The SaaS Apocalypse
The market is panicking. The disruption is real. The opportunity is enormous.
Jefferies traders are calling it the “SaaSpocalypse.” Their words: “get me out” style selling across the entire software sector.
The numbers are brutal. The iShares Expanded Tech-Software Sector ETF is down 20%+ year-to-date. ServiceNow dropped 28% YTD—down 50% from its peak despite beating earnings nine straight quarters. Salesforce is down 26%. Intuit down 34%. Thomson Reuters plunged 16% in a single day after Anthropic released a productivity tool for in-house lawyers on February 3rd.
Software has underperformed semiconductors by 20% over the last 20 days. That’s the largest gap since the dot-com bubble peak in February 2000.
The panic is real. The disruption is real. But the market isn’t distinguishing between the dying and the adapting.
The Structural Break
SaaS was built on per-seat pricing. More employees meant more licenses meant more revenue. The entire business model assumed human headcount would keep growing.
AI agents break this equation. If 10 AI agents do the work of 100 sales reps, you need 10 Salesforce seats, not 100. That’s a 90% reduction in seat revenue for the same work output.
Three forces are compounding this:
AI replacing point solutions. Many SaaS tools exist because a workflow was tedious enough to justify $20-200/month. Scheduling, basic document generation, simple analytics dashboards. AI handles these natively now as part of general-purpose tools.
Consolidation into platforms. Microsoft, Google, and Salesforce are embedding AI into existing suites. Enterprises want fewer vendors, not more.
Vibe coding. Displaced engineers and technical users can build focused internal tools that replace off-the-shelf SaaS. A CRM for general contractors was reportedly built in Replit in 20 hours. Not enterprise-grade, but good enough for a lot of use cases.
Enterprise IT budgets aren’t shrinking—they’re growing around 8%. But the growth is going to AI infrastructure and tooling. Hyperscalers alone will spend $470 billion on AI infrastructure in 2026. Every dollar going there is a dollar not going to another Salesforce seat, Workday module, or ServiceNow add-on.
SaaS is being starved, not killed outright. Budget reallocation rather than withdrawal.
Who Dies
The chaff is obvious: seat-based, workflow-heavy horizontal applications with shallow moats.
Companies where the entire value proposition was “we saved you from doing this tedious thing manually” with no deeper lock-in. The $30-50/month/seat tools for scheduling, expense tracking, lightweight project management, basic analytics.
These are one prompt away from being rebuilt internally. Many of these products didn’t need to exist as standalone businesses—AI just made that obvious.
Zoho’s CEO put it directly: “An industry that spends vastly more on sales and marketing than on engineering and product development was always vulnerable. The venture capital bubble and then the stock market bubble funded a fundamentally flawed, unsustainable model for too long. AI is the pin that is popping this inflated balloon.”
Who Survives
Companies with deep data assets, mission-critical infrastructure, and genuine platform ecosystems.
Salesforce is actually in a decent position—not because of the CRM software itself, but because of the ecosystem. Thousands of third-party apps on AppExchange. Deep integrations into enterprise tech stacks. Massive proprietary customer data. Decades of institutional knowledge embedded in how companies run.
The strategic question for survivors: can they shift from selling seats to humans to selling compute and orchestration for agents? If Salesforce makes Agentforce genuinely useful—AI agents operating within their ecosystem, acting on real customer data—they expand their market rather than defend it.
The premium holds if they become the platform agents run on rather than the tool humans click through.
Cybersecurity leaders survive regardless. You still need security whether humans or AI agents are doing the work. Companies where the value is domain expertise delivered through software, not the software itself.
The Second Wave
The current conversation focuses on software eating itself. The next wave is bigger.
Tens of thousands of displaced, highly technical engineers will turn their attention outward. They understand AI tools better than almost anyone. They know how to build products. And they’ll target industries that have never dealt with this competitive pressure.
Legal, accounting, consulting, real estate, insurance, healthcare administration, education. These industries were insulated by credentialing, institutional relationships, and industry-specific complexity. A displaced engineer partnering with one domain expert can now build something that threatens incumbents who’ve been comfortable for decades.
These small businesses aren’t trying to be unicorns. They’re building something specific, charging a fair price for real value. The economics work because the cost structure is tiny—no massive sales org, no offices, no management layers.
The AI democratization everyone talks about won’t happen top-down as some idealistic tech narrative. It happens bottom-up out of necessity. Displaced workers using powerful tools to survive.
The Opportunity Window
Right now there’s a brief period where incumbents are paralyzed, the market is panicking, talent is flooding onto the market, and the tools are powerful but most people haven’t figured out how to use them.
The gap between what’s possible and what most people realize is possible—that’s where the opportunity lives.
It closes over time as everyone catches up. But right now it’s enormous.
Chaos is a competitive advantage for those who can stomach it. Most people are risk-averse and freeze under uncertainty. The person who can tolerate ambiguity and make decisions with incomplete information has a massive edge.
Morgan Stanley’s SaaS basket has lagged the Nasdaq 100 by 40 percentage points since December 2024. Some of that is justified—companies that genuinely should be repriced. Some of it is panic selling healthy companies alongside dying ones.
By the time things settle down and the market starts doing nuance again, the opportunities will have been claimed. The window is now.