The claim that executive coaching changes behaviour is both true and, as typically stated, insufficiently precise to be useful. It changes some behaviours in some people under some conditions. The important questions for anyone investing in coaching, whether as a practitioner, a sponsor, or a recipient, are which behaviours, which people, and which conditions. The answers to these questions are clearer than most coaching advocacy acknowledges and more interesting than most coaching scepticism allows for.
The research on coaching effectiveness has grown substantially in the last two decades, from a relatively thin evidence base in the early 2000s to a body of work that, while still methodologically imperfect, supports several conclusions with reasonable confidence. The conclusions are both more modest and more specific than the general claim that coaching produces behavioural change would suggest. Understanding them precisely is the foundation of good coaching commissioning, good coaching practice, and the honest conversation about what coaching can and cannot deliver that most sponsor relationships would benefit from having.
The most consistent finding across the coaching effectiveness research is that coaching produces the most reliable behavioural change in the specific domain of interpersonal behaviour: how leaders communicate, how they listen, how they respond to challenge, how they manage conflict, how they build and sustain relationships. These are the behaviours that are most visible to others, most amenable to the kind of honest feedback that well-designed coaching generates, and most directly shaped by the leader’s internal experience of their role, which is the primary domain of coaching intervention. The research is considerably less consistent on the question of whether coaching produces changes in strategic thinking quality, in technical expertise, or in the core personality characteristics that shape interpersonal style from the inside.
The three things good coaching reliably changes
The first thing good coaching reliably changes is self-awareness of impact. The most consistent outcome of well-designed executive coaching across research studies and practitioner reports is an increase in the coachee’s accurate understanding of how their behaviour is experienced by others. This sounds modest and is in practice transformative, because the gap between leaders’ self-understanding and their actual impact is, as documented across multiple research programs, both substantial and systematically skewed in the direction of overestimating positive impact. The leader who discovers, through the honest stakeholder feedback that well-designed coaching generates, that their characteristic decisiveness is being experienced as dismissal of others’ input, has gained access to information that is both highly specific and highly actionable. They know what to change and they know why changing it matters to the people around them.
The second thing good coaching reliably changes is the quality and honesty of the leader’s developmental conversations. Coaching creates both the skill and the habit of honest developmental exchange, and the evidence consistently shows that these transfer from the coaching relationship to the leader’s other significant developmental relationships: their team conversations, their mentoring relationships, and the quality of the developmental input they both give and receive in their professional network. This transfer effect is one of the most important and least consistently measured outcomes of executive coaching, because its value compounds over time in ways that are difficult to attribute to a specific engagement but that are visible in the aggregate quality of the developmental culture that a population of well-coached leaders produces.
The third thing good coaching reliably changes is the leader’s relationship to their own development, specifically their willingness to engage with development as an ongoing process rather than as a periodic remediation of identified deficits. The coach-coachee relationship, when it is working well, is itself a model of the kind of honest, curious, ongoing developmental engagement that the coachee can take as a template for their own practice. Leaders who have experienced genuinely good coaching often describe it as having changed not only specific behaviours but their fundamental orientation to professional development: what it is for, what it requires of them, and what quality of engagement it can produce when it is done well.
The two things coaching cannot change
The first thing that coaching cannot change is the organisational environment in which the leader is operating. The most sophisticated coaching engagement produces a more self-aware, more behaviourally skilled, more developmentally oriented leader who is still operating inside the same structural incentives, cultural norms, and systemic pressures that shaped the behaviour the coaching was designed to address. When those systemic conditions are the primary driver of the behaviour in question, coaching that addresses only the individual leader will produce change that is partial and potentially temporary. The leader who has been coached to be more inclusive in their decision-making is still operating in an organisation where the performance management system rewards individual decisiveness. The one who has been coached to create more psychological safety is still operating in a culture where senior leaders model the opposite.
This is not a counsel of despair about the value of individual coaching. Individual change in the presence of unchanged system conditions is still genuine change and still worth pursuing. It is an honest acknowledgment of the limits of what the individual coaching engagement can produce in the absence of systemic change, and a caution about the use of coaching as a substitute for the organisational development work that some situations require.
The second thing coaching cannot change, and the limitation that is most important to acknowledge honestly in contracting, is the fundamental character structure of the individual. Coaching can develop behaviours, expand the range of available responses, improve the quality of relationship management, and increase the self-awareness that enables better choices in specific situations. It cannot transform the basic dispositional characteristics of the person, the introverted person into an extrovert, the person with a fundamental discomfort with conflict into someone who welcomes it, the person whose primary orientation is toward control into someone who is genuinely comfortable with distributed authority. These characteristics can be managed, worked around, and developed at the edges. They are not, within any reasonable coaching timeline, fundamentally altered.
The coaching engagement that is premised on producing this kind of fundamental character change is being sold something that cannot be delivered, and the coachee and sponsor should both be aware of this before the engagement begins. The more productive frame is to ask what specific behavioural changes are achievable within the coachee’s actual dispositional range, rather than what the organisation wishes the coachee to become regardless of who they actually are.
The conditions that determine which people benefit most
The research on who benefits most from coaching is consistent and instructive. The people who benefit most from executive coaching share several characteristics that are worth naming precisely for the guidance they provide to coaching commissioning decisions.
They are genuinely invested in the development goal, not merely compliant with the organisation’s decision to provide coaching. As discussed elsewhere, the motivation to change is the most important single variable in determining whether coaching produces genuine behavioural change, and it cannot be substituted for by the quality of the coaching or the comprehensiveness of the stakeholder feedback.
They have sufficient psychological security to engage with genuinely challenging feedback without either dismissing it or collapsing under its weight. The leader who can receive evidence that their behaviour has been producing effects they did not intend and did not want, hold that evidence with genuine curiosity rather than either defending against it or being destabilised by it, and think creatively about what they might do differently, has the specific relational capacity that good coaching requires. This capacity varies substantially by individual and is not reliably predictable from job level, professional success record, or stated openness to development.
They are operating in an organisational context that supports rather than actively undermines the behavioural changes the coaching is designed to produce. The leader whose coaching goals are directly contradicted by the implicit cultural norms of their organisation is working against a structural headwind that increases the difficulty and reduces the durability of any change achieved.
Measuring what coaching actually produces
The measurement of coaching effectiveness is one of the most neglected and most important aspects of executive coaching commissioning, and the gap between the sophistication of the coaching practice itself and the sophistication of its measurement is striking. Most coaching engagements are assessed through satisfaction surveys of the coachee, which measure the quality of the coaching experience rather than the quality of the behavioural change the experience was supposed to produce. The minority that measure behavioural change typically do so through self-report rather than through the stakeholder evidence that is the only reliable source of data on whether the leader’s behaviour has actually changed in the relationships where change was needed.
The measurement standard that coaching should be held to, and that well-designed stakeholder-centred coaching does hold itself to, is this: have the specific behaviours that were identified as most important to change actually changed, in the assessment of the specific people who are most affected by those behaviours? This standard is demanding, but it is the standard that actually measures whether the organisation’s investment has produced the outcome the investment was designed to produce. Everything else is measuring something easier and less useful.
Good coaching changes specific, named behaviours in specific, observable ways. It cannot change the organisation around the leader, and it cannot transform fundamental character. What it can do, when it is well-designed and honestly evaluated, is produce a more self-aware, more relationally skilled, and more genuinely developmental leader. That is worth having, precisely because it is specific and real.