The SaaS-pocalypse — AI Displacement, Overhiring Hangover, or Multiple Compression?

Narrative

Public SaaS growth rates have declined every single quarter since the 2021 peak, establishing the foundational case for structural deceleration independent of AI. Software forward P/E multiples have fallen to 22.7x by early 2026—below the S&P 500 overall market multiple for the first time in history, with total market capitalization lost across the software sector since the correction began in late January exceeding $2 trillion. The trigger was Anthropic's Claude Cowork announcement in February 2026, which sent public software stocks into a selloff that wiped out $285 billion in market value overnight. However, the Substack narrative fracture becomes evident when comparing fundamentals to valuation: Software and services earnings growth in the S&P 500 was projected to accelerate to 21% in 2026 from 17%, yet investors sold in response to the AI threat despite solid fundamentals. On the seat displacement question, evidence is mixed: Atlassian reported its first-ever decline in enterprise seat counts in early 2026, while Workday announced an 8.5% workforce reduction citing AI-driven efficiency gains, but Pave reports hiring for software engineers is stable with demand shifting toward experienced talent; if AI were systematically displacing engineers, hiring would be dropping, but it is not. The compression of per-seat pricing is real: purely per-seat pricing adoption plummeted from 21% to 15% over the last twelve months, as 70% of enterprises now demand usage-based or outcome-based contracts. Recovery narrative emerged by mid-April: forward P/E multiples compressed to 22.7x in March, triggering a 'valuation floor' created by massive private equity interest preparing take-private bids for mid-cap SaaS that had halved despite remaining cash-flow positive.