Abstract
This study examines the effectiveness of an explicitly nondirective team coaching intervention in enhancing team performance and individual teamwork competencies within self-managed teams. The intervention was implemented across multiple sessions with nine teams in three organizations and was evaluated using a convergent mixed-methods design, in which GROW functioned solely as a light process scaffold while coaches withheld content advice. Quantitative results showed significant improvements in conflict resolution and planning. Qualitative data revealed increases in team alignment, communication, and performance evaluation. These findings suggest that nondirective team coaching can foster autonomy, adaptability, and sustainable team development in dynamic work environments.
Introduction
Organizational design has progressively shifted from individual productivity to team dynamics, recognizing effectiveness as an emergent team-level outcome (Kozlowski & Ilgen, 2006; Salas et al., 2008, 2018). Despite advances in team management practices, many established Team Development Interventions (TDIs) (such as team training and some team coaching implementations), include directive elements in which facilitators act as experts imparting solutions (Lacerenza et al., 2018; Shuffler et al., 2018). Such direction can be useful, yet it may constrain the development of critical competencies like self-management and adaptability (Beal et al., 2003; Mesmer-Magnus & DeChurch, 2009).
To address this gap, the study investigates the effectiveness of nondirective team coaching in enhancing both individual competencies, such as communication, conflict resolution, and planning, and overall team effectiveness. The central hypothesis is that nondirective team coaching strengthens individual capabilities while fostering collective performance. Rather than proposing a new model, our contribution is to operationalize and empirically evaluate an explicitly nondirective implementation of team coaching in small-group settings, where GROW structures the conversation without coach-provided solutions.
A mixed-method quasi-experimental design is employed to analyze the impact of the intervention on both dimensions. Grounded in Carl Rogers' client-centered approach (Rogers, 1951), the coaching model promotes self-management and adaptability by enabling teams to access their own resources and insights. The findings aim to advance team coaching practices suited to contemporary organizational contexts, while offering practical value for leaders and teams navigating autonomy, complexity, and shared goals.
The remainder proceeds as follows. The literature review synthesizes team effectiveness, teamwork competencies, team coaching, and nondirectivity. The methodology details design, participants, intervention, and analyses. Results report quantitative and qualitative findings, followed by an Integration section. The discussion addresses implications and limitations, and the conclusions outline practical applications and future research.
Literature Review
Team Effectiveness Models and Teamwork Competencies
Team effectiveness has been a central topic in organizational research. Foundational models such as Hackman and Wageman (2005) framework emphasize conditions like compelling direction, enabling structure, and supportive context as drivers of high performance. The IMOI model (Ilgen et al., 2005) and its extensions (Mathieu et al., 2008) describe contextual factors and emergent states, illustrating how individual competencies contribute to dynamic team outcomes.
Regarding team competencies, drawing from the foundational study by Larson and LaFasto (1989), research on individual teamwork competencies has identified two key dimensions: technical skills and personal competencies. Building on this foundation, further explorations have been conducted by Stevens and Campion (1999), enhanced by Salas et al. (2009) and Cannon-Bowers and Salas (2014), identifying three core components to effective teamwork: (a) knowledge, which encompasses not just technical abilities but also skills necessary for collaborative efforts, such as conflict resolution, problem detection, and assertiveness; (b) skills, specifically those that can be developed through the exchange of information and feedback within the team, like task organization, closed-loop communication, and adaptability; and (c) attitudes, which reflect an individual's values and beliefs about teamwork, including team cohesion, mutual trust, and psychological safety. From these insights, teamwork competencies have been distilled into two primary dimensions by subsequent research (Aguado et al., 2014; Stevens & Campion, 1999): interpersonal competencies, including conflict resolution, collaborative problem-solving, and effective communication, and self-management competencies related to managing performance objectives and strategic planning.
Team Coaching
According to Shuffler et al. (2018), TDIs are defined as requiring intentional actions targeted at team performance trajectories. These actions aim to (1) improve and support struggling or failing teams, (2) maintain and sustain adequately performing teams, and (3) grow and maximize the capacities of teams ready to mature to a higher level of performance. Empirically demonstrated TDIs include for example, Team Task Analysis (Cannon-Bowers & Bowers, 2011; Lin et al., 2014), Team Composition (Mathieu et al., 2014), or Team Debriefing (Tannenbaum & Cerasoli, 2013). However, team coaching is the most recent TDI, and its research base is expanding rapidly, which still lacks rigor in terms of empirical experience (Hastings & Pennington, 2019; Jones et al., 2019; Rousseau et al., 2013; Widdowson et al., 2020) and on which this research is focused to provide knowledge in this gap.
