🚪 The SaaSpocalypse is a sorting, not a slaughter

Jefferies called it the SaaSpocalypse, and the label stuck. The IGV index has dropped 30% from its peak. ServiceNow shed 40% in four months, despite results that beat consensus. The catalyst was narrow, almost absurd: Anthropic released a legal tool, and the market concluded that general-purpose AI would eat specialised software whole, one seat at a time.

Bank of America spotted the contradiction. Investors are pricing in two beliefs at once: that AI investment will deliver poor returns, and that AI will be powerful enough to make SaaS obsolete. Pick one.

Then came Visma. The Norwegian vertical SaaS group was set to be London’s largest IPO this year. A €19bn valuation, €2.8bn in revenues, fourteen banks on the ticket. Hg, its British private equity owner since 2006, had chosen London over Amsterdam following the UK listing reforms. In late March, the deal was shelved until 2027. Not because the business had deteriorated, but because public markets have lost their appetite for anything with « SaaS » in the pitch.

The trouble with panic is that it flattens distinctions. Datev, the Nuremberg cooperative founded in 1966, turns over €1.3bn a year and runs 14 million payslips a month for 750,000 German businesses. The software encodes the Sozialversicherung, federal and regional tax regimes, twenty sets of overlapping rules. Personio, valued at $8.5bn, built its HR platform on German payroll complexity. Visma sells accounting tools to 2.1 million Nordic SMEs under regulatory frameworks that shift every fiscal year. No large language model is going to replicate this. The job is not to automate a workflow. It is to encode a jurisdiction.

Twenty-seven tax systems. As many labour law frameworks. GDPR underneath all of it. The regulatory fragmentation that makes scaling European software so painful also makes displacing an incumbent nearly impossible. A Californian competitor cannot localise a language model and call it a payroll product. Compliance is not a feature you bolt on. It is the product. Hg grasped this before the panic: sixty B2B portfolio companies across Europe, more than €100bn in assets. The Visma IPO is delayed, not dead.

Horizontal SaaS, the kind that sells generic collaboration at premium prices, faces a question it cannot answer. Vertical European SaaS, the kind that thrives because regulation is dense and local, does not face the same threat. The SaaSpocalypse is not an extinction event. It is a sorting. The companies that encode local complexity will remain standing, probably cheaper than they deserve to be, and harder to displace than the sell-off implies.