Everyone’s talking about the SaaS apocalypse.

On a recent episode of the All In Podcast, Marc Benioff sat down with Jason Calacanis, Chamath Palihapitiya, and David Friedberg and said it plainly: “The market’s rerated.” Top enterprise software companies are trading at 2x revenue. $180 billion in market cap has evaporated. The fear is that AI will make most software unnecessary.

Here’s the thing. The fear is partly right. But it’s pointed at the wrong target.

The Low End Is Getting Wiped Out

Chamath said it clearly: “The low end of the market is basically finished.”

He’s talking about the vertical point solutions. The niche SaaS tools that do one thing. The standalone app your team adopted because it solved a specific problem and nobody asked too many questions about how it connected to everything else.

Those tools are vulnerable. If an AI agent can do the job without a dedicated platform, the platform goes away. And increasingly, it can.

David Friedberg shared something interesting. At his company, they’ve spent the last six months dropping vertical software and doubling down on horizontal platforms instead. Not buying off-the-shelf workflow tools. Building specific workflows on top of foundational platforms that can actually support them.

That shift is happening quietly in a lot of companies right now.

The High End Gets Stronger

Here’s where the story flips.

Chamath made the case that the large horizontal platforms, the ones with deep customer relationships and years of data, are actually positioned to win. Not despite AI. Because of it.

Why? Because AI needs context.

Benioff said this directly when he explained the Informatica acquisition: “None of this stuff works if you don’t have context. The AI is very probabilistic. It can kind of figure things out but it needs to be grounded in real data. It needs to have that semantic layer.”

Read that again. AI needs to be grounded in real data.

That is the whole argument.

Your AI tools are only as smart as the data underneath them. If that data is fragmented across a dozen disconnected apps, your AI is guessing. It will be confidently wrong. And you will not catch it until something breaks.

What This Means for Mid-Market Companies

If you are a CFO or a CEO at a mid-market company, you are probably feeling pressure from two directions right now.

On one side: every vendor is telling you their product is “AI-powered” and you need to upgrade.

On the other side: you have heard enough about hallucinations and failed rollouts that you are skeptical anything actually works yet.

Both of those instincts have some truth to them.

Here is what is actually true: the companies that will get real value from AI in the next three to five years are the ones who built clean, connected, trustworthy data foundations first. Not the ones who bolted AI onto a fragmented stack.

Your system of record is not just an accounting tool. It is the data foundation that makes everything else possible. If your financials, operations, inventory, and customer data all live in one connected platform, you have something AI can actually work with. You have context.

If that data is scattered, you have a problem that AI cannot solve. It can only make it worse.

The Boring Answer Is the Right Answer

The most durable software investment you can make right now is not a flashy AI tool. It is a clean, integrated system of record.

This is why we work with NetSuite. Not because it is the only ERP, but because it is built as a single connected platform from the start. Financials, operations, inventory, CRM, and reporting all live in one place, sharing the same data. That is not a feature. That is the architecture Benioff described when he said AI needs a single source of truth to work well. Most mid-market companies trying to get there through integrations and bolt-ons are building that foundation one patch at a time. NetSuite starts there.

Get your data in order. Consolidate where you can. Make sure your ERP is actually being used the way it was designed to be used.

That is not exciting. But Benioff spent $8 billion buying Informatica because context is the bottleneck. He knows what the AI era actually requires.

The SaaS apocalypse is real for the point solutions that were always one good feature away from being replaced. For the systems that hold your company’s truth, the moment is just beginning.

Wondering whether your data foundation is ready for what comes next? Book a strategy session.