Regarding Team Coaching, there has been a recent trend toward narrowing and refining its definition, with over 10 definitions existing (Clutterbuck et al., 2019). While there is variation in terminology, most conceptualizations converge on the view that Team Coaching involves engaging with an entire work team as a collective process and ideally employing a nondirective, exploratory approach (Clutterbuck et al., 2019). Despite this theoretical emphasis, when considering the most utilized models of team coaching in empirical studies (Clutterbuck et al., 2019; Hawkins, 2021; Thornton, 2010; Wageman et al., 2005) and categorizing them based on Clutterbuck's model (Clutterbuck, 2007), which ranges from directive to nondirective interventions, it becomes evident that none of the interventions to date are purely nondirective. Instead, they are hybrids incorporating coaching elements alongside other TDIs. The eclectic nature of current team coaching models involves incorporating elements of training, mentoring, and facilitation. This often leads to the transfer of knowledge from the coach to the team, positioning the coach as the expert who imparts solutions (Clutterbuck, 2007; Thornton, 2010). While this can be beneficial in certain contexts, it may inadvertently impede team autonomy and limit opportunities for self-generated insights and strategies.
Despite the theoretical advantages, empirical research on team coaching interventions remains scarce. A review of the literature conducted by Carr and Peters (2013, 2019), Hawkins et al. (2022), and later expanded by Traylor et al. (2020), identified only nine studies that explicitly examine structured team coaching interventions. The remaining studies have primarily examined leadership behaviors associated with a coaching role instead of coaching interventions. Beyond conceptual models, recent empirical studies broaden the evidence base: a randomized controlled trial contrasting team coaching with team facilitation reported gains in cohesion and psychological safety (Passmore et al., 2024); applied research documents teamwork coaching in research-development teams (Stephens et al., 2024); evidence-based guidance on developing internal team coaches synthesizes competencies and training pathways (Fernández Castillo et al., 2024); and meta-analytic findings support team reflexivity as a mechanism linked to performance (Leblanc et al., 2024). Together, these studies strengthen the rationale for evaluating an explicitly nondirective protocol in small-group settings.
Nondirectivity in Coaching
The concept of “nondirectivity” originates from client-cantered therapy (Greene, 2017; Rogers, 1951), emphasizing the belief that individuals possess the intrinsic resources necessary for personal growth and problem-solving. In the coaching context, nondirectivity entails the coach adopting an attitude of empathetic listening, unconditional acceptance, and authenticity, refraining from imposing external interpretations or solutions on the coachee (Batten & Batten, 1967). Gallwey et al. (1997) introduced this coaching approach and advocated effectively managing one's internal dialogues. Subsequently, Renshaw and Alexander (2005) applied this technique to businesses and large corporations, laying the groundwork for the GROW model. The concept was further developed and popularized by Whitmore (2002), Thomson (2009, 2013), and Rodríguez (2021), who emphasized the importance of nondirective coaching as a core principle in this coaching style.
Although nondirectivity is widely recognized and applied in individual coaching (Ravier, 2021; Rogers, 1951; Whitmore, 2002) and acknowledged within the domain of team coaching (Clutterbuck, 2007), there is a notable absence of both theoretical and empirical studies specifically addressing explicitly nondirective team coaching. The only early reference to nondirectivity in group contexts appears in The Non-Directive Approach in Group and Community Work (Batten & Batten, 1967), which explored its application in fostering social development, but did not propose a replicable intervention model.
In summary, although team coaching shows promise, causal evidence on explicitly nondirective protocols in self-managed team settings remains limited. We therefore examine whether such a protocol is associated with targeted shifts in teamwork competencies and perceived team effectiveness.
Methodology
This study employed a mixed-method quasi-experimental design, which allows for the evaluation of causal relationships in real-world settings where random assignment is not feasible (Cook & Campbell, 1986). Specifically, a post-test within-subjects design was used to measure individual teamwork competencies quantitatively before and after the intervention, while team effectiveness was assessed qualitatively throughout the coaching process. While the absence of a control group limits the ability to establish causal relationships with certainty, the within-subjects approach combined with triangulated qualitative data supports the interpretation of observed changes as potentially attributable to the coaching process. Moreover, this approach preserves the natural team environment, thereby enhancing ecological validity, an essential consideration when studying behaviorally rooted interventions (Dumas & Sanchez-Burks, 2015). The research aimed to address two primary research questions: (RQ1) Are there differences in the individual competencies of team members after participating in the nondirective team coaching process? (RQ2) How does nondirective team coaching influence overall team effectiveness?
