$300 billion evaporated from software stocks in 48 hours. Wall Street calls it the "SaaSpocalypse." And it's hitting EPM vendors hard.
$300 billion evaporated from software stocks in 48 hours.
If you missed this headline in the last weeks, you were probably buried in month-end close and reporting. I understand the feeling.
Between February 3rd and 4th, 2026, the cloud software sector had its worst week in history. Atlassian dropped 35% in a single week. The iShares Tech-Software ETF retreated 30% from its late-2025 highs. The catalyst? Agentic AI platforms demonstrating they can handle complex business workflows without traditional software interfaces.
Wall Street calls it the "SaaSpocalypse." And it's hitting EPM vendors hard.
"If one AI agent can do the work of ten analysts, why are we paying for ten seats?"
That question captures the structural shift happening right now. The seat-based pricing model that built the SaaS industry is collapsing. And Enterprise Performance Management is in the blast radius.
Four forces are converging:
1. AI agents are automating human tasks at scale. AI can now draft variance explanations, handle support tickets, prepare journal entries, and write ad copy. OpenAI's o3 model costs dropped 80% in two months while accuracy improved. The economics are brutal: within three years, any routine rules-based digital task could move from "human + app" to "AI agent + API."
2. Per-seat pricing is dying. If AI handles the workload of ten finance analysts, charging for ten seats becomes absurd. The industry is scrambling toward outcome-based and usage-based pricing models. AI-native companies are abandoning seat-based pricing entirely.
3. "Disposable software" is real. Enterprises are realizing they can build micro-apps tailored to specific workflows. The cost and complexity barrier for custom software has dropped to near zero. Why pay for a bloated, generic subscription when you can spin up exactly what you need?
4. Data moats are the new moat. AI can replicate workflows. What it can't replicate: proprietary data and domain expertise. Financial reporting systemised models, the "source of truth", become the only defensible position.
I track three vendors closely: CCH Tagetik, Anaplan, and OneStream. Each is playing a different hand.
CCH Tagetik is all-in on "Expert AI." Their Intelligent Platform features "Finance Brain", AI that learns from customer data and applies financial logic. They're leaning hard into agentic workflows purpose-built for finance. Named Leader in the 2025 Gartner Magic Quadrant and the 2026 Nucleus Research CPM Value Matrix. Their bet: deep regulatory compliance expertise and embedded financial logic create barriers generic AI can't easily cross. It's a defensible position for now.
Anaplan is preparing to re-IPO, which takes a certain kind of courage given the timing. They just announced role-based AI agents and "CoModeler" (an AI that creates and tests scenarios instantly). General availability expected Q1 2026. The PE owners (Thoma Bravo) have been focused on margin expansion since the 2022 take-private. Now they're betting the market will buy an AI-enhanced connected planning story. The IPO filing in January, with the February selloff in full swing, is either brilliant contrarianism or terrible timing. We'll know soon.
OneStream made the most telling move: going private again. Just 17 months after their $490 million IPO at a $6 billion valuation, they announced a take-private deal with Hg for ~$6.4 billion. They've deepened their Microsoft AI partnership and are positioning as a unified platform with "one source of truth." But the retreat from public markets says something about how Wall Street is pricing AI transformation potential right now.
Here's what I'm telling clients, and how you should handle these conversations:
The leverage has shifted. Vendors are desperate to prove AI value. Pricing pressure is real. If you're negotiating renewals or new deals in the next 6-12 months, push harder.
Vendor will say: "Our AI features are included in your enterprise license."
You should ask: "What happens to our per-seat pricing when AI reduces our license needs by 30%? Will you restructure the contract, or are we locked in?"
Vendor will say: "We're investing heavily in AI capabilities."
You should ask: "Show me your roadmap and what pricing model will you use for those capabilities. If AI delivers the productivity gains you're promising, what should I expect in contract renewal?"
Watch for consolidation. Smaller EPM players will get acquired. Choose vendors with clear AI-native strategies and stable ownership.
Vendor will say: "We're well-capitalized and committed to the market."
You should ask: "Are you exploring acquisition offers? The OneStream flip happened 17 months post-IPO. Why should I expect stability from you?"
Interrogate the AI integration. 86% of North American finance teams are still in early stages of AI adoption. The vendors selling hardest are often the ones furthest behind.
Vendor will say: "Our AI is deeply integrated into the platform."
You should ask: "Is this embedded in the core, or a bolted-on copilot? What agentic capabilities exist today (not on the roadmap)? Which AI model and how does your AI learn from my data specifically?"
Protect your data. In regulated finance environments, "explainable AI" isn't a nice-to-have, it's a requirement.
Vendor will say: "Your data is secure and private."
You should ask: "Is my data used to train models accessible to other customers? What audit trail exists for AI-generated outputs? Can your AI explain its recommendations to my auditors?"
Bain's Technology Report 2025 mapped software workflows into five scenarios:
Scenario 4 is the one that should keep EPM vendors up at night. With its high automation potential, and easy for AI to replicate. The survival play? Scale your data moat and double down on AI automation.
The vendors who survive will be the ones who own unique, proprietary data and can demonstrate domain expertise that generic AI can't replicate. Deep financial consolidation logic. Regulatory compliance knowledge. Industry-specific workflows.
That's a smaller list than most software companies would like to admit.
The SaaSpocalypse isn't about AI replacing finance teams. It's about AI reshaping how finance teams buy and use software.
The seat-based SaaS model built a trillion-dollar industry. The $300 billion that evaporated in 48 hours? That's just the market's first repricing of what AI means for that model. More repricing is coming. EPM vendors are responding with AI features, partnerships, and strategic pivots. Some more convincingly than others.
For finance leaders, this is a moment of leverage. Vendors need you to validate their AI story. Use that.
Ask the hard questions. Negotiate the outcome-based deals. And pay close attention to which vendors are running towards AI, and which are running scared.
Got questions about your EPM vendor's AI strategy? Reply to this email. I read everything.
Want more insights like this?
EPM Insider drops every Tuesday. Practical perspectives on EPM, AI, and finance transformation.
Subscribe Free →This is Issue #001. Subscribe to be notified when new issues drop every Tuesday.
Subscribe to EPM Insider →