How do you evaluate clinical decision support tools before deploying them?

I'm working on a project where we've built a clinical decision support tool and are now figuring out how to test it properly before it goes anywhere near real patients.

There's a lot of guidance on usability testing and retrospective validation, but less on how to sequence these evaluations in practice.

I'd like to hear how others in health IT have approached this, especially when you have limited time and no access to a full clinical trial.