Project performance: an analysis based on open innovation and leadership

Open innovation rests on the intentional crossing of organizational boundaries in the pursuit and use of knowledge. Rather than relying exclusively on internal resources, firms combine what they already know with technologies, ideas, and expertise obtained through external relationships. This logic encompasses two distinct movements: inbound activities bring outside knowledge into innovation work, whereas outbound activities place internally developed knowledge in external contexts (Chesbrough, 2003; Moretti & Biancardi, 2020; Xu & Felzensztein, 2025).

Product development teams offer a useful setting for studying these movements because their work often depends on contributions distributed across organizational boundaries. Suppliers, customers, and other partners can provide expertise that affects technical decisions and the eventual delivery of project outputs (Barbic, Jolink, Niesten, & Hidalgo, 2021).

Existing research has predominantly examined openness at the firm level. A project perspective adds analytical value because concrete exchanges with external actors are frequently organized around temporary development efforts, where knowledge must be selected and used under specific performance constraints (Bogers, Chesbrough, & Moedas, 2018; Bahemia & Roehrich, 2023).

Another limitation is the tendency to aggregate open innovation into a single phenomenon. Treating inbound and outbound activities separately makes it possible to assess whether acquiring external knowledge and applying internal knowledge beyond the firm have different consequences for project outcomes (Barbic et al., 2021; Bahemia & Roehrich, 2023; Lago, Marcon, Ribeiro, Uhr, Olteanu, & Fichter, 2024). Accordingly, this study asks how each flow relates to product development project performance.

The automotive industry provides an appropriate empirical context because innovation is embedded in a network of technologically specialized organizations. Automakers and suppliers repeatedly exchange technical and market information while developing products and components, making both directions of knowledge movement observable (Gonzalez, 2025; Rodrigues & Rua, 2025).

Openness, however, is not determined solely by the availability of partners. Organizational conditions can encourage or discourage the use of external knowledge. Behaviors associated with resistance to outside ideas or reluctance to share internally developed knowledge may undermine collaborative innovation even when interorganizational links exist (Engelsberger, Bartram, Cavanagh, Halvorsen, & Bogers, 2023; Ismail, Pintado, Marco, & Lazzarotti, 2023).

For this reason, leadership may be important in converting interorganizational exchanges into useful project activity. Leadership that emphasizes learning, collaboration, and the purposeful use of knowledge may influence whether team members engage constructively with both incoming and outgoing knowledge flows (Costa, Pádua, & Moreira, 2023; Naqshbandi & Tabche, 2018; Vu, Alonso, Bressan, Tran, & Nguyen, 2024).

Knowledge-oriented leadership (KOL) is particularly relevant because it promotes communication and knowledge-related interaction. By supporting cooperation inside the team and relationships with external actors, KOL may reduce barriers that prevent knowledge from being acquired, interpreted, shared, or applied (Engelsberger et al., 2023).

This study therefore also examines whether KOL changes the performance implications of inbound and outbound openness. The argument is that leaders can shape the environment in which project teams connect with external sources and transform knowledge exchanges into coordinated action (Avença, Domingues, & Carvalho, 2024; Lappalainen, Aleem, & Sandberg, 2023; Naqshbandi, Tabche, & Choudhary, 2019).

From a resource perspective, external relationships can supplement capabilities that are not fully available inside the focal organization. Access to outside technologies and expertise may expand the inputs available for innovation and support the reconfiguration of existing resources when environmental demands change (Barney, 1991; Sysodiya, Johnson, & Grégoire, 2013; Cassânego, Moralles, Nascimento, & Tortorella, 2025).

The benefits of external search also depend on its configuration. Organizations may engage with many different sources or develop intensive relationships with a smaller set of partners. For project teams, both the diversity of sources and the capacity to absorb what those sources provide can influence the usefulness of externally obtained knowledge (Shahzad Dan, Imran, Holtkamp, Niemi, & Meyer, 2024; Viera, Ubinati, & Lazzarotti, 2025).

