Abstract
Until recently, there has been little published systematic empirical research into business coaching. This article reports on a systematic, critical review of 111 published empirical papers investigating business coaching theory, processes, and outcomes. The present article identifies a significantly larger body of empirical research than covered in previous reviews and uses a Systematic Review methodology (SRm) to conduct a comprehensive review of the available empirical evidence into business coaching effectiveness focusing on implications for theoretical development, practice (within human resource development) and further research in this area. This review identifies convergence around factors that contribute to perceived effective coaching practice but nevertheless highlights a number of issues to be resolved in further research. These include determining the primary beneficiaries of coaching, the factors that contribute to coach credibility, and how the organizational and social context impacts on coaching. Weaknesses in coaching research methodology and research gaps are also noted.
Introduction
Many business organizations make substantial investments in human resource development (HRD) programs hoping that these will increase productivity and be a source of competitive advantage for their organization (Fulmer, Gibbs, & Goldsmith, 2000). Business coaching is one type of HRD activity that has become very popular in recent years. In 2003, Capuzzi Simon claimed that there were tens of thousands of business coaches in the United States, and 3 years later, Fillery-Travis and Lane (2006) described business coaching as a US$2 billion per annum market. Liljenstrand and Nebeker (2008) described business coaching as “the fastest growing field within consulting” (p. 58).
In the present article, business coaching is defined as a formal relationship with a designated coach,
in which the coachee and coach collaborate to assess and understand the coachee and his or her leadership developmental tasks, to challenge current constraints while exploring new possibilities, and to ensure accountability and support for reaching goals and sustaining development. (Ting & Hart, 2004, p. 116)
Although much of the organizational coaching literature uses the term “executive coaching,” we use “business coaching” as a more inclusive description, incorporating all forms of coaching (including executive, management, team, and leadership coaching) taking place in business. Business coaching can be conducted by either external coaches or internal line managers or human resource management (HRM) and HRD staff (Brandl, Madsen, & Madsen, 2009; Teague & Roche, 2011). Although internal coaching is recognized within HRM/HRD as a critical skill for all managers to have as part of their repertoire of human resources (HR) tools (Swart & Kinnie, 2010), it is usually informal and often difficult to distinguish from mentoring (Doorewaard & Meihuizen, 2000). Discussions of business coaching thus usually distinguish it from mentoring by focusing on external coaches and by highlighting the formal, structured nature of the relationship between the coach and coachee (Witherspoon & White, 1996). Even here though there are overlaps, Garvey, Stokes, and Megginson (2009) pointed out that many mentoring programs, at least in the Unites States, are formally structured. Where coaching and mentoring differ, is that coaching is more short-term and focused on specific developmental issues, while mentoring takes a broader view of the person as a whole (Hawkins & Smith, 2013). Others stress that mentoring is often a voluntary activity within an organization, providing psychosocial support, while coaching is more often a paid, professional activity, with the coach acting as a facilitator of learning and development within an organizational context (Gray, Garvey, & Lane, 2016). Although acknowledging that coaches often have to work with psychological issues (Bachirova, Cox, & Clutterbuck, 2014), we see business coaching as working, largely, with functioning people, while therapy focuses on those with some form of dysfunction or disorder (Grant, 2001).
There are several different approaches to coaching, and any one coaching program can include a range of techniques and activities. Coaching programs can also vary in terms of the number, variety, and types of techniques that are used within the program, the length of the program, and the type of consultant employed as the coach (Blackman, 2006). The only restriction used in the present article is that the coaching is aimed at business or professional development (within a business) and that it is conducted with a coach from outside the immediate work unit and/or as part of a structured program.
Although there have been many articles written about what coaching is, advocating for its adoption, and describing how to coach, little critical, systematic, empirical research has been published about the effectiveness of business coaching for either the individual or the organizations they work for (Joo, 2005; Leedham, 2005). Few coaching programs have been formally evaluated within organizations (McDermott, Levenson, & Newton, 2007) and, until recently, little academic research has examined business coaching (Kilburg, 2001; Passmore & Gibbes, 2007). Business coaching has, however, become a topic of academic interest, and there has been a growing number of empirical studies published since 2005 (Ely et al., 2010; Jarvis, Lane, & Fillery-Travis, 2006). This emerging empirical literature offers an opportunity to begin to assess the many claims made for and about business coaching. This study reports on a systematic review of the published empirical research analyzing the processes and outcomes of business coaching. In doing so, it makes an important contribution to the current limited literature on the effectiveness of coaching in general and business coaching in particular. Furthermore, it offers a theoretical framework to guide both future research and practice by identifying the preconditions for coaching, the theoretical approaches taken to business coaching research, the processes and context of coaching, how coaching outcomes are measured, and the credibility of business coaching outcomes.
