CrossRef citations to date

Leadership coaching’s efficacy and effect mechanisms – a mixed-methods study

Pages 43-59
Received 22 Oct 2020
Accepted 28 Jan 2021
Published online: 08 Feb 2021


Due to concerns on the efficacy of formal leader development programmes such as low learning transfer back to the organisation, there has been an increasing shift towards individually owned leader development programmes within organisations. Whilst leadership coaching is one of these and is gaining in popularity, further studies are needed to validate its efficacy and explain its effect mechanisms. Drawing on adult learning theory and utilising a mixed-methods study design, we provide further evidence of coaching’s positive effect on organisational leaders and provide insights into how these occur. Thematic analyses of data obtained from nineteen semi-structured interviews indicate that through the coach supporting, listening and challenging the coachee, leadership coaching assists coachees to increase their self-awareness, confidence, clarity and focus, and adopt a wider perspective, and assists to explain coaching’s positive effect observed in our pretest-posttest study of 70 coached leaders. Combined, these results indicate leadership coaching closely resembles the ‘practices and discoveries’ facilitating intentional change and transformative learning which have been associated with desired and sustained change and provides strong support for coaching’s efficacy as a leader owned development intervention.

Practice points

  • To which field of practice area(s) in coaching is your contribution directly relevant? This study is relevant for leadership coaches and human resource development professionals responsible for developing organisational leaders.

  • What do you see as the primary contribution your submission makes to coaching practice? This study provides further evidence of coaching’s efficacy and insights into its effect mechanisms, with results indicating leadership coaching closely resembles ‘practices and discoveries’ facilitating intentional change and transformative learning, which indicates that leadership coaching may assist coachees to experience discontinuities and evolving shifts in perspective leading to a sustained, desired change.

  • What are its tangible implications for practitioners?

    • Understand organisational concerns on the efficacy of leader development programmes

    • Become more conversant with intentional change and transformational learning theories

    • Better able to explain how coaching assists leaders to learn and develop


It is widely recognised that leaders play an important role in their organisational success (Rosenbach et al., 2018), and may reflect why organisations invest most of their training budgets on leadership development (Lacerenza et al., 2017). However, organisations have expressed significant concerns about formal leader development programmes citing high costs and low transfer back to the organisation (Boyce et al., 2010). This may explain the rise in leader-owned development interventions (Clarke & Higgs, 2016) where, informed by adult learning theories such as andragogy (Knowles, 1975), the adult learner is placed at the centre of the experience, motivation to learn is internal rather than external, and experience is utilised as the source for learning.

One leader-owned development intervention gaining in popularity is leadership coaching (Bozer & Jones, 2018), with studies associating the practice with positive outcomes such as personal development, behaviour change and work performance (Athanasopoulou & Dopson, 2018; Jones et al., 2016; Zuñiga-Collazos et al., 2020). However, scholars contend the practice is ahead of its scientific understanding (Bozer & Jones, 2018) with calls for rigorous studies to investigate its theoretical framing (Egan & Hamlin, 2014) and effect on leaders (Grover & Furnham, 2016). We seek to address this gap by providing further evidence of its efficacy and insights into its effect mechanisms through a mixed-methods research design.

Theoretical framework

Leadership coaching

Informed by the coaching literature (e.g., Athanasopoulou & Dopson, 2018), we define leadership coaching as a one-on-one tailored learning and developmental intervention for organisational leaders that uses a collaborative, reflective, goal-focused relationship to develop and maintain positive change in personal development and leadership behaviour leading to the achievement of professional outcomes. In this context, leadership coaching outcomes align with organisational objectives, that is, a triadic relationship exists between the coach, coachee and organisation. Other interventions such as life coaching (no triadic relationship exists with the organisation), mentoring (mentor is more senior to the mentoree and often a subject matter expert), and internal/managerial coaching (coach is an employee of the coachee’s organisation and often the coachee’s supervisor), are excluded from this study. Further, we use the term leadership coaching rather than executive coaching, as the latter is considered more for executive leaders, and our interest is in all levels of leaders.

Scholars suggest one of coaching’s strengths is that the practice is informed by multiple disciplines including psychology, sociology, and adult learning (Maltbia et al., 2014). More recently (e.g., Spence et al., 2019; Taylor et al., 2019), coaching has been associated with two theories within andragogy that emphasise desired and sustained change, that is, intentional change theory (ICT: Boyatzis, 2008) and transformative learning theory (Mezirow, 1997).

