation. Beneath the visible applications and hardware lies an interconnected system of cloud platforms, enterprise software, identity services, cybersecurity networks and communications infrastructure on which an expanding range of commercial activity depends.
The companies controlling these systems occupy an unusual position. Their products are often essential to customers but largely invisible to the end users who benefit from them. An employee accessing a corporate application, a retailer processing an online payment or a manufacturer coordinating suppliers may rely on several infrastructure providers without directly interacting with any of them. As these dependencies deepen, the providers responsible for maintaining connectivity, security and operational continuity can become embedded in the economic processes of their customers.
That position creates the possibility of attractive recurring revenue, durable customer relationships and economies of scale. It also introduces important questions about concentration, competition and the ability of customers to retain control over their technology choices. A platform that simplifies business operations may create considerable value, but the same integration can make switching providers expensive or operationally disruptive.
For investors, the central issue is therefore not simply which companies will supply the next generation of technology. It is which businesses will control the essential interfaces through which that technology is accessed, secured and managed, and whether those positions can generate sustainable economic returns without being undermined by competition or regulation.
The economic importance of the digital control layer
Digital infrastructure is often described in physical terms: servers, fibre-optic networks, data centres and telecommunications equipment. Yet the economic value of these assets increasingly depends on the software and services that coordinate their use. Cloud management platforms allocate computing resources, identity systems determine who can access applications, cybersecurity services protect communications, and enterprise software connects information across departments and organisations.
These functions form a control layer between the underlying technology and the businesses using it. A company may be able to purchase computing capacity from several suppliers, but moving the applications, databases, security policies and operational processes associated with that capacity can be considerably more difficult. The provider managing these relationships may therefore possess an advantage that extends beyond the immediate technical characteristics of its product.
The distinction matters because technological capability and economic control are not necessarily the same thing. Infrastructure can become more powerful and less expensive while the software required to manage it becomes more deeply embedded in customer operations. Improvements in underlying technology may even strengthen established platforms when they allow those providers to introduce additional services into an existing customer relationship.
This helps explain the significance of integrated enterprise technology businesses. Their competitive position can arise not only from the quality of individual products, but also from the convenience, compatibility and administrative simplicity of purchasing several related services together. Customers may value having fewer suppliers, consistent security policies and a common operating environment, even when competing products offer advantages in particular functions.
The investment opportunity, where it exists, lies in the ability to convert that operational importance into recurring economic value. The risk is that customers, competitors and regulators increasingly challenge the dependence on which that value rests.
Cloud platforms and the economics of integration
Cloud computing provides one of the clearest examples of how infrastructure and software ecosystems reinforce one another. A cloud provider initially attracts customers through access to computing, storage and networking services, but the relationship can expand into databases, development tools, security, analytics and business applications. Each additional service may make the platform more useful while increasing the complexity of moving elsewhere.
The scale of these relationships is evident in Microsoft's 2026 annual report, which reported Microsoft Cloud revenue of $214.4 billion for the financial year ended June 2026, compared with $168.9 billion a year earlier. The company also disclosed approximately $678 billion of commercial remaining performance obligations at year-end, although these represent contracted future revenue rather than revenue already earned. The figures demonstrate the breadth of Microsoft's commercial relationships, while requiring careful interpretation because the reported cloud measure includes several distinct businesses.
A customer using a broad collection of cloud and enterprise services may benefit from common administration, integrated security and reduced implementation complexity. The supplier benefits from multiple revenue streams and greater opportunities to introduce complementary products. This relationship can support retention and expansion, although it does not establish that every additional service is competitively superior or economically profitable.
There are also important differences between switching costs and genuine network effects. A business may remain with a cloud provider because migrating applications is expensive, not because the platform becomes inherently more valuable as additional customers join. Network effects arise when participation by other users or developers improves the service itself; switching costs arise when changing suppliers imposes economic or operational burdens. Both can strengthen an incumbent, but they operate through different mechanisms and may respond differently to competitive pressure.
Cloud customers increasingly have reasons to resist excessive dependence. They may distribute workloads across multiple suppliers to improve resilience, negotiate more effectively or access specialised capabilities. Such arrangements can reduce concentration risk, although operating several platforms also creates additional management and security complexity. The long-term competitive question is whether integrated ecosystems will remain sufficiently valuable to justify that dependence as interoperability improves.
Enterprise software: where workflows become infrastructure
Enterprise software can be particularly durable because it becomes intertwined with the processes through which organisations conduct business. Financial reporting systems, customer relationship management platforms, procurement applications and human-resources software often contain years of operational information and reflect decisions about how employees work. Replacing them can require extensive data migration, retraining and changes to established procedures.