Nondirective Team Coaching Intervention
This model builds on Ravier’s (2021) work on individual nondirective interventions. It integrates three elements: (a) the conversational structure derived from the GROW model, (b) techniques like focused questioning, mirroring, recapitulation, and objective communication (Miller & Rollnick, 2012; Ravier, 2021), as well as (c) a procedural structure that outlines the length and intervals of each session, and applying them to the team level, results in the following model of nondirective team coaching intervention.
The coaching process is divided into three main phases, each with specific objectives and activities (Figure 1). The first phase, the Preliminary Phase, establishes the foundation for the coaching process and involves initial engagement with the client and the team. The objectives of this phase are to initiate contact with the client to understand their needs and expectations, assess whether team coaching is the appropriate intervention, clarify the nondirective coaching approach, and define the general objectives of the coaching process. Key activities include consulting with the client to listen to their reasons for seeking team coaching and their desired outcomes, collaboratively establishing a general goal for the coaching process, and ensuring that the team meets the necessary preconditions for coaching, such as readiness, awareness, responsibility, and voluntary participation, as emphasized by Ravier (2021).
Team Coaching Model Proposed. Source: Own Elaboration Based on (Ravier, 2021).
Exploratory sessions are conducted with higher-level stakeholders to refine objectives before engaging the team. During these sessions, specific, measurable, achievable, relevant, and time-bound (SMART) objectives are defined (Doran, 1981), the current reality is explored through targeted questioning, and success criteria are established. In addition, critical factors—referring to the attitudes, elements, or conditions deemed essential for achieving the objectives—are identified and prioritized to ensure alignment with the overall goals of the intervention.
The second phase, the Development Phase, involves conducting the coaching sessions with the team, focusing on achieving the defined objectives. Each session lasts between 90 and 120 min to maintain focus and engagement, and sessions are scheduled with intervals that allow for reflection and application of insights (at least 3 weeks between sessions). In the first coaching session, the team revisits and refines the objectives established during the preliminary phase, ensuring alignment with organizational goals and team needs. Through nondirective questioning, the team explores their current situation, challenges, and opportunities, fostering a deep understanding of their context. Moreover, success indicators are defined, which will serve as benchmarks to determine whether the objectives have been achieved upon completion of the process. Similar to the exploratory session, the team also identifies critical factors.
Subsequent coaching sessions focus on establishing second-order objectives that align with the overarching coaching goals. The team reflects on prior actions, learning experiences, and progress toward objectives, collaboratively developing strategies and action plans to address identified challenges. Each session concludes with a reflection on the team's learning outcomes and achievements, fostering continuity and ownership of the process. This evaluation helps maintain consistency with the intervention's core priorities and reinforces collective ownership of the process.
The final evaluation and closure session focuses on reviewing the general objective alongside the subobjectives, learning outcomes, and achievements from previous sessions. The success indicators established in the initial session are revisited to assess whether the overall coaching goal has been met. Additionally, the session consolidates the team's learning and accomplishments from the entire coaching process while ensuring confidentiality in the communication of results to stakeholders. Key activities include reflecting on the team's progress and growth, evaluating the success criteria defined during the preliminary phase, and facilitating a decision-making session. During this session, the team determines the next steps, such as continuing self-directed development, setting new goals, or identifying areas for further improvement.
The Finalisation Phase involves evaluating the coaching process in collaboration with stakeholders. Feedback is provided to team leaders or clients regarding the overall outcomes, while ensuring the confidentiality of specific team discussions is strictly maintained.
All teams were facilitated by the same certified coach following a standardized session script to reduce facilitator variability. Session notes were member-validated, and survey data were collected anonymously. GROW was employed solely as a light process scaffold (e.g., clarifying goals and exploring options), and the coach withheld content advice throughout.
Quasi-Experiment
The effectiveness of the coaching intervention is assessed using a mixed-method approach that combines quantitative and qualitative data. The quantitative component involves pre- and post-coaching assessments of individual competencies within the team, utilizing validated instruments such as the Teamwork Competency Test (TWCT) (Aguado et al., 2014). Subsequently, statistical analyses are conducted to determine the significance of observed variations (Dimas et al., 2016; Körner et al., 2017).
The qualitative component entails the ongoing collection of team members’ reflections and achievements throughout each coaching session. These data are recorded at the conclusion of each session, coded, and, where applicable, quantified to allow for integration with the quantitative results. Such triangulation of data enhances the reliability of the findings and affords a more holistic understanding of the coaching process (Creswell & Clark, 2017).
