You're going full circle here. He's not the one advocating for "modern evidence-based research and training", so it's not in his standards. And I'm not blaming him. Not everything is measurable, or at least measurable in ways that would enable researchers to extract meaningful answers or techniques. You said it yourself, productivity is a fuzzy end-goal.
If you do not believe it is possible to empirically derive a way to interview properly, then rationalism is your best choice, i.e. Intuition/Deduction. And contrary to common belief, rationalism is not a prison. It's just a decision process, in which you're perfectly allowed to question yourself using available empiric data.
If you do not believe it is possible to empirically derive a way to interview properly, then rationalism is your best choice, i.e. Intuition/Deduction. And contrary to common belief, rationalism is not a prison. It's just a decision process, in which you're perfectly allowed to question yourself using available empiric data.