Although openness can be considered strategically at the firm level, its implementation often occurs through projects. Organizational strategy establishes broad priorities, while project teams decide how knowledge will be sought, integrated, transferred, and used in day-to-day development work (Choi, Kim, & Roh, 2024; Zhang & Min, 2022).

Automotive development illustrates this multilevel process. Firms compete while simultaneously depending on technical coordination with other organizations. These relationships create opportunities for projects to receive external inputs and to direct internally developed solutions toward customers and partners (Gonzalez, 2024; Xu & Felzensztein, 2025).

Openness also involves a trade-off: knowledge sharing can create value but may expose strategically important assets. Firms must therefore balance collaboration with protection, and the appropriate balance may vary across stages of product development (Laursen & Salter, 2014; Bhatnagar, Dörfler, & MacBryde, 2023).

Project-level analysis is useful because projects differ from organizations as a whole in their temporary structure, defined objectives, formal constraints, and multidisciplinary composition. These characteristics affect how easily external knowledge can be incorporated into innovation activities (Bahemia & Roehrich, 2023; Gonzalez, 2023; Barbic et al., 2021; Dabic, Posinkovic, Vlaci, & Gonçalves, 2023).

Collaboration can supply a development project with expertise unavailable within the focal firm, but it also creates integration costs. Heterogeneous knowledge may be difficult to coordinate, particularly when participants lack the capacity to interpret and assimilate unfamiliar inputs (Markovic, Bagherzadeh, Vanhaverbeke, & Bogers, 2021; Shahzad et al., 2024; Tang, Fisher, & Qualls, 2021; Yang, Yang, & Li, 2023).

Inbound open innovation denotes purposeful efforts to bring externally located knowledge into innovation work. The process may begin with identifying relevant sources and continue through obtaining, interpreting, and incorporating useful inputs into the project (Chesbrough, 2017; Lago et al., 2024; Lappalainen et al., 2023).

Its potential contribution arises because outside technologies and expertise can complement a team's existing knowledge base. Yet access alone is insufficient; value is created when project members are able to assimilate and employ what has been obtained (Dabic et al., 2023; Naqshbandi & Jasimuddin, 2022).

Outbound open innovation concerns the deliberate deployment of internally developed knowledge beyond the firm's boundaries. This may include identifying external users, adapting technologies to partner needs, or commercializing knowledge in contexts outside the organization (Chesbrough, 2003; Lichtenthaler, 2009).

In development projects, outward knowledge deployment can be closely connected to the need to deliver solutions that meet market or customer requirements. Inventive capability supports the creation of knowledge for emerging needs, whereas desorptive capability concerns recognizing opportunities to transfer and exploit existing knowledge externally (Lee & Kim, 2019; Leitão, Pereira, & Brito, 2020; Shin, Kim, & Jeong, 2018).

These capabilities allow teams to respond to external demands either by generating new solutions or by adapting knowledge already accumulated by the firm. Consequently, outbound openness may link project expertise directly to identifiable customer and partner requirements (Leitão et al., 2020; Gonzalez, 2024; Lappalainen et al., 2023; Naqshbandi et al., 2019).

Project leadership must address both conventional delivery constraints and the knowledge processes required to solve development problems. The ability to meet scope, schedule, budget, and customer requirements can depend on how effectively a team acquires, exchanges, and applies knowledge (Moretti & Biancardi, 2020; Naqshbandi et al., 2019; Tang et al., 2021).

Research connecting leadership with innovation and knowledge management suggests that different leadership mechanisms can support complementary activities. Knowledge-oriented leadership integrates attention to knowledge creation, learning, sharing, and use rather than focusing exclusively on either change or control (Chaithanapat, Punnakitikashem, Oo, & Rakthin, 2022; Donate & Pablo, 2015; Donate, Mohíno, Appio, & Bernhard, 2022).