Participants
The research involved nine teams comprising a total of 40 members from one technology startup and two nonprofit organizations. The sample was selected through nonprobabilistic purposive sampling, based on organizational willingness to participate, and team stability conditions. While the sample size is modest, it is consistent with prior mixed-methods research in coaching and team development (e.g., Dimas et al., 2016; Körner et al., 2017), where in-depth intervention studies typically involve smaller, context-specific samples. All participants completed both pre- and postintervention assessments, and no attrition was recorded during the coaching process. To maintain confidentiality, the names of these entities are not disclosed. The teams were categorized into three types based on their organizational structures: (1) type 1 (startup, team 1): this team included the founders who participated as team members without a distinct hierarchical structure. (2) Type 2 (NGO, teams 2 and 3): these teams operated without designated leaders, despite having a board of directors overseeing the organization. (3) Type 3 (NGO, teams 4–9): each of these teams had an appointed leader managing the team under the guidance of a board of directors. All teams were self-managed but varied in their structural hierarchies, demonstrating that the coaching intervention could be applied across diverse organizational setups. Each team comprised 4 to 6 members and was classified as either “stable” or “relatively stable” (Tannenbaum & Salas, 2020).
Measures
For the quantitative component, individual teamwork competencies were assessed using the TWCT developed by Aguado et al. (2014), which builds upon the foundational work of Stevens and Campion (1999). The TWCT consists of 36 items rated on a four-point Likert scale ranging from 1 (never/rarely) to 4 (always/almost always). The competencies and subcompetencies are detailed in Table 1, delineating skills such as conflict resolution, collaborative problem-solving, communication, performance objective management, and planning. This instrument has demonstrated reliability and validity in measuring individual teamwork competencies across various organizational contexts. Each of these domains is further broken down into specific subcompetencies, allowing for a more nuanced assessment of teamwork behaviors, as detailed in Table 1.
Competency Framework of the TWCT: Codes and Descriptions of Competencies and Subcompetencies (Aguado et al., 2014).
Note: The codes in this table signify the following: KSA stands for Knowledge, Skills, and Attitude, while the letters A, B, C, … denote the order of the variables, and the accompanying numbers indicate the order of subvariables within each category.
Qualitative data was collected through the documentation of learning experiences and achievements during the coaching sessions. A certified coach facilitated each session and recorded observations, discussions, and reflections on a shared whiteboard visible to all team members. These notes captured the team's insights, challenges, and progress throughout the coaching process. After each session, the notes were transcribed and sent to the team members for validation, ensuring accuracy, and respecting confidentiality.
Procedure
An initial online meeting was convened with all prospective participants to explain the study's objectives, detail the methodological procedures, and address pertinent ethical considerations. During this session, participants were informed about the data collection methods, the measures in place to ensure anonymity and confidentiality, and their right to withdraw from the study at any point without repercussions. Teams that agreed to participate were provided with online access to the TWCT for the pre-intervention assessment. The coaching process itself spanned 5 to 9 months, comprising seven team coaching sessions and at least two sessions with senior organizational leaders, for a total of nine sessions. Each 90min session was conducted in person within a dedicated room equipped with a whiteboard for real-time notetaking and visualization. A minimum interval of 3 weeks between sessions was observed, enabling participants to reflect on and apply the insights gained before proceeding to the next meeting.
The coaching intervention followed the previously outlined nondirective coaching model, emphasizing the facilitation of team discussions, goal setting, and problem-solving without imposing external solutions. Upon completion of the coaching process, participants were asked to complete the TWCT once again to evaluate any changes in their individual competencies. In addition, qualitative data derived from the coaching sessions were compiled for subsequent analysis, with particular attention to the learning experiences and achievements reported by each team. All participants gave informed consent, and the study adhered to the ethical guidelines of the host institution (Universidad Politécnica de Madrid). Particular attention was paid to the researcher's dual role as external coach and principal investigator, which was acknowledged as a potential source of bias during the study design. To mitigate this, several safeguards were implemented: (1) all coaching sessions followed a standardized, nondirective protocol designed to minimize facilitator influence; (2) session notes were validated by participants to ensure transparency and reduce interpretive subjectivity; and (3) quantitative data were collected anonymously using a validated instrument. The combination of a structured intervention design and a nondirective coaching style facilitates replicability by other researchers and enables the comparison of results across studies, thereby supporting methodological consistency and reducing bias. Confidentiality and voluntary participation were ensured throughout the study.
