The global coaching industry generates more than $4 billion annually. Executive coaching has moved from a discretionary perk for a handful of C-suite leaders to a mainstream feature of corporate leadership programmes, board succession planning, and executive education. The investment is substantial and growing.
The question organisations should be asking — but often don't — is not whether coaching produces thoughtful conversations. It is whether it produces observable, lasting changes in leadership behaviour that others in the organisation can see and validate.
The honest answer is: sometimes. Under specific conditions. And far less often than the coaching industry would like you to believe.
The awareness trap
Leadership IQ's research, based on a study involving more than 1,200 employees, found that 84% of leaders showed no improvement in behaviour after their blind spots were pointed out. This is a striking number, and it challenges a foundational assumption of most leadership development: that awareness leads to change.
It does not. Not reliably. Not without something else.
The gap between insight and action is well-documented in psychology. Knowing that you interrupt people in meetings does not stop you from interrupting people in meetings. Knowing that your communication style is perceived as dismissive does not make you a better communicator. Awareness is a precondition for change, not a cause of it.
This is where the design of the coaching engagement matters more than the quality of the coach. A coach who creates brilliant self-awareness in a one-hour session every fortnight, but provides no structure for practising new behaviour between sessions, is producing insight without transfer. The leader leaves each session feeling illuminated and returns to their operating environment, where the same pressures, habits, and interpersonal dynamics reassert themselves within hours.
What the evidence actually says
A 2023 meta-analysis of randomised controlled trials published in Frontiers in Psychology examined the effectiveness of executive coaching across multiple outcome types. The findings were instructive: coaching had a stronger effect on behavioural outcomes than on attitudes or personal characteristics. In other words, coaching is most effective when it targets what people do, not just how they feel or what they believe about themselves.
The same analysis found large effect sizes for performance behaviours and goal attainment, and a moderate effect size for leadership behaviours specifically. The practical takeaway: coaching works best when it is focused on specific, observable behavioural goals — and less well when it is oriented toward general personal development or self-actualisation.
Separate research on coaching during organisational change found measurable improvements in goal attainment, solution-focused thinking, leadership self-efficacy, and resilience. Notably, coaching during periods of transition — when leaders are most destabilised and most receptive to change — produced the strongest effects.
The five conditions for behavioural change
Across the research, a consistent pattern emerges. Lasting behavioural change in senior leaders requires five conditions, all present simultaneously:
Accurate, specific data about the behaviour that needs to change. Not vague feedback ("you could be a better communicator") but precise, multi-source evidence of specific behaviours and their impact. This is where 360° feedback and psychometric profiling earn their value — they provide the diagnostic precision that gives coaching engagements a target.
Emotional engagement with the data. The leader must feel something — surprise, discomfort, recognition — in response to the feedback. Purely intellectual processing is insufficient. The debrief is where this happens: a skilled coach creates the conditions for the leader to sit with the feedback long enough for it to register emotionally, not just cognitively.
A specific, bounded development goal. Not "become a better leader" but "stop rescuing my direct reports from their own decisions within the next 90 days." The goal must be behavioural (observable by others), specific (clear enough to know when you're doing it), and bounded (a defined timeframe that creates urgency without overwhelm).
Deliberate practice in the real environment. This is the element most coaching engagements lack. The leader needs structured opportunities to practise the new behaviour in their actual working context — in real meetings, with real stakeholders, under real pressure — and to reflect on those experiences with their coach. Four to six structured coaching sessions are enough to trigger observable behavioural change, according to recent research. But only if those sessions are connected to real-world practice, not isolated in a coaching room.
Stakeholder follow-up. The leader should share their development focus with a small number of trusted colleagues and ask for ongoing, informal feedback on their progress. This creates external accountability, provides real-time data on whether the change is visible to others, and — critically — signals to the organisation that the leader is taking their development seriously. Research consistently shows that the frequency of stakeholder follow-up is the single strongest predictor of perceived leadership improvement.
What this means for organisations
The most damaging leadership blind spot identified in Leadership IQ's research was not a lack of charisma or executive presence. It was failure to follow through on commitments and operational priorities. When leaders repeatedly fail to follow through, teams hesitate to act, priorities shift unpredictably, and execution slows as employees wait for clarity.
This finding reframes the entire purpose of executive coaching. The goal is not to produce more self-aware leaders. Self-awareness without changed behaviour is an expensive form of introspection. The goal is to produce leaders whose behaviour is observably, measurably different in ways that improve how the organisation functions.
For organisations investing in leadership development, three questions matter:
Is the coaching grounded in multi-source evidence about specific behaviours, or is it based on the leader's self-reported development goals? The former changes behaviour. The latter often just reinforces existing blind spots.
Is there a structured mechanism for translating coaching insights into real-world practice and stakeholder accountability? If the coaching lives entirely in a one-to-one conversation with no connection to the leader's daily operating context, it will produce insight without transfer.
Can you demonstrate, with evidence that others can validate, that leadership behaviour has changed? If the only measure of coaching success is the leader's own satisfaction with the process, you are measuring the wrong thing.
Behavioural change in senior leaders is harder than the industry acknowledges and more achievable than sceptics assume. The difference is not in the quality of the conversation. It is in the rigour of the process around it.