The Practical and Theoretical Development of Business Coaching
According to Natale and Diamante (2005), there have always been coaches operating in business but the role and its associated activities were not formally recognized. This formal recognition of business coaching as a distinct staff development field of its own came about in the 1980s (Hudson, 1999; Kampa-Kokesch & Anderson, 2001; Kilburg, 1996a, 1996b; Olesen, 1996; Tobias, 1996). Harris (1999) described three periods in the history of business coaching. First, the period between the years 1950 and 1979, when a “few professionals used a blend of organizational development and psychological techniques in working with executives” (Kampa-Kokesch & Anderson, 2001, p. 207). Second, a middle period between 1980 and 1994 characterized by increasing and standardizing of professional services (Kampa-Kokesch & Anderson, 2001). The third period runs from 1995 to the present. In this period, there has been an increase in publications and the establishment of professional organizations for business coaching (Kampa-Kokesch & Anderson, 2001). As the practice of business coaching is considerably more advanced than the depth of research, there have been calls to direct more effort toward critically analyzing both the concepts and practices of business coaching (Diedrich & Kilburg, 2001) and the development of more theory-based research (Stone, 2007). Although there is no universal agreement regarding the definitions of HRD and HRM, it is generally accepted that HRM more often emphasizes policy, staffing, HR-related IT, compensation, and other policy-related issues. HRD is most often described to be development, learning, and performance focused at individual, group, organization, and larger system levels. Therefore, coaching can be classed as a key HRD-related activity (Beattie et al., 2014). Bachirova, Cox, and Clutterbuck (2014) described coaching as “a human development process that involves structured, focused interaction and the use of appropriate strategies, tools and techniques to promote desirable and sustainable change for the benefit of the coachee and potentially for other stakeholders” (p. 1).
Accordingly, there is an emerging field of empirical research that is addressing these problems with a growing number of published papers reporting research data (Jarvis, 2004; Jarvis et al., 2006), and this has generated several academic reviews. Webster and Watson (2002) highlighted the importance of critical literature reviews for theoretical development in the various fields in management. They note that literature reviews provide two key functions, the directing of new research through the identification of topics that have been examined in detail and of gaps in knowledge; and the description of cumulative patterns of results that can support theoretical development. The early reviews into business coaching, those conducted up to and including 2005, consistently reported a limited number of empirical papers. Kampa-Kokesch and Anderson (2001) found only seven empirical research–based papers, Feldman and Lankau (2005) were able to analyze 20 empirical studies, and Joo (2005) identified 11 research articles out of 78 academic papers on business coaching. Lowman (2005) and Bennet (2006) also provided early reviews but in each case they focused only on papers in special issues of journals or as presented at meetings dedicated to business coaching.
All these early reviews focused on what Passmore and Fillery-Travis (2011) have described as the first stage of the development of a new academic area—establishing definitions and exploring the scope of the phenomenon. Their second stage involves more consideration of how to evaluate or measure the concepts under study, development of research methods and initial development of theory. Four more recent reviews are clearly within this second stage, although each with a particular focus. Passmore and Gibbes (2007), for example, reviewed 23 empirical papers but concentrated their review on assessing the potential contribution of counseling research and theory to business coaching. Ely and colleagues (2010) reported on a larger number of studies (49) but selected only those studies that addressed or measured leadership development as an outcome from the coaching programs. In a similar fashion, De Meuse, Dai, and Lee (2009) reviewed empirical studies into coaching, seeking to assess return on investment (ROI) and so only examined six quantitative studies. In a similar fashion, three meta-analytic reviews have been conducted in recent years. Theeboom, Beersma, and van Vianen (2014) identified 18 quantitative empirical studies presented in sufficient detail to conduct a meta-analysis, while Sonesh and colleagues (2015) identified 24 studies suitable for meta-analysis, and R. Jones, Woods, and Guillaume (2015) analyzed 17 studies. In all cases, the results of the meta-analyses provide some limited evidence that coaching was associated with positive effects on outcomes. These limitations included a heavy reliance on self-reported outcomes and a lack of control groups suggesting that it was not possible to clearly link coaching to outcomes (Theeboom et al., 2014), stronger effects for coachee perceptions of the quality of the coach–coachee relationship than for other sorts of goals or outcomes and stronger effects for student samples than for executive and other nonacademic samples (Sonesh et al., 2015), and lower effect sizes for multi-source feedback measures (R. Jones et al., 2015). Taken as a whole, these reviews suggest that under certain very specific conditions, some positive relationships can be detected between coaching and individual benefits. The use of meta-analysis does, however, limit the number of studies that can be considered and so do not provide a very comprehensive overview of the available evidence.
In recent years, several wider reviews of business coaching have been published. Passmore and Fillery-Travis (2011) provided a comprehensive review of business coaching research up to the late 2010s, but details on how they selected papers for inclusion in their review are limited, and the review incorporated both research papers and other discussions of business coaching, with a continuing emphasis on connections to the counseling literature. Stern and Stout-Rostron (2013) provided a more traditional narrative literature review based on 263 papers about coaching from both a wide range of disciplines. The aim of this review was to describe coverage of 16 major topics in business coaching, and the review did not address issues of research quality or the nature of evidence about the topics examined. Bartlett, Boylan, and Hale (2014) examined definitions and descriptions of coaching by analyzing the key words of 533 papers. Again the review did not distinguish between descriptive, advocacy, and empirical research-based papers and did not address the nature of evidence about coaching.
A 2014 special issue of the journal Advances in Developing Human Resources offered a number of papers that reviewed and explored aspects of business coaching. Three were focused on specific topics including coaching measurement scales action learning (O’Neil & Marsick, 2014), and how coaching is integrated into HRD processes (Beattie et al., 2014). The first described the major theoretical traditions in coaching mapping these to current coaching practices (Cox, Bachkirova, & Clutterbuck, 2014). The second presented a framework on coaching definitions and competencies to inform the development of accreditation processes and coach credentials (Maltbia, Marsick, & Ghosh, 2014). Although each of these papers offers a synthesis of ideas based on reviews of previous research, and included both empirical studies and other documents, neither aimed to provide a comprehensive overview of the field. The Cox et al. (2014) paper, for example, provided an overview of the main theoretical approaches that can be linked to coaching practice but not a systematic assessment of how these theoretical approaches have been linked to evidence of the effectiveness of these coaching practices.