Data Analysis
This study adopted a convergent parallel mixed-method design (Creswell & Clark, 2017), wherein quantitative and qualitative data were collected simultaneously, analyzed independently, and then merged to provide a more comprehensive understanding of the coaching intervention's effects. The parallel collection of data enabled an in-depth exploration of both measurable changes in individual competencies and participants’ subjective experiences throughout the coaching process.
The quantitative component of the study focused on the TWCT. Pre- and postintervention scores were gathered for each participant to assess the coaching intervention's impact on individual teamwork competencies. Before conducting inferential tests, the psychometric properties of the TWCT were examined in the study sample to confirm reliability and validity, thereby ensuring the appropriateness of subsequent statistical procedures.
Data distributions were then assessed for normality using the Shapiro–Wilk test (Shapiro & Wilk, 1965). If the data conformed to a normal distribution, paired samples t-tests were performed to compare pre- and postintervention scores. In cases where the normality assumption was violated, the Wilcoxon signed-rank test served as a nonparametric alternative (Woolson, 2005). The hypotheses tested for each competency domain posited no differences in teamwork competency before and after the coaching intervention (null hypothesis) versus differences in teamwork competency over the same period (alternative hypothesis).
Effect sizes were calculated using Cohen's d to determine practical significance (Cohen, 2013). To mitigate the risk of type I errors resulting from multiple comparisons, the Bonferroni correction (Holm, 1979) was applied, thus increasing the robustness of the findings. Through these procedures, the statistical analysis offered insight into both the significance and magnitude of changes in teamwork competency across the different domains measured by the TWCT.
The qualitative dimension of the study aimed to capture the learning experiences and achievements reported by participants during the coaching sessions. Notes from each coaching session, which included the team's reflections recorded on a shared whiteboard, were transcribed verbatim for subsequent analysis. Table 2 presents the coding framework developed to classify the themes emerging from team reflections during the coaching sessions.
Categorization and Coding of the Learning and Achievements of the Teams in the Coaching Process Sessions Based on the Literature Review and the Effectiveness Model of Wageman et al. (2005).
These transcriptions were then imported into ATLAS.ti software (Friese, 2019), where they were coded and analyzed thematically. The initial coding scheme drew upon the team effectiveness model proposed by Wageman et al. (2005), encompassing elements such as goal alignment, internal communication, and performance evaluation. Once open coding identified salient segments of text related to these elements, axial coding was used to relate the emergent categories to one another. Selective coding then refined and integrated the categories into central themes that reflected the essence of the participants’ experiences (Table 2). Frequency analysis of the coded segments highlighted the most prominent themes.
Integration of Quantitative and Qualitative Data
Following the convergent parallel mixed-method design, the study integrated quantitative and qualitative findings to provide a comprehensive assessment of the coaching intervention's effectiveness. To enable direct comparison, a Focus Index was developed as a standardized indicator of the relative emphasis placed on each teamwork competency during the coaching sessions. This index was constructed by counting the frequency of qualitative codes associated with each TWCT domain across all teams and sessions. These raw frequencies were then normalized using Min–Max scaling (James et al., 2013), resulting in values ranging from 0 to 1. Higher Focus Index scores indicate that a given competency received greater attention and reflection throughout the coaching process.
The association between qualitative codes and TWCT competencies was established through a preliminary literature review and conceptual analysis. Specifically, each code emerging from the thematic analysis was mapped to the competency it most closely represented, based on the operational definitions provided in the TWCT framework (Aguado et al., 2014). This process ensured conceptual alignment and analytic consistency between the qualitative and quantitative components of the study.
For instance, if planning-related codes appeared 25 times and conflict resolution codes appeared 40 times, the normalized Focus Index values might be 0.62 and 1.00, respectively, reflecting their relative prominence in the teams’ discourse. Table 3 presents the mapping framework used to operationalize this integration.
Mapping Between Qualitative Codes and TWCT Competencies.
Results
Quantitative Findings
One of the objectives of this study was to assess changes in individual teamwork competencies following a nondirective team coaching intervention. A total of 40 participants from nine teams provided complete TWCT responses at pre- and postintervention time points.
The results revealed statistically significant gains in two specific competencies (see Table 4). Conflict resolution (KSA_A) exhibited a mean increase from 3.07 (SD = 0.41) to 3.26 (SD = 0.36), translating to a 6.10% improvement (adjusted p = .0093). Planning (KSA_E) also demonstrated a statistically significant change, with mean scores rising from 3.25 (SD = 0.33) to 3.41 (SD = 0.37), corresponding to a 4.74% improvement (adjusted p = .0381). These findings indicate that the coaching intervention had a targeted and positive influence on these competencies.
