Terraform is certainly useful and when it works it works. The problem is that when it doesn't, things have a tendency to go horribly wrong. The amount of hours saved by terraform are barely offset by the amount of time spent recovering broken production infrastructure or resolving state issues because it threw in the towel halfway through some operation. That's without even digging into plan and apply discrepancies.
So, declarative schema migrations, yay, and market yourself that way. But I would be careful with being "terraform for X". There's a lot of terraform trauma among people who run production systems.