The economics of these relationships differ from those of consumer applications. Enterprise customers generally evaluate software through reliability, compliance, integration and the cost of disrupting critical operations, rather than convenience alone. Once a platform becomes central to a company's workflow, the decision to replace it may involve risks extending well beyond the subscription fee.
This creates an opportunity for providers to expand within existing accounts. A supplier initially responsible for one business function may introduce additional modules, analytics capabilities or workflow automation, increasing the value of the customer relationship without acquiring an entirely new customer. Subscription arrangements can make revenue more predictable, while integrations and customer-specific configurations can support retention.
However, recurring revenue should not be confused with guaranteed profitability. Software companies must continue investing in development, security, customer support and sales, while competition can constrain pricing. A high renewal rate may indicate genuine customer satisfaction, but it may also reflect the practical difficulty of switching. Investors should consider whether retention is accompanied by growing customer value or merely by increasing dependence.
The more durable businesses are likely to be those whose platforms remain useful as technology changes. A system that can incorporate new capabilities while preserving established workflows may strengthen its position, whereas one that depends on customers being unable to leave can become vulnerable when alternative solutions make migration easier.
Cybersecurity and identity are becoming essential interfaces
The expansion of cloud services and distributed applications has altered the way organisations manage security. Traditional approaches based primarily on protecting a corporate network perimeter are less suitable when employees, customers and software systems interact across numerous locations and providers. Security increasingly depends on identifying users and devices, controlling access and monitoring activity across interconnected environments.
Digital identity is particularly important because it determines which individuals and systems are authorised to use particular resources. A provider responsible for authentication and access management can become closely integrated with a customer's applications, security policies and administrative processes. The resulting relationship may be commercially durable, but its value ultimately depends on reliability, compatibility and trust.
Okta's fiscal 2026 annual filing illustrates the importance of this integration. As of 31 January 2026, the company reported more than 20,000 customers and an integration network containing over 7,000 connections to applications and technology services. These figures indicate the breadth of the platform's relationships, although the number of integrations is not itself a measure of profitability or customer dependence.
The competitive dynamics of identity management are also distinctive. A broadly compatible independent provider may appeal to organisations seeking to avoid dependence on a single technology ecosystem. Conversely, large cloud and enterprise software companies can incorporate identity and security functions into wider commercial packages, potentially reducing the attractiveness of standalone products.
Cybersecurity businesses face a similar tension between specialisation and consolidation. Customers may prefer sophisticated individual tools for particular risks, but managing numerous suppliers can introduce complexity and additional costs. Platforms capable of combining multiple security functions may benefit from that demand for simplification, provided integration does not compromise effectiveness.
The importance of security does not guarantee attractive returns for every supplier. Threats evolve, customers reassess their requirements and technological advantages can diminish rapidly. Investors must distinguish between a service that is essential in principle and a company capable of delivering that service more effectively than competitors over an extended period.
Connectivity networks can create advantages beyond transmission
The internet depends on networks that move information between users, applications and infrastructure providers. Businesses operating these networks can benefit from geographical reach, relationships with telecommunications operators and the ability to process traffic close to its destination. As digital services become more distributed, performance and security increasingly depend on where and how information is routed.
Cloudflare provides an example of this approach. Its 2025 annual filing reported approximately 332,000 paying customers at 31 December 2025, compared with a substantially smaller base several years earlier. The company offers connectivity, security and application-related services through a shared global network, creating opportunities to serve multiple customer requirements using common infrastructure.
The economic attraction of such a model lies partly in the potential to spread network costs across a growing customer base. Additional services may make better use of existing infrastructure, while broader traffic visibility and operational experience can support improvements in security and performance. These advantages depend, however, on whether incremental revenue exceeds the additional expenditure required to maintain network capacity, develop products and support customers.
Network effects should again be considered carefully. More customers do not automatically improve every connectivity service, and traffic growth can increase costs as well as revenue. A provider may benefit from scale economies, extensive interconnection and accumulated operating expertise without possessing a self-reinforcing network effect in the strict economic sense.
Competition can also emerge from several directions. Large cloud providers operate substantial global networks, telecommunications companies retain control over important physical infrastructure, and specialist security vendors compete for particular functions. The advantage of an integrated connectivity platform will depend on its ability to provide sufficiently valuable combinations of performance, security and simplicity.
Regulation is beginning to challenge the economics of dependence
The same characteristics that make digital platforms commercially attractive have attracted increasing regulatory attention. High switching costs, integrated product ecosystems and control over essential services can create efficiencies, but they may also limit competition or make customers dependent on a small number of suppliers.