Here, KOL refers to leadership practices that recognize the value of employee knowledge, encourage learning and new ideas, and facilitate communication about knowledge-related issues. Such practices can support both the refinement of established expertise and the development of new capabilities (Donate & Pablo, 2015; Naqshbandi & Tabche, 2018).

Inbound activities may particularly benefit from these practices because external inputs must be evaluated and integrated into ongoing work. Leaders can support this process by creating channels for interaction, encouraging communication with partners, and facilitating the dissemination of useful knowledge inside the team (Dabic et al., 2023; Moretti & Biancardi, 2020).

Leadership may also affect outbound activities by helping teams interpret external requirements and decide how internal expertise should be deployed. Clarifying priorities, coordinating responsibilities, and maintaining attention to customer needs can improve the alignment between outward knowledge application and project objectives (Dabic et al., 2023; Gonzalez, 2024; Lichtenthaler, 2009).

Empirical evidence was obtained through a questionnaire directed at professionals involved in projects in Brazil's automotive industry. Participants occupied managerial, coordinating, or senior project-related positions. The sector was chosen because product development regularly involves interaction among organizations and the combination of knowledge generated inside and outside the focal firm. Responses were collected between July and October 2024.

Instrument quality was assessed before the main survey. Four academics experienced in innovation and leadership research reviewed the scales for clarity and appropriateness. A subsequent pretest involved 26 automotive project managers. Cronbach's alpha exceeded 0.70 for all constructs, providing initial evidence of internal consistency (Hair, Sarstedt, Mattews, & Ringle, 2016).

The sampling frame was derived from the SINDIPEÇAS directory, which listed 556 firms. We contacted 230 randomly selected organizations by telephone to introduce the study and identify suitable respondents and contact channels. Questionnaires and explanatory letters were then sent by email. The final dataset contained 137 usable responses, corresponding to a 59.56% response rate and exceeding the sample size commonly recommended for PLS-SEM applications (Hair et al., 2016). Table 1 presents the demographic data of the sample.

Possible nonresponse bias was examined by comparing early and late respondents. Independent-samples t-tests contrasted the first 30 and last 30 responding firms on KOL, inbound open innovation, outbound open innovation, and project performance. No statistically significant differences were detected (Armstrong & Overton, 1977).

We also assessed common method bias using Harman's single-factor procedure. Principal component analysis produced nine factors with eigenvalues above 1.0, and the largest factor explained only 13% of total variance. These results do not suggest dominance by a single common source (Podsakoff, Mackenzie, Lee, & Podsakoff, 2003).

Established measurement scales were selected according to their fit with the constructs in the model. Respondents indicated their level of agreement on seven-point scales ranging from strongly disagree (1) to strongly agree (7).

The questionnaire comprised five parts. The first collected descriptive information about firms, teams, and respondents. KOL was measured with six items adapted from Donate and Pablo (2015). Inbound and outbound open innovation were assessed using, respectively, the six-item scale of Sysodiya et al. (2013) and the four-item scale of Lichtenthaler (2009). Project performance was captured through nine adapted items addressing customer expectations and project objectives such as schedule, budget, scope, and quality (Carbonell & Escudero, 2013; Li & Huang, 2012).

The model also included team size, team tenure, and task interdependence as controls. Task interdependence was measured with the six-item scale proposed by Ali, Bahadur, Wang, Luqman, and Khan (2020). These controls account for differences associated with coordination complexity, accumulated team experience, and the degree to which members' tasks depend on one another (Gonzalez, 2024).

The hypotheses were tested with partial least squares structural equation modeling (PLS-SEM), allowing the measurement properties of the constructs and the structural paths among them to be assessed within the same analytical framework. The method is suitable for models that combine several latent relationships and operate with relatively modest samples (Hair et al., 2016).

Before evaluating structural paths, we conducted an additional assessment of common method variance. Harman's single-factor procedure yielded a first unrotated factor explaining 39.87% of the variance, below the commonly used 50% reference point (Ardura & Artola, 2020; Podsakoff et al., 2003).

