Coaching is built on dialogue, which is both its great strength and its hidden weakness. This paper sets out the scientific, commercial and ethical case for validated psychometric measurement in coaching, reviews the evidence base, answers the common objections, and offers a practical measure-coach-re-measure framework.
The coaching problem we don't like to talk about
Without an external reference point, even the most experienced coach is operating inside a shared bubble with the coachee — one in which the coachee's self-perception is the primary data source and the coach's pattern recognition is the primary instrument.
Three well-documented phenomena make that risky. Self-ratings on leadership behaviour correlate only modestly with the ratings given by peers, direct reports and managers, with meta-analytic estimates of self-other agreement typically between r = .14 and r = .35. People with the lowest competence in a domain tend to overestimate their ability most severely. And coaches, like clinicians, are vulnerable to confirmation bias and to over-applying whichever framework they trained in most recently.
Validated psychometrics do not eliminate these problems, but they introduce a second, independent data source that coach and coachee can both interrogate.
What the evidence shows
Meta-analyses of organisational coaching report positive effects across performance, well-being, coping, work attitudes and goal-directed self-regulation, with notably larger effects where coaching incorporates multi-source feedback.
Measurement-led engagements also report the strongest returns. The PwC/ICF global coaching study reported a median return of seven times the cost of the engagement; the earlier MetrixGlobal study put the average at 5.7 times. The common factor in the highest-return engagements was structured measurement at baseline and re-measurement at close. Without it, return on investment is asserted; with it, it is evidenced.
Why organisations should insist on measurement
A validated assessment at the start of an engagement creates a defensible baseline that survives changes in coach, sponsor or reporting line. Without it, the organisation is buying a relationship; with it, the organisation is buying a measurable change programme.
Common instruments across a coaching panel also make cohorts comparable, surface systemic patterns invisible at the individual level, and give the organisation an answer — rather than an anecdote — when a board asks what the coaching spend produced.
The measure-coach-re-measure loop
We recommend a simple, defensible loop on every engagement: baseline before the first session, a dedicated shared interpretation session, goal contracting tied to the data and agreed with the sponsor, six to twelve coaching sessions with the data referenced at agreed checkpoints, re-measurement at programme close, and a sustainment check six to twelve months later.
The case for psychometrics in coaching is, in the end, the case for taking coaching seriously as a professional practice rather than a craft. The instruments exist. The evidence exists. What remains is the discipline to use them.