Significance, Effect Size, and Descriptive Analysis of the Results Obtained in the Quantitative Analysis.
Note: p-value<.05*; p-value<0.01**; p-value<.001***.
While additional competencies such as Active Listening (KSA_C.1), collaborative problem-solving (KSA_B.1 and KSA_B.2), supervision (KSA_D.1), and feedback (KSA_D.2) showed positive percentage increases, none of these improvements remained significant after applying the Bonferroni correction. Consequently, the null hypothesis (H₀: no differences before and after the coaching intervention) was rejected exclusively for conflict resolution and planning, supporting the alternative hypothesis (H₁) that the coaching intervention produced gains in these domains.
In addition to examining statistical significance, the internal consistency of each TWCT subscale was assessed using Cronbach's alpha. Most subscales demonstrated acceptable or good reliability, with values ranging from 0.610 (active listening) to 0.803 (feedback). However, the internal consistency for collaborative problem-solving, both group (KSA_B.1) and individual (KSA_B.2), with alphas of 0.112 and 0.255, respectively. These values suggest that the items within each subscale may not adequately reflect a coherent latent construct in the current sample, which limits the confidence in interpreting observed changes as true improvements in these competencies. This is likely due in part to the limited number of items (three and two, respectively), which inherently constrains reliability estimates. Prior research indicates that low item counts and subtle semantic differences between items can reduce internal consistency (Aguado et al., 2014; Tavakol & Dennick, 2011). Consequently, results associated with these subscales should be interpreted with caution, and future studies might consider revising or expanding the item pools for collaborative problem-solving dimensions to improve psychometric robustness.
To determine the practical significance of observed changes, Cohen's “d” for paired samples was computed. Conflict resolution (KSA_A) showed a moderate effect size (d = 0.505), reinforcing its statistical and practical relevance. Planning (KSA_E) also showed a moderate effect (d = 0.415). Other competencies, such as active listening (d = 0.294), monitoring (d = 0.275), and collaborative problem-solving (group: d = 0.240; individual: d = 0.123), showed small or very small effects, which, while not statistically significant after correction, suggest emerging developmental trends. Feedback (KSA_D.2), though methodologically sound in terms of internal consistency, exhibited the smallest effect size (d = 0.107), pointing to potential challenges in promoting behavioral change in this area through nondirective approaches. These effect sizes are consistent with findings from meta-analyses of workplace coaching, which typically report small to moderate effects on teamwork-related outcomes (Jones et al., 2016; Theeboom et al., 2014).
Overall, these results support the targeted impact of the intervention on conflict resolution and planning, while highlighting differential responsiveness across competencies. This is consistent with the importance of combining statistical significance, psychometric reliability, and practical magnitude to evaluate the full impact of coaching interventions.
Qualitative Findings
A review of the frequency analysis revealed that performance evaluation emerged as the most frequently applied code, highlighting its central role in fostering reflective practices, shared accountability, and team learning during the coaching sessions (Figure 2). This pattern underscores how participants not only assessed performance but also engaged in deeper collective learning. Notably, several of the most frequently discussed themes, such as performance evaluation, planning, collaboration, and conflict resolution, correspond directly to the competencies assessed by the TWCT, reinforcing the alignment between participants’ reflections and the measurable improvements identified.
Bar chart represents the frequency with which codes appear in the achievements and learning from coaching sessions. Source: Own elaboration.
Efficiency and productivity and work processes also registered high frequencies, underscoring the emphasis participants placed on streamlining operations. Other notable areas included collaboration and teamwork, skill development, and goal definition, suggesting that the coaching discussions were wide-ranging and addressed various dimensions of team functioning. This convergence between what teams discussed and what was quantitatively measured reinforces the reliability of the observed improvements. In particular, the prominence of performance evaluation as a theme suggests that the coaching process encouraged sustained reflection on team effectiveness, accountability, and collective learning, beyond individual roles or contributions.
Integration of Quantitative and Qualitative Findings
The integration revealed particularly high Focus Index values for conflict resolution (KSA_A), which was linked to collaboration and performance evaluation, and for planning (KSA_E), which was closely tied to goal alignment, work processes, and efficiency. By contrast, collaborative problem-solving (KSA_B) and active listening (KSA_C.1) showed more modest index values, suggesting that these areas, while positively influenced, were not emphasized to the same extent.
