SaaS Isn’t Dying — It Was Already Dead

Your CFO just asked the question that ends boardroom friendships: “Why are we paying for five different SaaS tools when our own devs could build something better in a month?”

You laughed it off. Then your engineers, armed with AI coding assistants, came back with a prototype in two weeks. Now the laughter is uncomfortable. The SaaS you’ve paid for since 2019 suddenly looks like a luxury you can’t justify.

This isn’t a hypothetical. One HN user put it bluntly: “Two internal projects to replace Hubspot and Float. Hubspot will be kept, Float will be replaced.” That’s the new reality. In-house replication isn’t a punchline anymore. It’s a strategy.

And here’s what nobody wants to admit: many SaaS companies aren’t dying because of AI. They were already zombies, shambling along on cheap debt from the zero-interest-rate era.

That’s the open secret. During ZIRP, investors threw money at any SaaS that could hit a growth target. Startups built features, not moats. They priced products for venture-scale growth, not for the actual value delivered. For a decade, the math didn’t have to work — the next funding round always did.

Now the money’s gone. The CFO is scrutinizing every subscription. And AI has turned your internal dev team from a cost center into an in-house SaaS factory.

This is the twist that terrifies founders: the same AI tools that SaaS companies use to ship their products are now being used by their own customers to build those products for themselves.

If your SaaS can be recreated by a handful of engineers in a quarter, it never had a moat. It had a vacation from reality.

Data sovereignty is the other nail in the coffin. Regulators, security auditors, and even your own CISO are asking: “Why is this customer data sitting on a third-party server when we could hold the keys ourselves?” Bringing data in-house increases security, robustness, and legal control. That’s not ideology — that’s risk management.

But let’s be precise: not everything will be built in-house. The HN poster said they’re keeping Hubspot. Some tools are too complex, too deeply integrated, or too far outside a company’s core competency. The pattern isn’t “all SaaS dies.” It’s “commodity SaaS dies.”

If your product is a spreadsheet with a pretty front-end, a calendar with a subscription, or a chat widget that a team of three could clone in weeks, your days are numbered. The moat your pitch deck claimed — “first-mover advantage,” “brand,” “UX polish” — is now a puddle.

What survives? The tools that have genuine network effects, proprietary data advantages, or workflows so domain-specific that replicating them would be a decade-long distraction. Those are real moats. Everything else is temporary.

The end of ZIRP didn’t just raise interest rates. It raised the bar for what counts as a defensible business.

So if you’re a founder, stop asking “Is SaaS dying?” and start asking “If my customers had a hundred thousand dollars and three engineers, would they build my product instead of paying me?” If the answer is yes, it’s not AI killing you. It’s the truth.

If you’re an operator, the message is better: the era of getting held hostage by bloated SaaS contracts is ending. You have options. The tools are in your hands, and for the first time in a decade, the economics are on your side.

The SaaS zombie apocalypse isn’t a metaphor. It’s happening. And the survivors won’t be the ones with the most funding — they’ll be the ones with actual gravity.