Briner and Denyer (2012) distinguished between several different types of academic literature reviews. The first type is a traditional narrative review which seeks to provide a comprehensive overview, often presented in chronological order of what has been studied in a field and what major conclusions can be drawn from this research. The reviews by Kampa-Kokesch and Anderson (2001), Feldman and Lankau (2005), Joo (2005), and Passmore and Fillery-Travis (2011) are examples of this type of review. A second type of review is the meta-analytic review (e.g., De Meuse et al., 2009; R. Jones et al., 2015; Sonesh et al., 2015; Theeboom et al., 2013) which is necessarily constrained to particular types of studies and the nature of the data reported by those studies. A third type of review is one based on themes which presents the results of a set of papers relevant to a specific theme or question (Briner & Denyer, 2012). In the previous section, several such reviews were noted. Finally, there are what Briner and Denyer (2012) refer to as selective, expert reviews which provide an argument about issues or concepts supported by a selection of references (e.g., Cox et al., 2014; Maltbia et al., 2014). Although this latter type of review can offer synthesis and direction to a field, they are not focused on systematically and critically evaluating the available evidence in a field. and it has been suggested that these expert reviews could be supported and extended by the final category of systematic reviews (Briner & Denyer, 2012)—as in our study.
Briner and Denyer (2012) defined a systematic review as one that
addresses a specific question, utilizes explicit and transparent methods to perform a thorough literature search and critical appraisal of individual studies, and draws conclusions about what we currently know and do not know about a given question or topic. (p. 329)
This type of systematic review is common in fields such as medicine, law, and psychology but is only recently being adopted in business and management (Rojon, McDowall, & Saunders, 2011) despite calls for, and explanations of, SRm in management beginning more than a decade ago (Tranfield, Denyer, & Smart, 2003). SRm can be contrasted with other more traditional approaches to literature reviews including meta-analytic approaches that reanalyze data from quantitative studies, argument/thematic reviews which argue for a particular standpoint supported by selective references to research, and expert/narrative reviews that focus on an individual synthesis of an area (Briner & Denyer, 2012). Although each of these approaches has its benefits, they also have limitations. Meta-analytic reviews, for example, are limited to very specific types of quantitative studies reported in sufficient detail to facilitate analyze and so often cannot consider the full range of evidence available on a topic (Rousseau, Manning, & Denyer, 2008). Thematic and expert reviews are often limited by a failure to explicitly describe how papers were selected for review and how they were analyzed, making them open to potential bias on the part of the review authors (Rojon et al., 2011). According to M. Jones, Coviello, and Tang (2011), this approach is especially useful in newer areas where practice and advocacy may not be well connected to emerging research, as is the case for the effectiveness of business coaching.
The present article sought a unique approach to critically examining the issue of coaching effectiveness using a SRm which has not been used previously in the business coaching literature. A systematic review is critical because it evaluates the quality of the evidence provided in the empirical literature rather than just describing the conclusion of the papers reviewed. In the area of business coaching, such an approach supports a more comprehensive but focused review incorporating a larger body of empirical evidence than has previously been examined in narrative or meta-analytic reviews. As a systematic review it includes all the available empirical papers and an assessment of the quality of the evidence in these papers which offers a way to test the claims made in the thematic and expert reviews. In this latter context, it can be seen as a form of triangulation. Given the dramatic increase in academic empirical studies evident in recent years, the present review provides an assessment of the effectiveness of coaching based on a larger body of empirical evidence than was previously available, a comprehensive overview of the field that is not limited by a specific focus on one aspect of coaching, and offers direction for a more evidence-based conceptual framework to guide future research and practice. A core contribution of the present review is that it is based on a specific type of review methodology that has not been previously used in the field.
SRm: An Overview
The use of SRm is aimed at avoiding these biases by having reviewers follow and clearly report a systematic process for the search, selection, inclusion, and analysis of a comprehensive set of studies in an area (Briner & Denyer, 2012). The four core principles of SRm are that the process is systematic and organized to avoid researcher bias, there is a transparent and explicit audit trail for decisions, the review is replicable and easily updateable, and it synthesizes the evidence to answer specific preset questions (Briner & Denyer, 2012; Denyer & Tranfield, 2009; Rousseau et al., 2008; Tranfield et al., 2003). Although there are no standardized procedures set down for SRm examples, all share the following four major steps: identifying a clear review question focused on assessing the available evidence on a key topic; there is a clear protocol used to locate and select relevant studies and the steps in this protocol are designed to minimize reviewer bias; there is a critical appraisal of the studies included; and this analysis is synthesized to reveal new knowledge in an area (Briner & Denyer, 2012; Denyer & Tranfield, 2009; Rousseau et al., 2008; Tranfield et al., 2003).