ICT suggests leader development occurs within a complex system involving behaviour, thoughts, feelings and perceptions, is a non-linear and discontinuous process, and assists to explain desired and sustainable change (Boyatzis, 2008). As depicted in Figure 1, Boyatzis (2008) argued that when a leader is supported to move through five ‘discoveries’ that is, discoveries resulting from articulating one’s ideal self, comparing this to one’s real self, then developing a plan to bridge the gap and practising the desired changes, this leads to the leader experiencing discontinuity resulting in a ‘tipping or trigger’ point and a sense of urgency to change and develop. Taylor et al. (2019) argued leadership coaches are change agents, and therefore facilitate self-driven and sustained change by applying the principles of ICT leading to increased feelings of autonomy, relatedness, and competence and therefore motivation to change.

Figure 1. Intentional Change Theory (ICT: Boyatzis, 2008, p. 304).

Transformative learning, or as it is sometimes known, transformational learning, was popularised by Jack Mezirow in 1991 and is argued to go beyond content or process learning, to ‘deep learning’ or ‘inner work’ where learners critically reflect on their assumptions and beliefs to transform or develop their thinking (Taylor & Laros, 2014). Mezirow (1997) defined transformative learning as ‘reflectively transforming the beliefs, attitudes, opinions, and emotional reactions that constitute our meaning schemas or transforming our meaning perspectives’ (p. 223), and argued that awareness of one’s frame of reference, critical reflection, and dialogue, are important elements of transformative learning. E. W. Taylor and Laros (2014) detailed its application further and proposed that transformative learning could be facilitated through six ‘practices’, that is emphasising personal experience, critical reflection, dialogue, adopting a holistic orientation, awareness, and authentic support from others. Moons (2016) and Spence et al. (2019) suggest these practices are core elements of leadership coaching, and lead to evolving shifts in perspective by the coachee and therefore sustained change.

Hypothesis development

Our initial efforts focus on coaching’s efficacy, that is, we present reasons supporting associations between participation in leadership coaching and enhanced individual-level leadership outcomes including open-mindedness, leadership self-efficacy, emotional intelligence, behaviour, and effectiveness. Our principal argument centres on the similarities between leadership coaching and the ‘practices and discoveries’ facilitating intentional change and transformative learning.

Coaching and open-mindedness

‘nobody is as smart as everybody’. (Taylor, 2007, p. 32)

As highlighted by Taylor’s quote, organisational success is increasingly being recognised as the collective efforts of all employees rather than the efforts of a single leader, indicating open-mindedness is an important leadership attribute for today’s rapidly changing and complex environments (Tiffan, 2010). We define open-mindedness as a ‘willingness to revise one’s views’ (Hare, 1979, p. 20), and whilst traditional education methods are considered to have little effect on open-mindedness, Hare (2011) argued that open-mindedness could be developed through open discussion and modelling.

We argue that through the coach using ‘practices’ similar to those facilitating transformative learning such as confidential dialogue, critical reflection, and unconditional support (Stout-Rostron, 2014), this assists coachees to challenge their assumptions and beliefs, consider different perspectives and increase self-awareness, leading to increases in open-mindedness. Supporting this perspective, one of the primary objectives of coaching is to create behaviour change through increasing coachee self-awareness (Bozer & Sarros, 2012). This is likely to enhance a person’s ability to recognise and manage bad habits of thought such as biases that may inhibit open-mindedness (Riggs, 2010).

Hypothesis 1. Participation in coaching is associated with increases in open-mindedness.