A graphical plot was generated to illustrate the relationship between the percentage change in TWCT scores (pre- vs. postintervention) and the corresponding Focus Index for each competency (Figure 3). In the plot, competencies with an asterisk (*) represent those where pre–post changes were statistically significant after adjustment. The Y-axis reflects the percentage change in the mean TWCT score for each competency, indicating the measurable effect of the coaching intervention. The color gradient of the bars, based on the Focus Index, represents the frequency and emphasis with which each competency was addressed in the coaching sessions. Competencies that received greater attention in the coaching sessions generally exhibited larger improvements in mean TWCT scores. This pattern supports the notion that targeted, in-depth discussions and reflections on specific skills foster more competency development. This pattern is consistent with the idea that competencies receiving more focused attention and collective reflection during coaching may have benefited from more intentional practice and reinforcement, resulting in greater observable gains.
Integration of TWCT score changes and Focus Index across competencies. Source: Own elaboration.
By juxtaposing quantitative score changes with the qualitative Focus Index, the study offers compelling evidence that the nondirective coaching intervention was most successful where teams devoted time and effort to discussing and refining particular competencies. Conflict resolution and planning stood out as areas with both qualitative emphasis and statistically significant gains, while Supervision showed promising quantitative improvement aligned with moderate emphasis on the qualitative data. Overall, these integrated findings underscore the importance of combining quantitative metrics with qualitative context to capture the full spectrum of an intervention's impact on teamwork competencies and team dynamics.
Discussion
This study provides empirical evidence on the impact of nondirective team coaching in enhancing individual teamwork competencies and team effectiveness. The results indicate significant improvements in conflict resolution and planning competencies, suggesting that fostering a self-directed learning environment enables teams to develop problem-solving capabilities without reliance on external directives. These findings support RQ1, demonstrating that nondirective coaching facilitates competency development, particularly in areas requiring structured decision-making and adaptability.
These results are consistent with prior empirical evidence showing that nondirective or strength-based coaching interventions of medium duration tend to produce small but significant improvements in interpersonal and self-management competencies (Theeboom et al., 2014). Additionally, the qualitative insights reveal that teams increased their focus on performance processes, supporting RQ2 by highlighting the positive influence of nondirective coaching on team effectiveness. The present study contributes by operationalizing an explicitly nondirective intervention, reinforcing prior theoretical assumptions that autonomy-driven coaching enhances self-management and problem-solving skills (Edmondson & Harvey, 2018).
Compared to existing team coaching models, our approach introduces a structured nondirective methodology, which contrasts with the directive emphasis of Hackman and Wageman (2005). Their model, widely used in team effectiveness research, suggests that coaching should focus primarily on task performance rather than interpersonal processes. In contrast, our study demonstrates that nondirective coaching supports not only performance-related outcomes but also competencies such as planning and conflict resolution, which are critical for self-managed teams.
Similarly, Hawkins’ (2021) Systemic Team Coaching model integrates coaching with organizational ecosystems, emphasizing stakeholder engagement and leadership alignment. While our approach does not extend coaching beyond the team level, it aligns with Hawkins’ principle of continuous team learning through iterative coaching cycles.
On the other hand, Clutterbuck's (2007, 2019) models, including PERILL, focus on structured facilitation and balance directive and nondirective elements. Our explicitly nondirective model builds upon Clutterbuck's concept of self-sustaining team learning, reinforcing the idea that teams can develop autonomously without external facilitation, provided they have the right support mechanisms.
Unlike traditional coaching models that rely solely on retrospective reflections or external feedback, this approach enables teams to take ownership of their learning process. This method aligns with best practices in coaching evaluation (Busetto et al., 2020; Creswell & Clark, 2017), offering a replicable framework for both researchers and practitioners.
While the study hypothesized that all teamwork competencies would improve, results showed that feedback and collaborative problem-solving did not exhibit statistically significant improvements. This suggests that competencies may respond differentially depending on their nature in an explicitly nondirective framework, reinforcing the idea that certain skills may require structured interventions (Salas et al., 2007). Additionally, qualitative data revealed that teams focused more on performance-related processes than interpersonal dynamics, an insight that may inform future refinements of nondirective coaching methodologies.
Although only conflict resolution and planning exhibited statistically significant improvements after adjusting for multiple comparisons, other competencies such as active listening, supervision, and collaborative problem-solving showed consistent positive trends. While these changes did not reach significance under Bonferroni correction, they may nonetheless reflect emerging patterns of growth that warrant attention. Prior research has shown that small, incremental behavioral changes often precede more robust developmental effects (Cohen, 2013). Moreover, qualitative data from this study reinforces perceived progress in these domains, suggesting that measurable change may unfold gradually and might be more evident in longitudinal assessments.