Thorpe, Holt, Macpherson, and Pittaway (2005) summarized the steps suggested for a systematic review by Rousseau et al. (2008) and Tranfield et al. (2003) into three main steps with several actions within each step and the present article followed this procedure. The first main step is to define the objectives of the review and establish the guiding questions. Building on the previous reviews of coaching and addressing the gaps in their approaches, this systematic review of empirical research into the effectiveness of business coaching was guided by the following research questions:
The second main step is to conduct the research review, and this consists of searching for relevant research, using criteria to decide on which studies to include in the review, coding the included studies, and analyzing these studies (Thorpe et al., 2005). The search procedure for the present review built upon the work of Grant (2005b) who conducted a review of peer-reviewed behavioral science journals and identified 175 articles that included some discussion of business coaching; however, only 24 of these were reports of empirical research. For the present article, these 24 articles were taken and supplemented a search for articles published in peer-reviewed journal since March 2005. The decision to focus on peer-reviewed journals was consistent with reviews by M. Jones et al. (2011), and Keupp, Palmié, and Gassmann (2012). Searches of the electronic academic databases Google Scholar, ProQuest (which included InformIT, PsycINFO, and PsycARTICLES), Expanded Academic ASAP (which included InfoTrac), Emerald Insight, Ingenta Connect, SAGE Premier Journals, ScienceDirect, SpringerLink, Taylor & Francis Online, and Wiley Online were conducted using the terms “business coaching” and “executive coaching.” References within key articles and the existing review papers and Grant’s (2011) annotated bibliography were also used to identify any additional papers. Following the guidelines of Parris and Peachey (2013), results of the searches were judged against three inclusion criteria: the papers were in English, they addressed the effectiveness of business or executive coaching, and they reported on empirical research. For this article, empirical research is defined as “the process of developing systematized knowledge gained from observations that are formulated to support insights and generalizations about the phenomena under study” (Lauer & Asher, 1988, p. 7). Based on this definition and, consistent with both the suggestions of Briner and Denyer (2012) and a critical realist research approach, the review included both quantitative and qualitative research. One researcher applied the selection criteria and ambiguous papers were checked by a second researcher before a joint inclusion decision was made. The application of this protocol resulted in a final set of 111 empirical papers for review. Tables listing the reference details for these studies and summarizing their key characteristics are available on request from the corresponding author. Of these 111 papers, 15 were published prior to 2000, with the remaining 96 published since 2004, supporting claims made about an upward trajectory in this area by Ely et al. (2010). Thirty-five articles were published in 2012 to 2015 further demonstrating that the field of coaching research is developing and that considerably more evidence is now available for consideration than previous narrative reviews such as those provided by Passmore and Fillery-Travis (2011).
Findings
In this section, we first report on the methodological and theoretical approaches taken by the business coaching articles reviewed. The findings are then organized by the first four research questions; the fifth research question focused on gaps is answered by considering the findings overall and is discussed in the “Discussion” section.
Methodological Approach
These 111 research papers were then organized into three groups according to the research designs used to evaluate or study business coaching. This approach follows the suggestions made by Passmore and Fillery-Travis (2011) that there is a link between different methodological styles and the types of questions addressed. Within each category, each study was coded according to the nature of the sample, the method used, and major findings. It is important to note that there is, however, considerable variability in research quality in each grouping.
The first group consisted of case studies and one group posttest only designs (54 papers were reviewed). These types of approaches have been classified as preexperimental, and while they can provide valuable insights into the process of coaching and indications of patterns and trends (Kilburg, 2004), they do not provide independent evidence that a change has occurred and they can be confounded by numerous sources of bias (Shadish, Cook, & Campbell, 2002). The studies reported in the first group were either case studies describing a one-on-one set of coaching sessions or an analysis of the target coaching exercise using interviews and questionnaires. Although a variety of scales and measures were used to evaluate the coaching almost all relied upon coachees’ and/or coaches’ self-reported subjective evaluations of effectiveness. The limitations of using only subjective participant evaluations to assess the success of coaching have been discussed in detail elsewhere (Ely et al., 2010; Passmore & Fillery-Travis, 2011; Passmore & Gibbes, 2007) and include issues such as rating the coach instead of the effectiveness of the coaching process and everything it involves, participants who are going through the coaching process at the time they evaluate it are more likely to rate different aspects more highly compared with those who have time to reflect. Thus, it is not possible from this set of studies to make very strong claims about the effectiveness of coaching in general or its effectiveness compared with other training opportunities. These studies do, however, offer insights into the process of coaching, and it is possible to explore the links between characteristics of the coachee, coach, and coaching process and differences in perceived outcomes (Kilburg, 2004; Lowman, 2005). Furthermore, it could be argued that in each specific study, the sample size and nature often limits generalization of the results. Taken as a whole, however, the set of studies provides consistent results across a wide range of different samples and coaching programs.
The second group consisted of quasi-experimental studies that included either a one group pretest–posttest design or a nonequivalent control group design (34 papers were reviewed). This group of studies offers some potential to begin to assess the overall effectiveness of coaching programs, but again this is limited to participants’ evaluations on a small range of outcomes (Shadish et al., 2002). All the studies in this group focused on coaching techniques only. One particular issue with the studies in both of these groups is that of the Hawthorne or halo effect where improved performance results simply from having someone pay attention to respondents’ attitudes and feelings (Dyck & Neubert, 2010).
To avoid this possibility and to examine claims made about business coaching versus other training or development options (Research Question 2), a third group of studies that combined coaching with other techniques was also identified (23 studies were reviewed). The majority used coaching as part of a more extensive training program, and evaluating coaching was not the sole purpose of the research. It had been hoped that this group of studies might provide some insights into the relative effectiveness of coaching versus other training techniques but more detailed analysis found that this was not possible. Typically, these studies evaluated the whole program with only limited information provided on evaluations of individual components. The discussions of results did, however, provide some additional insights into the contributions of different aspects of coaching to perceived program success. For example, a study by Thach (2002) of 281 executives found that the combination of multi-rater feedback and individual coaching do increase leadership effectiveness up to 60% according to direct report and peer postsurvey feedback.