Coaching and leadership self-efficacy

Leadership self-efficacy reflects individual’s confidence in their abilities, skills and knowledge in leading others (Hannah et al., 2008), with literature indicating it can be developed through four sources identified by Bandura (1977), that is, enactive mastery (practice), vicarious experiences (modelling others), verbal persuasion, and emotional arousal (physiological information). We argue that processes common to coaching (Kampa-Kokesch & Anderson, 2001) are similar to the ‘discoveries’ facilitating intentional change, and as coachees progress through these, they are likely to increase their leadership self-efficacy through each of Bandura’s four sources. To illustrate, following the identification of gaps between a leader’s ideal and real self as reflected for example by the Goal and Reality phases of the GROW coaching framework (Whitmore, 2002), coaches assist coachees to develop a learning agenda to bridge identified gaps and to implement their learnings through practice such as role-playing with the coach and practising in the workplace (Finn, 2007). The coached leader is therefore likely to experience small successes (enactive mastery) as they practise and implement their learnings leading to increased leadership self-efficacy (Bozer & Jones, 2018). Similarly, through the coach encouraging the coachee to discuss interferences perceived as inhibiting their leadership performance, and focus more on strengths and solutions (Grant, 2013), this is likely to lead to lower feelings of anxiety, stress and frustration (physiological information), and therefore enhanced leadership self-efficacy.

Hypothesis 2. Participation in coaching is positively associated with increases in leadership self-efficacy.

Coaching and emotional intelligence

Emotional intelligence (EI) reflects an individual’s ability to recognise and regulate emotions in themselves and others (Cherniss & Goleman, 2001), with evidence associating the construct with teamwork effectiveness, job performance and transformational leadership (Cartwright & Pappas, 2008; Farh et al., 2012). Goleman (2000) argued that unlike the intelligence quotient (IQ) which is largely genetic, EI could be learned at any age and comprises of distinct competencies. For example, Cherniss and Goleman (2001) identified four main EI competences, that is, two relating to personal competence (i.e., self-awareness and self-management) and two relating to social competence (i.e., social awareness and relations management).

We argue that coaching using techniques similar to ‘practices’ facilitating transformational learning such as awareness-raising, critical reflection, dialogue and authentic support, is likely to increase a coachee’s awareness of their and other’s emotions, and following the development of a learning agenda, with practice and support from the coach, is likely to lead to an increased ability to recognise and manage emotions. For example, the coach’s use of reflective questioning (Stout-Rostron, 2014) is likely to raise a leader’s self-awareness (Wasylyshyn, 2003), such as the impacts of their emotions and behaviours on others. This is likely to lead to increases in EI competencies such as emotional self-awareness (Bozer & Sarros, 2012) which is considered the most essential EI competency (Goleman, 1998). Following this increased awareness, through the leader practising learnings in the workplace or role-playing with the coach (Stout-Rostron, 2014), this is likely to lead to enhanced emotional proficiency (Kaplan et al., 2014).

Hypothesis 3. Participation in coaching is associated with increases in emotional intelligence.

Coaching and positive leadership behaviour

Informed by leadership behaviour literature (e.g., DeRue & Myers, 2014; Walumbwa et al., 2008), to assess coaching’s efficacy, we group positive leadership behaviour into four categories, that is, those that are authentic-, task-, relations- and change-oriented.

As one of leadership coaching’s primary purposes is to facilitate sustained shifts in behaviour to enhance performance (Athanasopoulou & Dopson, 2018), we argue leadership coaching can influence these categories of leadership behaviour. To illustrate, informed by literature indicating coaching techniques closely resemble ‘practices’ facilitating transformative learning such as the use of dialogue, critical reflection and awareness-raising (Moons, 2016; Spence et al., 2019), coachees are likely to increase the awareness of the impacts their behaviours have on others, leading to evolving shifts in perspective and sustained behaviour change. For example, by coaching increasing a leader’s self-awareness and gaining a wider perspective (Taylor et al., 2019), they are more likely to adopt a more considered and balanced approach as exemplified in relations- and authentic-oriented leadership behaviours (Avolio & Gardner, 2005; Yukl et al., 2002).

Similarly, informed by ICT and reflected in commonly used coaching models such as Whitmore’s (2002) GROW model, coaching facilitates leaders to discover and clarify differences between their ideal and real self, and develop and implement action plans to address these gaps. Coachees are therefore likely to more ably clarify their objectives and expectations of others and monitor the effectiveness of their plans to achieve their desired goals (Whitmore, 2002) as exemplified in task- and change-oriented leadership behaviours (Yukl et al., 2002).

These perspectives are supported by literature indicating leadership coaching is associated with increased trust in subordinates (Ladegard & Gjerde, 2014), communication (Spence et al., 2019) and envisioning and articulate change (Bozer & Jones, 2018) indicating coaching is associated in increases in relations- and change-oriented leadership behaviours. Similarly, leadership coaching has been associated with constructively viewing difficult issues and achieving goals (Blackman, 2007) indicating an association with enhanced task-oriented leadership behaviour, and increased self-awareness, self-regulation and adopting a wider perspective (Athanasopoulou & Dopson, 2018; Smith, 2015), reflecting authentic leadership behaviour (Avolio & Gardner, 2005).

