Evidence-based management — Rousseau's case for treating HR like a science

60-Second Summary

EBMgt = best available evidence + context + stakeholders + judgement.
'Evidence' includes scientific research, internal data, practitioner experience, and stakeholder values — not one of these alone.
Most HR fads (forced ranking, MBTI, open offices) fail an EBMgt review.
The discipline is a 5-step loop: Ask, Acquire, Appraise, Aggregate, Apply, Assess.

Medicine moved from tradition to evidence base in the 20th century and saved millions of lives. HR has not yet had its moment. Denise Rousseau (Carnegie Mellon, ex-president of the Academy of Management) is the closest thing the field has to its Hippocrates.

What EBMgt is (and isn't)

EBMgt is the conscientious, explicit, and judicious use of four sources of evidence: scientific research, internal organisational data, practitioner expertise, and stakeholder values. It is not 'data-driven HR' nor 'best practice copying' — both are subsets and both fail when used alone.

The 6-A loop

The disciplined practice

1
Ask
Translate the practical issue into an answerable question.
2
Acquire
Systematically search the four evidence sources.
3
Appraise
Judge each source's quality and relevance.
4
Aggregate
Weigh the evidence as a whole.
5
Apply
Adapt to context, decide, document the reasoning.
6
Assess
Evaluate the outcome and feed back into evidence base.

Why HR fads keep winning

Compelling stories beat boring evidence.
Consultants sell certainty; researchers sell nuance.
Boards want simple narratives at the AGM.
EBMgt is slower in the first cycle and faster forever after.

The litmus test

Before adopting any new HR practice, ask: 'What is the evidence, and what would change my mind?' If neither has an answer, you're buying belief, not method.

What EBMgt is (and isn't)

The 6-A loop

Why HR fads keep winning

Read next