Conceptual/Theoretical Approaches
The majority of the 111 studies reviewed did not design their research guided by a particular theoretical model or conceptual framework, although the development of the coaching program that was evaluated may have been guided by any number of implicit concepts and theories. Seven concepts or theories emerged from those studies that did explicitly use a theory or concept to guide their research including psychotherapy (Wasylyshyn, Gronsky, & Hass, 2006), cognitive behavior therapy (Cleary & Zimmerman, 2004), approaches designed to improve self-efficacy (Baron & Morin, 2010; Evers, Brouwers, & Tomic, 2006), mindfulness (Spence, Cavanagh, & Grant, 2008), concepts related to leadership (Konczak, Stelly, & Trusty, 2000; Levy, Cober, & Miller, 2002), action learning (Douglas & McCauley, 1999), and the use of attitude and behavior change theories from the field of persuasive communication (Blackman, 2008, 2010). Cox et al. (2014) argued that concepts from adult learning offered potential for coaching practice and could be linked to many of the existing theories used in practice. The present review indicated that, with the exception of the Douglas and McCauley (1999) study using action learning, this area has not been given much consideration by academic researchers providing support for Cox et al.’s (2014) recommendations.
This diversity of theoretical approaches reflects the larger field where ideas range from broad philosophies, such as humanistic psychology (Eggers & Clark, 2000), through to very specific concepts such as leadership, which are connected to multiple theories (Ely et al., 2010), to possible links to other areas of activity such as counseling psychology (Passmore & Gibbes, 2007), which also in turn are connected to multiple theories related to areas of process and outcome, supervision and training, career development, and health to name a few. Many specific coaching activities are recommended for their inclusion of cognitive behavioral therapy (Grant, 2005b) and psychodynamic approaches (Foster & Lendl, 1996), but almost none have been systematically analyzed or tested in coaching research.
The practice of coaching draws from many of the theoretical constructs and psychological models and practices of mentoring, training, and counseling with the counseling the most common, given the background of many coaching consultants. Within the field of counseling, there exist many different theoretical approaches including psychoanalysis, Adlerian therapy, Gestalt therapy, cognitive behavior therapy, and rational emotive therapy (Corey, 2012). Although these are not simple alternatives to each other, they share a common focus on the individual being counseled or coached and challenging how they think about themselves and their situation, and while not seeing these individuals as sick, there is a tendency in the counseling literature to assume that the individual being counseled or coached has a problem (Corey, 2012). Other authors in the coaching literature on coaching tend to see the organization as having the problem and thus look more toward instruction and learning theories as a source of theoretical framework (Fournies, 1999; Wrench, McCroskey, Berletch, Powley, & Wehr, 2008). In this approach, theories from the area of persuasive communication are beginning to emerge as options for coaching practice and research (Peltier, 2001; Wrench et al., 2008). The two theoretical traditions are not necessarily in conflict, and merging the two is likely to produce the best overall outcome.
Research Question 1: Benefits or outcomes of coaching
The following is a summary of the benefits often claimed for coaching offered as an extensive list that includes outcomes relevant to the coachee and to the employing organization. For the coachee, benefits are said to include the following: improving/establishing a work life balance (Gegner, 1997), improving psychological and social competencies, improving/advancing career development (Wales, 2003), improving self-awareness and assertiveness (Bozer, Sarros, & Santora, 2014; Gatling, Castelli, & Matthew, 2013; Wales, 2003), increasing confidence (Wales, 2003), developing relationships/networks/interpersonal skills (Wales, 2003), adapting to change more effectively, helping to set and achieve goals (Smither, London, Flautt, Vargas, & Kucine, 2003), role clarity (Kim, Egan, Kim, & Kim, 2013; Sonesh et al., 2015), and changing behaviors (Grant, 2005a; Wasylyshyn, 2003). For employers, benefits for the organization include increased productivity (Olivero, Denise Bane, & Kopelman, 1997), acting as a support mechanism for other training programs (Wales, 2003), improvements to communication (Graham, Wedman, & Garvin-Kester, 1993), and effectiveness of organizations or teams (Hagen & Garvrilova Aguilar, 2012). There are also suggested outcomes that benefit both the coachee and the employer and these include improvements in interpersonal dynamics/teamwork (Hagen & Garvrilova Aguilar, 2012), higher levels of motivation (Sonesh et al., 2015), increased job satisfaction/job retention/company loyalty (Ellinger, Ellinger, & Keller, 2003; Olivero et al., 1997), and increased skill levels (Wales, 2003), able to deal with stressful situations more effectively and improved leadership and management.
All of the reviewed studies reported that the target coaching program was seen by most participants as effective in some way (Parker-Wilkins, 2006). Overall, there was support for a range of benefits and outcomes with the exception of direct evidence of career advancement, although this was often assumed to be likely to happen in the future. Additional benefits in terms of coachee skills and behaviors were also proposed:
Improved ability to delegate appropriately (McGovern et al., 2001);
Better stress management (Ladergard, 2011; Wales, 2003);
Enhanced ability to address workplace conflict and work autonomously (Blackman, 2008; Gray, Ekinci, & Goregaokar, 2011; Smith Glasgow, Weinstock, Lachman, Dunphy Suplee, & Dreher, 2009);
Increased personal accountability (Kralj, 2001);
More innovation and creativity (Norlander, Bergman, & Archer, 2002; Styhre, 2008);
Higher levels of initiative (Blackman, 2010; Sonesh et al., 2015); and
More flexibility and adaptability (Hall, Otazo, & Hollenback, 1999).
It is important to remember, however, that these conclusions were based almost entirely on participant evaluations conducted during and/or immediately after the coaching experience.
It can be argued that perceived or subjective effectiveness is likely to be an important prerequisite for actual changes in coachee attitudes and behavior, but there is little evidence that describes the nature or extent of this link. Certainly, research into training effectiveness more generally in HRD suggests that this is a complex relationship (Santos & Stuart, 2003) and changes in coachee behavior and attitudes may not directly contribute to better outcomes for employers.
