I've been watching the "SaaS is dead" conversation with an eyebrow raised. Af...

I've been watching the "SaaS is dead" conversation with an eyebrow raised. After digging into the data, the market (software stocks are down 20% YTD), reading a bunch, I think the narrative is half right. But it's half wrong in a way that matters.

SaaS isn't dying. It's splitting in two.

One half: AI wrappers. Thin UIs over APIs. Tools some founder vibe-coded over a weekend. These are commodities now. Switching cost is zero because the value is zero.

The other half: Systems of Record. ERPs, CRMs, compliance-heavy platforms where your actual data lives. These aren't going anywhere. And this is the part the Twitter thread bros miss: migration is organizational trauma, not a technical problem.

Alton Brown has a rule: no unitaskers in the kitchen. That avocado slicer? Gone. Your oven? Non-negotiable. Systems of record are your oven. AI wrappers are the drawer full of gadgets you forgot you bought.

SaaS that solves a painful, specific problem? With data lock-in and workflow integration baked in? That's not dying. That's your oven.

But here's the flip side: if a tool is core to your secret sauce and you're only using 10% of what you're paying for, now might be the time to own it. Tokens are cheap. Building a focused internal tool that does exactly what you need — and nothing else — has never been more accessible. If your Salesforce bill is six figures and you're using three features, the math has changed.

Generic workflow tools that AI can replace with a prompt and an API call? That's dead. It just doesn't know it yet.

Seeing this split in your own stack?