Analysing Human Performance

Klaus Wittkuhn has written an excellent article on the systemic approach required in human performance analysis. This article appears in the March 2004 edition of "Performance Improvement" published by ISPI. The wealth of practical advice in PI is one more reason to become a member of ISPI (unabashed promotion here).

Wittkuhn discusses an aspect of performance analysis that has been bothering me for a while – how can you take a systemic approach when there are overlapping systems as well as multiple sub-systems in any organisation? Where do you start and where do you finish?

Wittkuhn discusses the idea of emerging properties (e.g. the whole is more than the sum of the parts) but also provides a template for intervention, that is practical but considerate of the fact that you cannot engineer human performance. Human performance is an emergent property of an organisation, and is affected by multiple variables.

Witthuhn’s approach for improving performance is to first address what he calls the "Steering Elements". These "ensure that the right product is delivered at the right time to the right place", and include – Management, Customer Feedback, Consequences, Expectations and Feedback.

Once the steering elements have been addressed, then look at the "Enabling Elements" – Management (again), Design, Resources and Support.

Only after the steering and enabling elements (the non-human factors) have been aligned, should you look at work performance. The rationale here is that it is only within an optimized system that we can expect optimal human performance. As Wittkuhn states:

It is not an intelligent strategy to train people to overcome system deficiencies. Instead, we should design the system properly to make sure that the performers can leverage all their capabilities.

This is the most succinct operationalization of performance technology that I have yet read, and I hope that it also makes sense to you. If not, please comment.

2 Responses to “Analysing Human Performance”

Leave a Reply

  • (will not be published)