Blackman’s (2006) study asked coachees to list their personal goals for the coaching program they were involved in and the goals they thought their organization might have for the coaching and found there was little overlap between the two. It is possible that an effective coaching program helps the coachee reach their personal goals which may then lead to them leaving the organization. This is supported by a consistent finding across a number of studies (see Gegner, 1997, for an example) where coaching was seen as contributing to a number of personal benefits outside the workplace such as improved family communication and a better work life balance resulting from better time management at work. The review reveals an interesting tension within coaching that has been given little attention to date—the tension between the effectiveness of the coaching for the individual coachee versus the benefits and outcomes for the organization. It is possible that measures of coaching effectiveness that focus solely on elements such as ROI and productivity within the workplaces may not capture all the outcomes of value to the coachee.
Research Question 2: Business coaching compared with other techniques
Much of the research reported in the 111 studies was conducted on specific coaching programs, often run by one of the researchers and so there was little systematic comparative research. As a result there was no clear evidence that coaching is more effective than other training or development techniques and a tendency to assume that if participants rated the coaching as effective that this was defacto evidence of its effectiveness over other techniques. One study did, however, find evidence that when coaching was combined with other more traditional development options, then there was a significant increase in coachee productivity when compared with the traditional option alone (Olivero et al., 1997). This suggests that coaching may be an important tool to assist in the transfer of learning from the development options. However, further studies are clearly needed in this area.
Research Question 3: Factors contributing to effective outcomes
The studies reviewed for this question were organized into a set of themes that emerged from the results and conclusions reported in the individual papers. The identification of these themes followed the same steps as used in content analysis and thematic coding for developing grounded theory (Grbich, 2013). The resulting themes are discussed in more detail in the following sections but are very similar to those proposed by Cox et al. (2014) in their structural analysis of coaching.
Characteristics and behaviors of effective coaches
The findings with regard to effective coach characteristics and skills across the three groups of studies reviewed can be summarized into four consistent themes—integrity, support for the coachee, communication skills (Rekalde, Landeta, & Albizu, 2015) and behaviors, and credibility. Important elements of integrity were trustworthiness (Blattner, 2005; Graham, Wedman, & Garvin-Kester, 1994; Gregory & Levy, 2011; Judge & Cowell, 1997; Kombarakaran, Yang, Baker, & Fernandes, 2008; Kralj, 2001; O’Brion & Palmer, 2010; Orenstein, 2006; Peterson, 1996; Rekalde et al., 2015) and maintenance of confidentiality (Blackman, 2010; Douglas & McCauley, 1999; Rekalde et al., 2015; Segers, Vloeberghs, Henderickx, & Inceoglu, 2011; Taie, 2011). Support for the coachee included empathy (De Haan, Culpin, & Curd, 2011; Gregory & Levy, 2011), being nonjudgmental (Blackman, 2010; Judge & Cowell, 1997), and being able to put issues into a larger perspective (Ahern, 2003; Graham et al., 1994; Orenstein, 2006; Styhre & Josephson, 2007). Communication skills were particularly important (Graham et al., 1994; Longenecker, 2010; Rekalde et al., 2015; Thach, 2002) with particular significance attached to the provision of clear performance expectations (Ahern, 2003; Graham et al., 1994; Orenstein, 2000) and detailed and regular feedback (De Haan, 2008b; Graham et al., 1993, 1994; Gregory & Levy, 2011; Judge & Cowell, 1997; Karsten, 2010; Longenecker, 2010; Miller, 2003; Olivero et al., 1997; Orenstein, 2006; Strayer & Rossett, 1994; Taie, 2011; Wasylyshyn, 2003). In the case of these first three themes, the conclusions are consistent with those made by Joo (2005) and Passmore and Fillery-Travis (2011).
Characteristics of the coachee
In comparison with the coach, the coachee has been given far less attention in research to date. The review found some limited evidence that coachee motivation expressed through amount of effort invested and an open attitude to change (Audet & Couteret, 2012) in the coaching and self-efficacy are important characteristics linked to positive coaching outcomes (Anderson, 2013; Baron & Morin, 2010; Berg & Karlsen, 2012; Blackman, 2006; Bozer, Sarros, & Santora, 2013; De Haan, Duckworth, Birch, & Jones, 2013; Levy et al., 2002). It is important to note that most, but not all, of the coachees in the 111 studies reviewed were high achievers and/or took on the coaching voluntarily, suggesting little variation in motivation and confidence was available to be examined limiting conclusions. There was some evidence, however, that even within this group, there were factors that acted as barriers to coachees’ being able to invest as much time and effort into the process as desired (Blackman, 2006). These included family commitments and an inability to find time in the existing workload.
There has also been some debate in the literature about who is likely to benefit the most from coaching with some authors linking coaching to the development of leaders and managers (Caligiuri, 2006; Riggio & Lee, 2007) and others arguing that coaching is particularly useful for new, young, and less experienced staff (Feldman, 1995). The studies reviewed included a wide range of coachees, from different sectors and at different levels of experience and seniority, and there was no evidence that any one type of coachee benefited more than the others.
Features of the coach–coachee relationship
Two consistent themes emerged from the review with regard to the coach–coachee relationship. The first theme was the need to find a good match between coachee and coach (Ahern, 2003; Brodie Gregory & Levy, 2012; De Haan, 2008b; Gan & Chong, 2015; Graham et al., 1994; Gray & Goregaokar, 2010; G. Jones & Spooner, 2006; Peterson & Millier, 2005; Wasylyshyn, 2003; Wasylyshyn et al., 2006). Few details were provided though on what was required to make a good match. The second theme was that coachees believed that better outcomes were more likely when the coach developed an individual or personalized program for the coachee (Blackman, 2010; Blattner, 2005; Gregory & Levy, 2011; Kiel, Rimmer, Williams, & Doyle, 1996; Orenstein, 2006; Peterson, 1996; Wasylyshyn et al., 2006; Winum, 2005).
