A more useful question is: what kinds of SaaS are actually getting funded today?
Over the last couple of years the barrier to building software has dropped significantly. AI tools make it easier to build interfaces, automation layers, and basic functionality quickly. As a result, products whose differentiation lives mostly in UI or thin workflow automation are harder to defend.
That's why generic productivity tools, project management platforms, CRM clones, and thin AI wrappers built on top of existing APIs are finding it harder to raise funding and survive.
The companies that still attract strong interest tend to have a few characteristics:
- products embedded in real operational processes
- ownership of structured or proprietary data
- deep domain expertise in a specific industry
Reproducing an interface is relatively easy. Reproducing years of accumulated data, integrations, and operational trust is much harder.
Another shift is pricing and distribution. Seat-based SaaS models are under pressure, especially when AI agents can replace or compress human workflows. Consumption-based pricing and value-linked pricing models are becoming more common.
For startups this means the bar is higher. A large codebase is no longer an advantage by itself. What matters more is speed, focus, and a clear understanding of the workflow you’re improving.
Curious how others here see it, especially founders building SaaS products today.