Toronto, Canada - February 23, 2025: Mobile apps on a smartphone: ServiceNow, Salesforce, and HubSpot.
getty
Every CTO in America is fielding the same question from their board right now: which of our software vendors is about to become irrelevant? It’s a fair question. In February, $285 billion in SaaS market cap evaporated in 48 hours after AI agents demonstrated they could execute complex business workflows autonomously. The financial press dubbed it the “SaaSpocalypse,” and enterprise software stocks posted their worst quarter since the 2008 financial crisis.
But the panic is broader than the problem. The market is treating all software companies like they’re standing in the same burning building. They’re not. Some are holding the match. Others are selling the fire extinguisher. And if you’re a business leader trying to figure out which of your vendors belongs in which category, the distinction has never mattered more.
Make no mistake: genuine forces are reshaping enterprise software. Per-seat pricing, the business model that powered two decades of cloud growth, is facing compression. When an AI agent can do the work of three knowledge workers, enterprises don’t need three seats. Margins built on switching costs are being challenged by AI coding tools that make it cheaper and faster to build alternatives.
Companies like Notion, Asana, and Monday.com should be concerned. Their core products (task management, project tracking, lightweight collaboration) sit squarely in the kill zone. These are categories where AI agents don’t augment the software; they replace the need for it entirely. When an agentic system can manage projects, assign tasks, track dependencies, and update stakeholders without a human ever opening a browser tab, the value proposition of $10 per seat per month collapses.
But extending that logic to all enterprise software mistakes a category correction for an extinction event.
While SaaS stocks were cratering, enterprise software spending quietly rose 15% to $1.4 trillion in 2026. Global IT spending is projected to hit $6.3 trillion. AI spending alone will reach $2 trillion this year. The money isn’t leaving software. It’s migrating to different software. And the companies capturing it share a specific profile.
Palantir’s revenue grew 70% year-over-year last quarter, with U.S. commercial revenue up 137% and adjusted operating margins at 57%. CrowdStrike grew revenue 22% to $4.8 billion and added over $1 billion in net new ARR. AI doesn’t eliminate the need for cybersecurity. It dramatically increases it. ServiceNow’s AI product, NowAssist, generated $600 million in contract value in 2025 and is on pace to exceed $1 billion this year. Salesforce’s AgentForce ARR soared 169% in a single quarter.
These aren’t survivors. They’re accelerants. AI didn’t threaten their business models. It gave them a new product line to sell on top of an existing platform that enterprises already can’t live without.
If you’re a CEO or CTO evaluating your software stack, the question that matters is brutally simple: is this vendor a feature or a platform?
Features get absorbed. Platforms absorb. The difference comes down to three things.
Proprietary data moats. Palantir doesn’t just sell software; it sits on operational data that becomes more valuable as AI models consume it. CrowdStrike’s Threat Graph ingests trillions of security signals daily. No AI agent built on a foundation model can replicate a decade of proprietary telemetry. If your vendor’s value is rooted in data you can’t get elsewhere, they’re defensible.
Workflow gravity. ServiceNow and Salesforce aren’t tools people use. They’re systems enterprises can’t operate without. When a vendor is the system of record for IT operations or customer relationships, AI doesn’t replace them. It flows through them. Ask yourself: if this software disappeared tomorrow, would your operations stop? If yes, that’s a platform. If you’d just find a workaround by Thursday, that’s a feature.
Pricing power in the AI era. The companies winning have moved beyond per-seat models entirely, layering consumption-based AI pricing on top of existing subscriptions. ServiceNow charges for NowAssist. Salesforce charges for AgentForce. CrowdStrike charges for Charlotte AI. They’ve turned the AI disruption narrative into a revenue multiplier. If your vendor hasn’t shipped an AI product you’re willing to pay for, that tells you something about their position.
The SaaSpocalypse isn’t a reason to freeze software budgets. It’s a reason to reallocate them. Every enterprise should be conducting a ruthless audit of its software stack through the feature-versus-platform lens.
For the platforms, the vendors with proprietary data, deep workflow integration, and credible AI products, this is the moment to deepen the relationship. They’re going to be more important to your operations in three years than they are today, and their AI capabilities will compound on your existing investment.
For the features, the lightweight tools with horizontal functionality that an agentic system could replicate, start planning the transition. Not because the tools are bad, but because the economics are about to shift beneath them. Why pay per-seat for project management when an AI agent manages projects natively inside the platform where your work already lives?
I’ve said for months that we aren’t witnessing a SaaSpocalypse. We’re witnessing the birth of the agentic platform era, where the value of enterprise software isn’t the interface but the intelligence layer underneath. The companies building that layer, whether they’re Salesforce embedding AgentForce or CrowdStrike embedding Charlotte AI, are positioning themselves as the operating system for how work gets done in the age of AI.
The companies that treat this as a platform shift, not a feature addition, will define the next decade of enterprise software. The ones still selling seats to a workflow will find that the workflow no longer needs them.
That’s not an apocalypse. It’s an evolution. And the leaders who understand the difference will make better technology decisions, better vendor bets, and better strategic investments than the ones reacting to headlines.
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。