Hypothesis 4. Participation in coaching is associated with increases in positive leadership behaviour.

Coaching and leadership effectiveness

Literature suggests learning and development interventions centred within ICT and transformative learning theory are highly relevant to the development of leadership effectiveness (Boyatzis, 2008; Mezirow, 1997). This and coaching’s associations with these theories (Lee & Roberts, 2010; Moons, 2016) and the common objective of leadership coaching to improve individual and organisational performance, suggests coaching is associated with increased leadership effectiveness.

This is supported by meta-analyses indicating coaching has a positive effect on affective, cognitive, skill-based, and individual-level leader outcomes (e.g., Grover & Furnham, 2016; Jones et al., 2016), and by association, leadership effectiveness.

Hypothesis 5. Participation in coaching is associated with increases in leadership effectiveness


To investigate coaching’s efficacy and effect mechanisms, an explanatory sequential mixed method approach was adopted (Snelson, 2016). First, quantitative data sourced from 70 leaders and 233 of their subordinates and supervisors pre- and post-coaching were utilised to assess coaching’s efficacy. Second, an inductive thematic analysis (Braun & Clarke, 2006) of narrative data from nineteen semi-structured interviews with coaches and coachees was undertaken to further assess coaching’s efficacy and to provide insights into its effect mechanisms.

Quantitative study

Participants and procedures

For the quantitative study, eight coaches who were known to the principal researcher identified 82 organisational leaders (coachees) predominately in Australia who were soon to commence leadership coaching and met the study’s eligibility criteria. Of these, 70 coachees volunteered to participate in the study and returned full data sets. 49% were male, 51% female, 73% had a Bachelor or higher degree, were on average were 44.2 years old (ranged from 28 to 66 years) and had an average of 13.2 years of leadership experience. 21% classified themselves as front line supervisors, 59% as middle managers and 20% as senior or executive managers. Each participant (N = 70) and their raters, that is, their subordinates (N = 175) and supervisors (N = 58), completed online questionnaires pre- and post-coaching.

Following the completion of pre-coaching questionnaires, coachees received on average six 60–90-minute one-on-one leadership coaching sessions funded by their employing organisation over an average of four months and identified their primary goal from leadership coaching to be developmental (46%), skill development (30%), performance improvement (11%) or other (13%). Coaching were delivered by one of eight professional leadership coaches known to the principal researcher, and had on average 14 years of leadership coaching experience, were external to the coachee’s organisation and volunteered to participate in the study.


The study’s five constructs were operationalised using questionnaires from prior leadership studies due to their established reliability and validity.

Open-mindedness was assessed with five items developed by Tjosvold and Poon (1998) and assessed by the coachees. Each item utilises a seven-point Likert scale (1 = A little like me, 7 = A great deal like me) and includes items such as ‘To what extent do you express your own views openly?’ (Cronbach’s alpha; pre = .766, post = .739).

Leadership self-efficacy was assessed with the Leader Efficacy Questionnaire (LEQ) developed by S. Hannah and Avolio (2013). The LEQ was completed by coachees, and consists of a 22 item 11-point Likert questionnaire (0 = not at all confident, 100 = totally confident) and includes items such as ‘As a leader, I can motivate myself to perform at levels that inspire others to excellence’ (Cronbach’s alpha; pre = .904, post = .927).

Emotional intelligence was assessed with the Genos Emotional Intelligence Inventory - Concise (GEII) developed by Palmer and Stough (2014) and was assessed by the coachees. The GEII consists of a 31 item 5-point Likert questionnaire (1 = almost never, 5 = almost always) and utilises items such as ‘When someone upsets me at work I express how I feel effectively’ (Cronbach’s alpha; pre = .890, post = .918).

