The End of the B2B Software Annuity: How AI is Reshaping Industry Valuations and Growth Trajectories
For two decades, a predictable and lucrative model defined the business-to-business (B2B) software landscape: the "slow lane." This segment, characterized by steady, albeit unexciting, growth, offered a stable annuity stream. Companies in this space enjoyed high net revenue retention (NRR), robust profit margins, sticky customer contracts, and data locked within their applications, ensuring predictable cash flow. While no longer considered growth stocks, these software firms effectively functioned as bonds with a modest upside, a reliable investment that attracted significant attention from private equity firms, which acquired many such companies with clockwork regularity. This established paradigm, however, is now facing a fundamental disruption, driven by the accelerating capabilities of artificial intelligence (AI).
The core of this once-unshakeable annuity was built upon the concept of "seats" – individual user licenses sold to businesses. As customer workforces expanded, so did the number of seats, leading to consistent revenue growth and high NRR, often exceeding 110%. This "land and expand" strategy, where initial sales led to organic growth as clients added more users, was the engine of the B2B software era. Contractual recurring revenue, coupled with high switching costs and years of workflow integration, made these software assets appear inherently durable. Investors were willing to pay a premium for this reliability, as B2B software historically traded at a significant markup compared to the broader S&P 500.
However, the market’s perception of this annuity has shifted dramatically. A significant cohort of publicly traded software companies is now experiencing growth rates below 5%, a stark departure from previous expectations. Companies like Dropbox, Zoom, DocuSign, and PagerDuty, once bellwethers of consistent revenue, now find themselves in single digits or even declining growth. This slowdown is not merely a temporary lull but appears to be a fundamental re-evaluation by the market, which is questioning the durability of the cash flow these companies generate. They are no longer viewed as slow-growth entities but as businesses potentially facing the "end of growth," with an added layer of uncertainty regarding the long-term safety of their existing revenue streams.
The advent of sophisticated AI, particularly generative AI and AI agents, is the primary catalyst for this seismic shift. These intelligent systems are fundamentally altering the economics of software consumption. AI agents are increasingly capable of performing tasks that previously required multiple human users, thereby directly challenging the seat-based licensing model. If an AI agent can effectively handle the workload of ten support representatives, a company will logically reduce its need for ten individual seats, opting instead for a single human overseer and the AI agent. This dynamic transforms expansion from a tailwind into a headwind, potentially reversing the growth trajectory for applications reliant on user counts. Dropbox’s recent guidance for a revenue decline serves as a somber, albeit slow-motion, illustration of this potential reversal.
Furthermore, the financial resources of enterprises are being reallocated at an unprecedented pace. Investment in AI infrastructure and development is soaring. For instance, Anthropic, a leading AI safety and research company, has seen its annualized run rate surge to over $19 billion, a substantial increase from $9 billion at the end of 2025. Projections indicate that approximately 75% of new hyperscaler infrastructure spending in 2026, estimated to exceed $450 billion, will be directed towards AI initiatives. This massive influx of capital into AI means that budget dollars previously allocated to traditional B2B software categories, such as CRM seats, IT service management (ITSM) modules, or storage tiers, are now being rerouted to AI-powered solutions. Chief Information Officers (CIOs) operate within finite budgets, and every dollar invested in AI agents represents a dollar that cannot be used to expand existing seat-based contracts.
The Impact on Valuation: A Shift from Growth to Terminal Value
The implications of sub-5% growth extend beyond mere revenue deceleration; they fundamentally alter the mathematical underpinnings of software company valuations. In a typical discounted cash flow (DCF) model, a substantial portion, roughly 85% to 95%, of an enterprise’s value is derived from its terminal value – the projected value of the business beyond the explicit forecast period. When a company exhibits robust growth rates, such as 30% annually, the focus remains on near-term expansion, and the terminal value is a less critical component of the valuation equation.
However, as growth plummets to around 4%, the terminal value becomes the dominant factor. The market’s attention shifts from "how fast are you growing?" to a singular, critical question: "is this annuity durable?" For two decades, the answer was an unequivocal yes, supported by entrenched workflows, long-term contracts, and high customer stickiness. This perceived reliability allowed software companies to command a premium valuation compared to the S&P 500.
This premium has now evaporated. As of early 2026, the software sector is trading at approximately 22.7 times forward earnings, a valuation that has fallen below the S&P 500 for the first time in recorded history. The iShares Expanded Tech-Software Sector ETF (IGV), a benchmark for software stocks, has seen a decline of around 30% from its peak in September 2025, representing a loss of approximately $2 trillion in market capitalization. This downturn is distinct from previous market corrections, such as the 2008 financial crisis, where the underlying economic fundamentals of software remained intact and a swift rebound followed. Today’s market is not suggesting that software is temporarily overvalued; rather, it is expressing a profound uncertainty about the very nature of the software annuity itself.
