The enterprise software industry is experiencing what may prove to be the most consequential structural repricing since the dot-com correction. Since the start of 2026, approximately $2 trillion in market capitalisation has evaporated from companies that spent the previous decade building what appeared to be unassailable positions in enterprise software. Salesforce and Adobe have seen their share prices decline by more than 25 per cent. Price-to-sales ratios across the sector have compressed from 9x to 6x — levels not witnessed since the mid-2010s. Wall Street has christened it the SaaSpocalypse. But this is not a panic. It is a pattern. And Clayton Christensen mapped it precisely.

The Framework: Disruption Theory Applied

Christensen's theory of disruptive innovation, articulated in The Innovator's Dilemma, describes a specific mechanism by which incumbent firms lose market dominance. The disruption does not arrive as a superior product competing on established performance metrics. It arrives as a cheaper, simpler alternative that initially underperforms on the dimensions incumbents measure — but outperforms on dimensions the market is beginning to value more highly.

AI agents are the disruptive technology. They do not replicate what a human operator does inside Salesforce or Adobe Creative Cloud; they redefine the unit of work entirely. Where the incumbent model charged per seat — assuming that software value correlated with the number of human operators — the disruptive model charges per outcome, per action, or per resolution. The fundamental assumption that software requires human operators has collapsed, and with it, the revenue architecture of an entire industry.

This is Christensen in its purest form. The incumbents are not failing because their products are inferior. They are failing because their business models are optimised for a world that no longer exists.

The Evidence: Seat Counts Are Falling

The data confirms the disruption thesis. When a single employee equipped with AI agents can accomplish the work previously requiring five software licences, the per-seat model does not merely compress — it inverts. Customers are actively reducing seat counts rather than expanding them, precisely because AI-augmented workers accomplish more with fewer tools.

Salesforce's response has been instructive. The company has introduced its Agentic Enterprise Licence Agreement, a hybrid model charging $0.10 per autonomous action via Flex Credits. Microsoft has layered agent-metered billing atop its existing seat licences. Zendesk and Intercom now charge per resolution alongside traditional subscriptions. Every major vendor has arrived at the same messy hybrid, attempting to bridge two fundamentally incompatible revenue logics.

This is what Christensen would recognise as the incumbent's dilemma in its most acute form: the company can see the disruption clearly, yet cannot cannibalise its existing high-margin revenue stream quickly enough to match the economics of the new model. Salesforce's existing subscription business generates predictable, high-margin recurring revenue. Its action-based pricing generates micro-transactions of uncertain scale. The market's scepticism — reflected in that 26 per cent share price decline — is rational.

The Strategic Landscape: Who Survives?

BCG's January 2026 research offers a sobering complement to this analysis. Only 5 per cent of companies are creating substantial value from AI at scale, despite 88 per cent of firms using it in at least one function. This mirrors Christensen's observation that incumbents typically pursue sustaining innovations — incremental improvements within existing business models — whilst ignoring the disruptive trajectory until it is too late.

The companies most likely to survive this disruption are those willing to do what Christensen prescribed: establish autonomous units with independent business models. This is precisely what we observe in the firms thriving amid the SaaSpocalypse — AI-native companies that never built their revenue architecture around seat counts in the first instance. They charge for outcomes delivered, workflows completed, or value generated. Their cost structure assumes AI agents as the primary operators, with humans in supervisory roles.

The Harvard Business Review argued last month that when every company can access the same AI models, organisational context becomes the competitive advantage. This is correct, but it understates the magnitude of the shift. Context is not merely a differentiator — it is the new basis of competition, replacing the installed base and switching costs that protected incumbents for two decades.

The Contrarian View

The bear case for this analysis is not trivial. Enterprise software incumbents possess three assets that disruptive entrants lack: deep integration into mission-critical workflows, vast proprietary data sets, and established trust relationships with procurement departments. Christensen himself acknowledged that disruption is not inevitable when incumbents control distribution and customer relationships. Salesforce's installed base of over 150,000 enterprise customers represents formidable inertia, and the transition to outcome-based pricing — however painful — may ultimately succeed precisely because those customer relationships provide the runway to experiment.

Moreover, the $285 billion valuation wipeout may prove to be an overcorrection. Markets are notoriously poor at pricing structural transitions in real time, and the hybrid pricing models now emerging may stabilise faster than current sentiment suggests.

The Strategic Imperative

For executives navigating this disruption, the prescription is clear. First, audit your exposure: if your enterprise software costs are predominantly seat-based, model the financial impact of a 30 to 50 per cent seat reduction over 24 months. Second, demand outcome-based pricing from vendors — the leverage has shifted decisively to buyers. Third, and most critically, examine your own business model for per-unit assumptions that AI agents may disrupt. The SaaSpocalypse is not confined to software companies. Any business model predicated on the assumption that value scales linearly with human headcount is vulnerable to precisely the same mechanism Christensen described. The seat was never the product. The outcome was. The market is simply, and belatedly, recognising that fact.