Leadership behaviour was operationalised utilising the Managerial Practices Survey (MPS: Yukl, 2012) to measure task-, relations- and change-oriented leadership behaviours, and Avolio et al.’s (2007) Authentic Leadership Questionnaire (ALQ) to measure authentic leadership behaviour. Both instruments were completed by the coachee’s subordinates as they were considered most likely to observe the coachee’s behaviours due to their frequent work interactions with their manager. The MPS utilises a 13-item 5-point Likert scale (1 = not at all, 5 = to a very great extent) with items such as ‘Please describe how much your supervisor explains task assignments and member responsibilities; sets specific goals and deadlines for important tasks; explains priorities for different objectives; explains rules, policies, and standard procedures’ (Cronbach’s alpha; pre = .942, post = .944). The ALQ utilises a 16 item 5-point Likert questionnaire (1 = not at all, 5 = frequently, if not always) with items such as ‘My manager admits mistakes when they are made’ (Cronbach’s alpha; pre = .949, post = .945).

Leadership effectiveness was assessed by the coachee’s subordinates and supervisor with four items developed by Hooijberg and Choi (2001) and one item by G. Yukl (personal communication, February 10 2015) measuring perceived leadership effectiveness. Three items utilise a 6-point Likert scale (1 = very ineffective, 6 = very effective), one utilises a 5-point Likert scale (1 = well below average, 5 = well above average) and one a 10-point Likert scale (1 = the least effective leader I have known, 10 = the most effective leader I have known), with items such as ‘Please indicate the overall effectiveness of the leader in carrying out his/her job responsibilities’ (Cronbach’s alpha; pre = .922, post = .920).

Qualitative study

Participants and procedures

Nineteen participants (twelve female and seven male) comprising of twelve coaches and nine coachees were interviewed for the qualitative study. Of the twelve coaches, eleven resided in Australia, one in Britain, were full-time professional leadership coaches affiliated with international coaching and or HRD institutions, were known to the researcher, and agreed to participate in the qualitative study. The nine coachees were nominated by the participating coaches, varied from supervisory to CEO in their seniority, resided in Australia, and had recently participated in leadership coaching.

Each participant participated in a voluntary one-on-one semi-structured interview between 40 and 60 min in duration. The same seven questions were asked of all participants to ensure themes that arose were not related to the questioning, and included questions such as ‘What do you think changes in a leader as a result of coaching?’ and ‘How do you think coaching brings about these changes?’ Whilst the principal researcher had access to additional coaches and coachees to interview, following interviews with twelve coaches and seven coachees, little new information or insights were provided in the last of these interviews, i.e., theoretical saturation (Sekaran & Bougie, 2013), therefore this sample was deemed adequate for analysis.


Quantitative study

Following assessments of scale reliability and data distributions using IBM SPSS version 25, paired t-tests were used to determine whether mean differences between pre- and post-coaching construct scores were statistically significant (alpha < 0.05). presents means and standard deviations of constructs pre- and post-coaching, results of paired t-tests and Cohen’s repeated measures effect sizes.

Table 1. Means, standard deviations and effect sizes.

Results indicate means for all constructs increased following participation in coaching and were statistically significant, providing support for our five hypotheses, that is, participation in coaching is associated with increases in open-mindedness (H1), leadership self-efficacy (H2), emotional intelligence (H3), positive leadership behaviour (H4) and leadership effectiveness (H5). Of the self-reported constructs, the largest effect size was observed for leadership self-efficacy (d = 1.01) followed by emotional intelligence (d = 0.82). Of the constructs assessed by the coachees’ subordinates and supervisors, the largest effect size was observed for task-oriented leadership behaviour (d = 0.32).

Qualitative study

Following the six-step approach for qualitative research developed by Braun and Clarke (2006), 29 first-order themes were identified from the question ‘What changes during coaching?’. These were reduced to five second-order themes to assist interpretation, that is, data indicated coaching is associated with increasing a leader’s self-awareness, confidence, clarity and focus, and assists the leader to feel more energised and adopt a wider perspective. For example, the exemplary quote ‘The main thing that I see time and time again is the capacity to see and train themselves to see another perspective’ indicates coaching is associated with the leader’s capacity to see a wider perspective. Second-order themes and exemplary quotes are presented in .

Table 2. Themes, frequency, and exemplary quotes.

Similarly, four second-order themes were identified from the question ‘How do you think coaching brings about these changes?’, indicating the coach brings about change by providing a supportive and confidential relationship with the leader, assisting the leader to own their development, by listening, reflecting and challenging the coachee, and assisting them to gain increased clarity and focus. For example, exemplifying the second-order theme ‘supportive and confidential relationship’, one coach likened coaching to a rare relationship saying, ‘ … such a delight to have someone who is non-judgemental and confidential, and a sounding board for their ideas, and reflecting back on the blind spots. It is such a rare relationship for them.’