Elements of the coaching process
The majority of the reviewed studies examined specific coaching programs in detail and so it is not surprising to find that the largest number of results related to the required or important features of the coaching process. These features included the following:
Establishing the benefits of coaching at the start of the program (Baron & Morin, 2010; Bond & Seneque, 2013);
A focus on setting clear goals or outcomes to be achieved (Baron & Morin, 2010; Blackman, 2008, 2010; De Haan et al., 2011; Evers et al., 2006; Grant, Green, & Rynsaardt, 2010; Gray et al., 2011; Judge & Cowell, 1997; Moen & Federici, 2012; Orenstein, 2006; Scriffignano, 2011; Smither et al., 2003; van Oorsouw, Embregts, & Bosman, 2013);
Providing coachees with a clear understanding of their strengths and weaknesses and an action plan for resolving issues (Judge & Cowell, 1997);
Using coaching with other forms of organizational consultation (Sauer, 1999);
Regular communication and contact between the coach and coachee (Jowett, Kanakoglou, & Passmore, 2012; Sauer, 1999);
Maintaining a long-term focus (Gan & Chong, 2015; Grant et al., 2010; Hall et al., 1999; Judge & Cowell, 1997; Styhre, 2008);
A focus on continuous learning (Cleary & Zimmerman, 2004; De Haan & Nieb, 2012; Douglas & McCauley, 1999; Scherer, Canty, Peterson, & Cooper, 1995; Strayer & Rossett, 1994); and
Ongoing evaluation of progress so the program can be adjusted (Cleary & Zimmerman, 2004; Olivero et al., 1997).
Features of the organizational context
The main theme in the results relevant to the organizational context was that the coachee’s employment organization needs to share the responsibility for the coaching goals and outcomes (Peterson, 1996; Wasylyshyn et al., 2006). This can be achieved through adopting a number of change programs (Kombarakaran et al., 2008; Schnell, 2005), including working on aligning individual and organizational goals (Kim et al., 2013; McGurk, 2011; Winum, 2005), providing support to the coachee to engage in the program (Kombarakaran et al., 2008; Levenson, 2009; Orenstein, 2006; Rocereto, Mosca, Forquer Gupta, & Rosenberg, 2011; Smith Glasgow et al., 2009; Winum, 2005), ensuring commitment from senior management for the coaching program (Baron & Morin, 2010; Blackman, 2010; Gray et al., 2011; Smither et al., 2003), and providing an appreciative environment (Maritz, 2013).
Research Question 4: Understanding coach credibility
Research Question 4 focused on the evidence available to inform key debates on the factors that affect coach credibility. Although there is agreement that coaches must be seen as credible to be effective (Joo, 2005; Judge & Cowell, 1997; Passmore & Fillery-Travis, 2011; Segers et al., 2011; Sue-Chan & Latham, 2004), there is debate around how this credibility can be established. This debate has been between those who believe that coaches do not need skills or expertise in the management or organization sector in which they work but rather need skills in counseling psychology or therapy (Ajzen, 1992; Beutler, Machado, & Neufeldt, 1994; Lafferty, Beutler, & Crago, 1989; Petty & Cacioppo, 1984), and those who argue that coaches must have skills or expertise in either executive management development and/or in the sector in which they are practicing (Berglas, 2002; Eggers & Clark, 2000; Evers et al., 2006; Hagen, 2010; Joo, 2005; Passmore & Fillery-Travis, 2011; Smither et al., 2003; VanFleet, 1999). The evidence available in the 111 studies reviewed here tends toward the latter conclusion, that skills or expertise in the sector or in executive management are preferable (De Haan, 2008a; Judge & Cowell, 1997; Kiel et al., 1996; Orenstein, 2006; Smith Glasgow et al., 2009; Wasylyshyn, 2003). There was also support for the idea that credibility and effectiveness are enhanced by a coach external to the organization that the coachee works for (De Haan, 2008a; Judge & Cowell, 1997; Oberschachtsiek & Scioch, 2015).
Discussion
The “Introduction” section to the present article established five key questions to guide this review. For each question, this summary provides an overall conclusion noting consistency with previous reviews, highlighting new knowledge generated by this review, and suggests directions for future research.
The first question addressed in the review was whether or not business coaching was an effective development tool and in what areas was it particularly effective. The review established some preliminary evidence that business coaching can be effective across a range of different outcomes. This evidence was limited, however, to participants’ self-reported evaluations. Despite numerous discussions in the literature about desirable methodologies for evaluating effectiveness, and a significant increase in the number of studies being conducted into coaching, there has been little change in the approaches taken in the published research. These conclusions are consistent with those of previous reviews (Ely et al., 2010; Feldman & Lankau, 2005; Joo, 2005; Passmore & Fillery-Travis, 2011; Passmore & Gibbes, 2007).
The present review identified, however, a new issue for consideration—the tension between coaching benefits for the individual coachee and coaching benefits for the organization and, indeed, society. This tension arises from elements unique to coaching which are linked to taking a holistic perspective and a focus on the individual and their goals and perspectives and has been given little attention in previous literature. Research is needed that more explicitly measures coachees’ goals and how consistent these are with the organizational goals. These then need to be compared for possible variations and differences of perceived coaching effectiveness. Kaufman and Keller (1994) added to Kirkpatrick’s (1959) four levels of evaluation with a fifth, advising that interventions need to take account of their broader impact on society.