A further diagnostic compared the confirmatory structure with the theoretical specification through partial correlation. The dependent variable result remained below 10%, providing no indication of a serious common-method problem (Chin, Thatcher, Wright, & Steel, 2013). Collinearity diagnostics were also acceptable: item VIF values ranged from 1.46 to 2.63 and did not exceed the 3.3 reference value (Kock, 2015).

Initial estimation identified three indicators with loadings below 0.70 (KOL4, I-OI3, and PP6), which were excluded from the final model (Hair et al., 2016). The retained indicators showed composite reliability, Cronbach's alpha, and rho_A values above 0.70, AVE values above 0.50, and statistically significant loadings exceeding 0.70 (Dijkstra & Henseler, 2015).

Construct distinctiveness was evaluated with both the Fornell–Larcker criterion and HTMT. For every construct, the square root of AVE exceeded its correlations with other constructs, and all HTMT values were below 0.85 (Henseler, Ringle, & Sarstedt, 2015).

Bootstrapping in SmartPLS 3 was used to estimate the structural paths. The model explained 67.8% of project-performance variance (R² = 0.678), indicating substantial explanatory power and exceeding the 0.20 reference discussed by Chin et al. (2013).

Predictive relevance was examined with Stone–Geisser's Q². The positive value obtained for project performance (Q² = 0.45) indicates predictive relevance for the endogenous construct.

According to Table 2, both direct hypotheses were supported. Inbound open innovation had a positive association with project performance (β = 0.197, p < 0.05), while the outbound relationship was substantially larger (β = 0.517, p < 0.001). Thus, H1 and H2 received empirical support.

The interaction results also supported the proposed moderation hypotheses. KOL strengthened the relationship involving inbound openness (β = 0.295, p < 0.01) and showed an even larger interaction with outbound openness (β = 0.480, p < 0.001). Neither control variable had a statistically significant effect on project performance.

The study contributes to project-level research on openness by separating two knowledge movements that are often combined in aggregate measures. Examining them independently shows that the effects associated with acquiring external knowledge need not mirror those associated with deploying internal knowledge outside the firm (Barbic et al., 2021; Bahemia & Roehrich, 2023; Scaringella & Scaringella, 2025).

The distinction between inbound and outbound flows is especially useful when the objective is project performance. Inbound activities may broaden the solution space available to a team, but they can also introduce interpretation and coordination costs. External information may use different technical languages, assumptions, or standards and therefore require time before it becomes useful. Outbound activities involve a different challenge: the team must identify how its accumulated knowledge can be translated into a solution that has value for a customer, partner, or other external recipient. Thus, the two directions involve different managerial tasks and may influence outcomes through different mechanisms.

The estimates indicate asymmetric relationships with project performance. Inbound openness was positive but modest, whereas outbound openness showed a considerably stronger association. This pattern adds nuance to prior project-level findings, which have not always reported uniform benefits from external knowledge exchanges (Dabic et al., 2023; Tang et al., 2021).

One explanation is that the two flows impose different demands on teams. Bringing in unfamiliar knowledge requires search, interpretation, assimilation, and coordination before it can influence project work. Outbound activity, in contrast, may begin with expertise already available to the team and direct it toward a concrete external use.

These integration demands can be particularly consequential in temporary projects. Teams operate under deadlines and changing membership, and resistance to unfamiliar ideas may further complicate the incorporation of outside knowledge. Consequently, the potential benefits of inbound openness may be partly offset by the effort required to make external inputs usable (Engelsberger et al., 2023; Tang et al., 2021).

The stronger outbound coefficient is consistent with the delivery orientation of product development. Applying internal expertise to a specified customer or partner need creates a relatively direct connection between knowledge activity and an observable project output (Barbic et al., 2021; Dabic et al., 2023; Gonzalez, 2024).