Two second-order themes were identified in response to the question ‘Are these changes likely to be sustained?’ and indicate coaching’s effects are likely to be sustained, as exemplified in the quote ‘Yeah, I think what I’ve learnt from [coach] will stick because they’re kind of personal growth. It’s quite deep change’, however nearly half the respondents suggested follow-up sessions would be beneficial.

In response to the question ‘What influences coaching’s effect?’, three second-order themes were identified and indicate coaching’s effect is influenced by the coach-leader relationship, the leader’s developmental environment and the leader’s capacity for development. For example, exemplifying the influence of the leader’s developmental readiness theme, one coach said, ‘The person’s openness and capacity for self-awareness and preparedness to be vulnerable I would say are key factors.’


Quantitative study

Results of our quantitative study provide further evidence of coaching’s positive effect and respond to calls to extend the coaching literature into the domains of emotional intelligence (Erdös et al., 2020). For example, our results indicate coaching is associated with enhanced emotional intelligence, particularly the EI competence dimensions emotional expression (d = 0.72) and emotional management of others (d = 0.67). These results suggest coaching may be capable of influencing emotional intelligence building on the limited number of coaching studies researching coaching’s effect on this construct (e.g., Chapman, 2005), and maybe an important effect acknowledging EI’s potential correlation with job performance (O’Boyle et al., 2011).

Similarly, our study adds further evidence supporting definitions of coaching that suggest coaching assists to bring about behavioural change leading to enhanced performance (Athanasopoulou & Dopson, 2018). Of the four behaviour categories assessed in this study, the largest effect was observed for task-oriented leadership behaviour (d = 0.32) indicating coaching may have a larger effect on behaviours relating to clarifying, planning, monitoring and problem solving than those such as making decisions in line with core values as exemplified in authentic behaviour.

Qualitative study

Building on our quantitative study assessing coaching’s efficacy and to gain insights into coaching’s effect mechanisms, we conducted a thematic analysis from responses received from semi-structured interviews with twelve coaches and seven coachees. Results indicate that through the leadership coach providing a supporting confidential relationship with coachees, and utilising techniques such as active listening and critical reflection, coachees increase their awareness, confidence, clarity, and focus. These themes provide further evidence of coaching’s efficacy and provide insights into coaching’s effect mechanisms.

For example, coaching’s positive associations with increases in open-mindedness and emotional intelligence observed in the quantitative study are supported by the study’s thematic themes ‘self-awareness’ and ‘wider perspective’ as exemplified in the quotes ‘Coaching keeps awakening them until they stay awake’ and ‘Perspective … freedom from the attachment to their identity and their self-structures and getting caught up in their own need and getting themselves out of the way’. Similarly, coaching’s association with leadership self-efficacy is evident in the quote ‘Leadership confidence hands down … self-belief in their ability is the biggest increase’.

Coaching’s association with enhanced leadership behaviour is also evident. For example, the thematic themes ‘confidence’ and ‘clarity and focus’ as exemplified in the quote ‘First and foremost, I feel in control of myself … I think this whole process really kind of taught me how to be much more effective rather than frantic, much more in control and being able to really understand the bigger picture’, lend support to coaching’s association with increased task-oriented leadership behaviour such as providing clearer directions, improved planning and more closely monitoring performance.

Our thematic analysis also assists to explain coaching’s effect mechanisms, with themes closely resembling the ‘discoveries and practices’ facilitating intentional change and transformative learning (Boyatzis, 2008; Mezirow, 1997), lending support to contemporary scholars (e.g., Spence et al., 2019; Taylor et al., 2019) who associate coaching with these adult learning theories to explain how coaching leads to desired and sustained change. For example, six of the seven ‘practices’ facilitating transformative learning (Taylor & Laros, 2014), that is, support, dialogue, wider perspective, awareness and critical reflection are mirrored in the second-order themes of our thematic analysis. To illustrate, the second-order theme ‘self-awareness’ is exemplified in the quote ‘It is just that general heightened level of self-awareness about how they navigate their world and how they interact with others’. Themes also indicate coaching’s effect mechanisms are similar to the five ‘discoveries’ facilitating intentional change (Boyatzis, 2008), that is, discovering the ideal self, the real self, then planning and practising to address identified gaps whilst being supported by others. For example, the process of discovering the ideal and real self with the support of others is evidenced in the quotes ‘We simply show them the gap between where they are and where they need to be in order to be effective’ and ‘[Coach] was so beneficial to me because he was supportive and also really challenging, and he really allowed me to take a step back and re-identify with the person that I wanted to be’.

