Executive coaching or executive assessment?

Coaching works, on average

Rebecca Jones, Stephen Woods and Yves Guillaume pooled 17 studies of workplace coaching by internal or external coaches in 2016. Coaching improved outcomes overall, with an effect size of 0.36. The largest effect was on attitudes and motivation, measured by the coachees’ own reports, at 0.51. Skills, mostly measured by observation, improved less, at 0.28.

Those are real effects. They are also averages. The meta-analysis did not separate coaching aimed at a specific behavior from coaching that started without one.

Feedback can make performance worse

Avraham Kluger and Angelo DeNisi reviewed 607 effect sizes from studies of feedback on performance in 1996. Feedback helped on average, with an effect of 0.41. But over 38 percent of the effects were negative, and sampling error could not explain them. Feedback did least when it drew attention away from the task and toward the person receiving it.

What happened when coaching included 360 feedback

Jones and her colleagues also compared coaching that included multi-source feedback, the 360 ratings that often open an engagement, with coaching that did not. Both helped. Coaching without the 360 had an effect of 0.88. Coaching with it had 0.21, and the confidence intervals did not overlap.

The study does not say why. We think it fits Kluger and DeNisi’s finding. A 360 report is feedback about the person. It can pull a coaching relationship toward the self before anyone has defined the task. That is an inference, and a small set of studies supports it.

Begin with the uncertainty

If the organization can identify the behavior that has to change and show that it costs performance, coaching is the right tool. Coaching then means practice, feedback on the task, and a date to check the result.

If a capable executive began struggling after a promotion, a restructuring or a change in tempo, the source is less clear. It may lie in the person, the role, the people around the role, or the fit between them. Coaching started at that point has nothing specific to work on, so the assessment should come first.

What the assessment should hand the coach

A useful assessment ends with a brief a coach can work from. The brief gives one behavior to try, the signal that will show whether it held, and a date to review it. It also states the evidence that would show the diagnosis was wrong. P×E produces that shape of recommendation, and a coach can take it from there.