The European Commission's 25 June 2026 preliminary assessment proposed that Amazon Web Services and Microsoft Azure should be designated as gatekeepers under the Digital Markets Act. The Commission cited their established market positions, extensive ecosystems and potential lock-in effects. This was a preliminary regulatory position, not a final finding that the companies had breached competition rules.
The European Union's Data Act introduces a related challenge through its provisions on switching between data-processing services. According to the European Commission's explanation of the legislation, the rules require providers to address obstacles to switching and provide for the removal of qualifying switching charges, including certain data-egress charges, from 12 January 2027. The precise treatment of transfers depends on the circumstances, and ordinary ongoing services should not be confused with switching operations.
These developments matter because regulation can alter the economics of customer retention. If customers can transfer data and workloads more easily, incumbents may need to compete more actively on service quality, pricing and innovation rather than relying on the expense of migration. Greater interoperability could also benefit specialist providers that previously struggled to compete with integrated platforms.
Yet lower switching costs do not necessarily eliminate competitive advantages. Customers may continue to prefer established providers because of reliability, functionality, support or the benefits of integration. Indeed, reducing artificial barriers could make genuine product differentiation more economically important, rewarding businesses that retain customers through superior service rather than contractual or technical restrictions.
The consequences will depend on implementation, enforcement and customer behaviour. Investors should distinguish between announced regulatory intentions, legally applicable obligations and measurable changes in competition. A regulation may alter the long-term structure of an industry without producing an immediate change in reported revenue or profitability.
Recurring revenue is valuable only when it produces economic returns
Digital infrastructure businesses are often valued partly on the stability of their subscription and usage-based revenues. Recurring payments can improve financial visibility, support customer relationships and reduce the uncertainty associated with repeatedly acquiring new business. However, the quality of those revenues depends on the costs required to retain customers and maintain the underlying service.
A subscription may be contractually recurring while remaining economically vulnerable. Customers can reduce usage, negotiate discounts or consolidate suppliers, while providers may need to increase spending to meet security, performance and regulatory requirements. A growing revenue base is therefore not sufficient evidence of improving economic returns.
Investors should examine retention alongside pricing, customer concentration, operating margins and free cash flow. Remaining performance obligations can provide useful information about contracted business, but their value depends on the timing, profitability and fulfilment of those obligations. Similarly, strong customer growth can be encouraging without demonstrating that acquisition costs will be recovered through future earnings.
Capital intensity also differs considerably across the digital infrastructure landscape. Software and identity providers may require relatively limited physical infrastructure but substantial development and sales expenditure. Cloud and connectivity operators may depend on significant investment in equipment and networks. Comparisons based solely on revenue growth or operating margins can therefore obscure important differences in the capital required to sustain each business.
The strongest economic positions are those where customer value, retention and capital efficiency reinforce one another. A platform that delivers essential services, expands economically within customer accounts and generates attractive cash returns may justify a durable competitive position. One that must continually increase expenditure simply to defend existing revenue may be less valuable than its apparent market importance suggests.
The next contest will be over control and interoperability
The development of digital infrastructure is unlikely to produce a single winner across every layer of the technology economy. Cloud platforms, enterprise software, cybersecurity, identity and connectivity perform different functions, and customers have legitimate reasons to combine integrated services with specialist suppliers. The competitive landscape is therefore likely to remain a mixture of large ecosystems and businesses focused on particular capabilities.
The more important question is where economic control ultimately resides. A cloud provider may own the underlying computing environment, while an enterprise software company controls the workflow, an identity provider manages access and a connectivity network governs how information moves between them. Each occupies a potentially valuable position, but each also depends on the reliability and cooperation of other parts of the system.
Interoperability will become increasingly significant. Customers want the flexibility to adopt new technology without rebuilding their entire operating environment, while regulators are seeking to reduce barriers that prevent them from changing suppliers. Providers able to participate effectively in open and interconnected systems may find new opportunities, even as established incumbents face pressure on some traditional sources of customer dependence.
For investors, this suggests a more demanding approach than simply identifying companies exposed to expanding digital activity. The relevant questions concern the durability of customer relationships, the source of switching costs, the strength of integration advantages and the ability to sustain attractive returns as competition evolves. The ownership of critical infrastructure can create an economic advantage, but it does not guarantee that advantage will remain intact.
The next technology battleground may therefore be less about who builds the most computing capacity than who establishes the trusted systems through which businesses operate. The companies that succeed will need to combine reliability, security and integration with a willingness to adapt as customers demand greater flexibility. Ultimately, the most valuable digital infrastructure may be that which becomes indispensable through the quality of its service rather than through the difficulty of leaving it.