This structured coaching model not only offers a promising alternative to traditional directive interventions and includes a replicable measurement protocol for assessing coaching effectiveness. By integrating a dual-measurement methodology, the study contributes to the empirical foundation for nondirective interventions and highlights their potential relevance in dynamic environments where adaptability and self-regulation are essential. These results align with recent studies showing that conflict resolution and planning competencies tend to improve most clearly after sustained reflection and team-led learning cycles (García-Galán et al., 2024). Meta-analyses also indicate that coaching typically produces small to moderate effect sizes in team-related outcomes, particularly in communication and coordination (Jones et al., 2016; McEwan et al., 2017), reinforcing the value of even modest gains in complex team environments.
Therefore, from a theoretical perspective, it extends the empirical foundation of nondirective team coaching by demonstrating its effectiveness in competency development and team performance. Unlike traditional models that incorporate directive elements, this study validates a structured, explicitly nondirective framework, expanding the scope of team coaching methodologies especially in the context of self-managed, adaptive teams (Clutterbuck et al., 2019; Rousseau et al., 2013; Whitmore, 2002; Widdowson et al., 2020). Likewise, from a practical perspective, the findings provide organizations with a structured nondirective coaching model that can be applied to enhance team autonomy, adaptability, and strategic decision-making. The model provides teams with a structured space for self-reflection and iterative learning aligned with the coaching process. Finally, from a methodological standpoint, this study integrates a dual-measurement approach, combining the TWCT for quantitative assessment and qualitative session analyses for a deeper understanding of team learning dynamics. This structured and replicable framework enhances the rigor of coaching intervention evaluations, addressing gaps in previous studies that relied primarily on self-reported measures or case studies (Dimas et al., 2016; Körner et al., 2017).
Conclusion
In multisession team coaching programs, organizations can strengthen planning and conflict-resolution routines by installing brief reflection rituals between sessions (retrospectives, check-ins) and scheduling performance-evaluation reviews; set cadence and fidelity by running approximately seven sessions over 5 to 9 months with intervals of at least 3 weeks, defining success indicators early, using a standardized script, and member-validating session notes; and embed gains through explicit team routines (e.g., planning boards, decision logs, and conflict-handling protocols) so that improvements translate into day-to-day practice.
A key contribution lies in addressing a gap in the literature: the scarcity of empirical research on nondirective team coaching. While most models blend directive and nondirective elements, our study implements an explicitly nondirective protocol supported by light process scaffolding (e.g., GROW/SMART) while coaches withhold content advice. Teams may nonetheless ascribe expert power to coaches; our protocol seeks to minimize, rather than eliminate, such attributions. Results indicate that this approach is particularly effective for fostering planning and conflict resolution, key competencies for dynamic, uncertain environments. These findings reinforce the theoretical value of nondirectivity as a driver of team autonomy and sustained learning.
Methodologically, the study's strength is its use of a convergent mixed-methods design that integrates quantitative assessments (TWCT) with qualitative thematic analysis grounded in team effectiveness theory. Reliability was acceptable across most subscales, and effect sizes confirmed significant gains. Qualitative insights deepened the interpretation of how the coaching influenced individual and collective development. This triangulated approach enhances the validity of findings and supports future replication.
Limitations include a small sample of nine teams from startups and nonprofits, limiting generalizability. The medium-term duration (5–9 months) offers insights into short-term shifts but not long-term retention. The lack of objective performance metrics restricts conclusions on productivity outcomes. Findings may partially reflect coach effects. Although using a single coach and a standardized protocol reduces between-facilitator variance, we cannot isolate coach-specific contributions. Self-report data may be subject to social-desirability toward the coach–researcher; anonymity and member-validation mitigate (but do not remove) this risk. Qualitative interpretations may also reflect researcher-as-coach bias; we mitigated this through participant validation and a prespecified coding scheme.
Future studies should apply the model in varied organizational and cultural settings, incorporate objective measures of team effectiveness, and assess long-term sustainability of gains. Research might also explore hybrid models to determine optimal combinations of directive and nondirective coaching.
In summary, this study offers a foundational contribution to the understanding of nondirective team coaching. By emphasizing autonomy, iterative team learning, and a rigorous mixed-methods design, it proposes a replicable framework with value for both scholars and practitioners. The findings suggest the potential of nondirective coaching to enhance adaptability, ownership, and strategic capability in today's complex team environments.
Footnotes
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Center for Special Education Research, Institute of Education Sciences (grant number R324B170017).
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