It has also been suggested that business coaching is more effective than other forms of training and development. For example, some authors have argued that business coaching provides better transfer of training to the workplace than other techniques because it is grounded in the coachee’s workplace (Bacon & Spear, 2003; Eggers & Clark, 2000; Joo, 2005; Olivero et al., 1997). Similarly, Van Velsor and Leslie (1995) argued that coaching avoids problems of overdependence that can be a problem with mentoring, and Day (2001) concluded that coaching is preferable to job assignment because coaching addresses overall development and is not limited to a particular set of skills. The second question asked whether business coaching was more effective than other training or development options. The review of 111 empirical studies did not find any clear evidence that business coaching was a more effective staff development option. Again, this was not a novel conclusion. The process used in the present review of dividing the studies into categories based on approaches to research design used did identify a number of studies where coaching was part of a larger training program. As a result, there was some emerging evidence that coaching may be useful to assist transfer of learning from other training options. This conclusion prompts a call for more research into the instructional learning aspects of coaching and that future research needs to adopt more field experimental techniques to systematically compare coaching to other training and development options.
The third question sought the factors that contribute to effective coaching. In addition to the claims made about the effectiveness of business coaching overall, there are also numerous claims made about the details or the characteristics of coaching that contribute to this effectiveness. Figure 1 provides a summary of the most common claims made in this area organized under headings of characteristics of the coach, the coachee, the coaching practice, the coach–coachee relationship, and the wider context or barriers. The framework is consistent with that used in previous reviews (Bennet, 2006; Passmore & Fillery-Travis, 2011) and is based on Kilburg’s (2001) analysis of coaching practice. Webster and Watson (2002) argued that the use of such a framework is preferable to an historical or annotated bibliographic review style as it allows for clearer links to be made to potential concepts and theories. There is consensus that in the case of coaches, integrity, support for the coachee, and good communication skills are critical characteristics, and these features have been noted in previous reviews (Passmore & Fillery-Travis, 2011). The results of our systematic review outline the empirical support found by Maltbia et al. (2014) remembering that overall the evidence of effectiveness is limited in many or the studies reviewed. In addition, Maltbia et al. (2014) offered quite specific direction as to particular communication skills necessary for effective coaching, and the available studies rarely examined the feature of communication in detail. This offers an opportunity for additional research that analyzes the proposed competencies in more detail.

Conceptual Framework.
The fourth question sought to explore the factors that contribute to the credibility of coaching. The present review found evidence and an area for future research suggesting that credibility is enhanced by expertise in the relevant business sector and by being external to the coachee’s work organization. While anecdotal accounts of positive personal experiences of coaching abound, the long-term credibility of coaching must rely on evidence-based studies with robust research, including quasi-experimental and experimental designs based on both cross-sectional and longitudinal designs.
The fifth question was about gaps in the existing research, and despite a considerable increase in the number of empirical papers published, few of the gaps identified in previous reviews have been addressed. More specifically, there continues to be a need for research that assesses the effectiveness of coaching using measures other than subjective evaluations from participants and that uses valid experimental methods, to better understand the impact of coachee versus organizational goals on the coaching experience. Research into coaching effectiveness also needs to explore longer term outcomes as well as immediate impacts using longitudinal rather than cross-sectional research designs. De Meuse et al. (2009) and Ely et al. (2010) offered more details on suggested methodologies. These methodologies also need to be applied to studies that systematically examine the effectiveness of coaching relative to other training or development options.
Overall, this review can be summarized in a comprehensive conceptual framework (see Figure 2 that also shows the areas that offer opportunities for delivering evidence-based improvements). This can include options for improving processes and their interface with organizational contexts and can identify preconditions for coaching effectiveness.

Summary of factors contributing to effective coaching practice.
Conclusion
The present article sought a unique approach to critically examining the issue of coaching effectiveness using a SRm which has not been used previously in the business coaching literature. This approach supported a more comprehensive and focused review incorporating a larger body of empirical evidence than has previously been examined in narrative or meta-analytic reviews. Given the dramatic increase in academic empirical studies evident in recent years, the present review provides an assessment of the effectiveness of coaching based on a larger body of empirical evidence than was previously available, a comprehensive overview of the field that is not limited by a specific focus on one aspect of coaching.
After completing the systematic review, a number of reasons can be put forward to explain the lack of progress in the coaching research reviewed in this article. First, there are practical barriers. Many studies are conducted by coach practitioners in real-world settings making it difficult and often unethical to randomly assign participants to different experimental and control conditions. Even where quasi-experimental designs can be developed, they are likely to involve longer time frames, and the employer organizations may be reluctant to fund coaching programs in these situations. The current state of coaching reflects a continuing concern among researchers involved with establishing coaching as a legitimate HRD activity and with the development of coaches as professionals. To mature as a field, the research needs to move beyond these concerns and focus on developing sound theoretical models based on more sophisticated research.
Webster and Watson (2002), in their guide to good research reviews, cautioned against being overly critical of an area and this seems particularly pertinent to the field of coaching research. It would be easy to suggest that despite a significant increase in the number of empirical studies, little actual progress has been made in demonstrating the effectiveness of coaching or providing guidelines for coaching practice. Alternatively, it could be argued that the growth in studies that evaluate coaching programs in a number of sectors and locations suggests that coaching has now established itself as a legitimate HR development tool and that coaching programs can be positively evaluated by a variety of participants and methods. The present challenge for researchers in this area is to accept this conclusion and begin to examine coaching in more detail and with a more critical approach. This article offers direction for a more evidence-based conceptual framework to guide future research and practice.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
