Programs don’t have an evaluation-intent problem — they have a workflow-friction problem.
Use micro-interactions that can be completed in seconds after clinical contact.
Start with simple prompts and deepen only after repeated interactions.
When enough data is collected, trigger the formal evaluation automatically.
Result: better data quality, less coordinator chase, better learner feedback.