The Bifurcation of the Software Sector: Infrastructure vs. Applications
The narrative of a monolithic B2B software market facing a universal decline is an oversimplification. The current landscape is characterized by a significant bifurcation, with two distinct categories emerging: AI infrastructure and AI-dependent applications.
On one side, companies providing the foundational infrastructure for AI are experiencing a resurgence and accelerated growth. Cloudflare, a leading cloud security and performance company, has guided for 28-29% growth. Snowflake, a cloud-based data warehousing company, reported 30% product growth. Twilio, a cloud communications platform, posted 20% growth, its fastest in over three years, driven by the surge in AI voice workloads. DigitalOcean, a cloud infrastructure provider for developers, is forecasting 19-23% growth. These companies benefit directly from the increased adoption and usage of AI. As AI agents consume more computing power, storage, bandwidth, and data, the demand for the underlying infrastructure that supports these operations intensifies.
Conversely, applications built on the traditional seat-based model are increasingly drifting towards a terminal state. HubSpot, a customer relationship management (CRM) platform, offers a compelling example of this trend. Over five consecutive quarters, the company has reported decelerating customer count growth: 21%, then 19%, 18%, 17%, and finally 16%. While HubSpot remains a strong and valuable company, the consistent downward trajectory in customer acquisition indicates a fundamental shift in its growth dynamics. This pattern is indicative of applications that are increasingly being commoditized or augmented by AI, reducing the need for extensive human interaction and, consequently, fewer user licenses.
The implications of this bifurcation are profound. The concept of a single, unified "software category" is becoming obsolete. Instead, the industry is now divided into two distinct ecosystems: infrastructure that powers AI innovation and applications that are being reshaped, and in some cases, cannibalized, by AI.
The Demise of Graceful Decline and the Insufficiency of Profitability
In the pre-AI era, mature B2B software companies could often navigate a period of slower growth with a degree of grace. They could focus on optimizing operations, generating strong free cash flow, and returning capital to shareholders, even if top-line growth stagnated. Profitability was a significant factor that could sustain their valuation and investor confidence during these transitional phases.
However, the current AI-driven disruption leaves little room for graceful decline. The speed at which AI capabilities are advancing and being integrated into business processes means that companies are facing an existential threat to their core business models. Simply achieving profitability is no longer a sufficient safeguard against obsolescence or significant market de-rating. The market is now demanding more than just financial stability; it requires demonstrable adaptation and innovation in the face of AI’s transformative power.
The data presented in the original article, while dated (referencing screenshots from 2026), illustrates this point starkly. Companies like Dropbox and PagerDuty, cited as examples of those struggling with sub-5% growth, represent the challenges faced by applications whose value proposition is directly impacted by AI. Dropbox, a file-sharing and collaboration platform, sees its value proposition threatened by AI’s ability to integrate with and extract information from various data sources, potentially reducing the need for centralized cloud storage as a primary function. PagerDuty, an incident response platform, faces a similar challenge as AI becomes more adept at predictive maintenance and automated issue resolution, diminishing the reliance on human intervention for alerts and escalations.
The historical reliance on seats as the primary driver of recurring revenue has proven to be a vulnerability in the age of AI. As AI agents become more sophisticated, they can automate tasks, streamline workflows, and reduce the need for individual user licenses. This directly impacts the "land and expand" model, where growth was predicated on adding more users over time. When AI can perform the work of multiple users, the expansion potential diminishes, and in some cases, can even lead to a contraction in seat-based revenue.
The shift in enterprise spending towards AI infrastructure is a clear indicator of where future growth lies. Companies that are building the tools, platforms, and infrastructure that enable AI development and deployment are well-positioned to capture significant market share. Conversely, application providers that fail to integrate AI into their offerings or whose core functionality is rendered redundant by AI face an uphill battle.
The market’s reaction to these trends has been swift and decisive. The de-rating of software stocks that are not perceived as AI beneficiaries reflects a fundamental reassessment of risk and reward. Investors are now prioritizing companies that are directly enabling or benefiting from the AI revolution, leading to a bifurcated market where the future prospects of software companies are increasingly defined by their relationship with artificial intelligence. The era of the predictable B2B software annuity is over, replaced by a dynamic and rapidly evolving landscape where adaptability and AI integration are paramount for survival and success.
