The SaaSpocalypse as Disruption Case

In the AI Story

The structural conditions for disruption were present and accumulating for years before the threshold was crossed. Enterprise SaaS platforms had overserved their customers through relentless feature accumulation. The median user of an enterprise platform used between five and fifteen percent of available features. The population of small businesses, individual professionals, and specific departments who needed simple tools rather than comprehensive platforms was enormous — and largely served, inadequately, by platforms whose pricing and complexity reflected requirements they did not share.

AI-generated custom tools entered this market from below. A marketing manager who needed a CRM for twenty clients could build one in an afternoon at the cost of an API subscription, rather than paying Salesforce hundreds of dollars per seat per month for features she would never use. A small accounting firm could build its own workflow automation rather than licensing complex enterprise platforms. Each of these substitutions was individually modest; collectively, they represented a structural shift in the software market.

The mini-mill parallel held with precision. SaaS companies initially welcomed the loss of low-end customers. These were the least profitable customers; ceding them improved the product mix and quarterly metrics. Each company moved upmarket, emphasizing enterprise features, deeper customization, and premium services. The pattern was rational given the existing value network. It was also, structurally, the same pattern integrated steel mills followed from 1965 to 1990.

The market recognized the structural inevitability before the full trajectory had played out. The February 2026 repricing reflected not current revenue losses but the market's forward-looking assessment of where the trajectory was heading. Fred Pope's analysis — "AI-assisted development isn't a sustaining innovation that makes existing SaaS products incrementally better. It's a disruptive force that enables entirely new entrants to build comparable products at a fraction of the cost" — captured the market's recognition in a sentence.

The outcomes varied by firm position. Companies whose value was primarily functionality — features that perform specific tasks — were exposed because AI generates equivalent functionality at near-zero cost. Companies that built genuine ecosystems — accumulated data, integrations, institutional relationships, workflow knowledge — had defensible positions above the commoditizing code layer. The distinction was not between "good" and "bad" SaaS companies but between companies whose moats operated at the functionality layer (which was being commoditized) and those whose moats operated at the ecosystem layer (which AI could not easily replicate).

Origin

The SaaSpocalypse as a contemporary event is documented in Segal's The Orange Pill (2026). The application of Christensen's framework to the event draws on analysis by Fred Pope, the Christensen Institute's ongoing research, and Aswath Damodaran's valuation work.

Key Ideas

Textbook low-end disruption. The SaaSpocalypse exhibits every structural feature Christensen documented in earlier industries.

Compressed timeline. What took two decades in steel took eight weeks in software market recognition.

Overserving as precondition. Years of feature accumulation created the gap between platform capability and median user need that disruption exploited.

Ecosystem survives, functionality doesn't. Firms with moats above the code layer retained value; firms with moats at the code layer did not.

Market forward-looking. The repricing reflected where trajectories were heading, not where current revenues had landed.