The only things academics produce are publications. Those are the only widgets we make. Productivity in teaching would be what, the number of students taught? Then a large section is always more productive than a small one. Power-point lectures to 400 students are the most productive. Once again, then, teaching and research are defined so differently that the very definition of productivity is wildly different: the graduate prof with a lot of publications is likely to teach fewer courses with fewer students and be less productive, however brilliant her teaching.
Economists talk about productivity as how much work each employee can churn out per hour, so that is a measure of efficiency.
One of my goals this summer is to learn to distinguish, by ear, the various palos of flamenco, because I want to teach a course on this some day. This is productive work but not efficient. All the work on the scholarly base seems highly unproductive, in fact, because it doesn't lead directly to articles and books, and only some of it filters down into classes taught. I could never justify all my work on the scholarly base in the management-speak of "productivity."