Our thematic analysis also indicates that coaching leads to sustained change as exemplified in the quotes ‘ … it [coaching] is actually changing the brain, rewiring the brain, and the changes become hard-wired’ and ‘Yeah, I think what I’ve learnt from [coach] will stick … it’s kinda deeper than that, about self-awareness, so I think it will be quite hard to undo’. This may be an important aspect of coaching’s efficacy acknowledging concerns on the sustainability of learnings and development from traditional leadership development training (Boyatzis, 2008).

In sum, this study provides further evidence of coaching’s efficacy, however, its greatest contribution is likely to be the insights provided into coaching’s effect mechanisms. In particular, the similarity between the study’s second-order themes and the ‘discoveries and practices’ facilitating intentional change and transformative learning lends support to contemporary scholars (e.g., Spence et al., 2019; Taylor et al., 2019) who associate coaching with these adult learning theories to explain how coaching leads to desired and sustained change.

Practical implications

Organisations have expressed concerns on the high cost of formal leader development programmes and their low transfer back to the organisation (Boyce et al., 2010), yet recognising the importance of leadership to their organisations, continue to invest the majority of their training and development budgets on developing their leaders (Lacerenza et al., 2017).

Whilst this study provides further evidence of coaching’s efficacy, such as coaching’s association with increases in leadership self-efficacy, emotional intelligence and task-oriented leadership behaviour, arguably our largest contribution for practitioners is our insights into coaching’s effect mechanisms, that is, how coaching works.

Results of our thematic analysis indicate the techniques and processes leadership coaches utilise to develop leaders, closely resemble the ‘discoveries and practises’ facilitating intentional change and transformative learning. These are established theories within adult learning and development and indicate that leadership coaching may assist coachees to experience discontinuities and evolving shifts in perspective leading to sustained and desired change. This may be relevant to practitioners from two perspectives. First, there is a wide concern on the transfer and sustainability of learnings from organisational training programmes (Khan et al., 2015), with some suggesting learning transfer reduces to 15% within a year following leadership training (Wexley & Latham, 2002). Second, this may assist practitioners to explain what is sometimes referred to as the ‘black box’ of coaching (Feldman & Lankau, 2005), that is, explain how coaching works to assist furthering its application within organisations.

Limitations and future research

There are several limitations of this study that need to be acknowledged and potentially addressed in future research. First, a control group was not utilised therefore alternate explanations beyond participation in coaching may have contributed to changes observed such as maturation and test effects (Marsden & Torgerson, 2012). Future studies would benefit by adopting randomised control group experimental designs to assess causality (Marsden & Torgerson, 2012).

Second, limited controls were placed on the coaching provided and the eligibility of participants for this study, therefore differences between the coachees, their organisational environment and the objectives of the coaching provided, have the potential to influence coaching outcomes (de Haan et al., 2013). Informed by the results of this study, future studies would benefit by narrowing their focus when investigating coaching’s efficacy and effect mechanisms, and by incorporating additional controls and eligibility requirements.

Third, we support calls for future research to investigate leadership coaching’s efficacy relative to other development interventions (e.g., Smither, 2011). This will assist to increase our understanding of the comparative efficacy of leader development interventions and provide valuable insights for leader development scholars and practitioners alike.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contibutors

Peter Halliwell is a PhD Candidate at the University of Newcastle, Australia, and a senior leader at an Australian energy utility. His research interest lies in leadership and leader development.

Rebecca Mitchell is a Professor and Assistant Dean at Macquarie University, Sydney, Australia. Her research interest lies in organisational behaviour, with a particular focus on health and wellbeing at work and has received several international awards for her research.

Brendan Boyle is an Associate Professor at the University of Newcastle, Australia. His research interests include international HRM and knowledge management/sharing in organisations and has received several awards for his research and teaching.

Additional information


This research was supported by an Australian Government Research Training Program (RTP) Scholarship.


Alternative formats