This reasoning is particularly relevant in automotive supply relationships. Product development frequently occurs through interdependent activities distributed among automakers, first-tier suppliers, and specialized technology providers. A project team may simultaneously receive specifications, technical feedback, and complementary expertise while also adapting its own knowledge to requirements defined elsewhere in the supply chain. The relative importance of each flow can vary according to the position occupied by the firm, the maturity of the technology, and the stage of development. Rather than assuming that more openness always produces better results, this study considers how distinct forms of openness become associated with performance in this organizational setting.

A project-level perspective also helps clarify why openness should not be treated simply as a firm characteristic. Organizations may endorse collaboration and still differ considerably in the way individual projects search for, absorb, or deploy external knowledge. Project teams face concrete decisions about timing, partner selection, information sharing, and the allocation of attention to external demands. Consequently, the practical consequences of openness depend on how these decisions are embedded in a particular development effort. Examining knowledge flows at this level therefore connects innovation strategy with the activities through which a product is conceived, developed, tested, and delivered.

The relative value of each flow may also vary across the development cycle. Early stages can require broader exploration, whereas later stages may emphasize protection and the application of accumulated knowledge. In automotive supply chains, however, the movement of solutions toward downstream customers remains central throughout project execution (Bhatnagar et al., 2023; Laursen & Salter, 2014).

Industry structure offers a further interpretation. Automotive suppliers typically concentrate on specific components or technological domains and therefore possess specialized knowledge bases. Their development challenges may often concern adapting and applying this expertise rather than repeatedly searching for entirely new knowledge outside the firm.

Automakers also occupy a coordinating position by specifying performance targets and development requirements. Suppliers consequently direct substantial project effort toward satisfying those requirements, creating strong incentives to deploy knowledge outward through technical solutions and commercialized components. This arrangement may help account for the larger outbound effect.

The moderation findings indicate that leadership influences the conditions under which knowledge exchanges become productive. KOL strengthened both relationships, suggesting that leaders can help teams evaluate external inputs, connect expertise with relevant partners, clarify customer requirements, and direct knowledge toward appropriate applications. The larger outbound interaction is consistent with the importance of customer alignment in automotive development.

From a theoretical standpoint, these observations reinforce the value of separating the direction of knowledge movement from the organizational conditions that enable its use. Open innovation research frequently emphasizes access to external resources, whereas project research highlights coordination under constraints. The present results bring these perspectives together by showing that knowledge flows and leadership jointly shape project-level outcomes. Future research could examine whether the relative effects of inbound and outbound activities change across project phases, technological uncertainty, or positions within supply networks. Longitudinal designs may also clarify whether leadership influences the acquisition of knowledge differently from its later integration and exploitation.

An additional implication concerns the temporal characteristics of projects. Unlike permanent organizational units, project teams operate within a limited period and often combine participants with different functional backgrounds. Knowledge exchange must therefore be coordinated quickly enough to influence decisions before major technical and managerial choices are finalized. Leadership that gives explicit attention to knowledge-related processes may reduce the risk that valuable information remains isolated within a particular individual or organization. At the same time, such leadership can establish priorities that prevent excessive openness from creating unnecessary communication burdens. KOL may consequently improve not only the volume of exchange, but also its selectivity and timing.

The moderation results further indicate that leadership should be understood as a contextual capability rather than simply an individual attribute of the project manager. Knowledge-oriented leaders can influence the conditions under which team members interpret external information, communicate across organizational boundaries, and decide whether existing knowledge should be retained, modified, or applied elsewhere. Their role is consequently relevant at several points in the knowledge-flow process. They can encourage the search for useful external inputs, create opportunities for discussion, reduce resistance to unfamiliar ideas, and help translate customer requirements into priorities that guide technical work.

By contrast, the stronger outbound relationship can be interpreted as evidence of the importance of problem–solution alignment. When project teams direct their knowledge toward a clearly identified external requirement, the connection between innovation activity and project outcomes may become more immediate. This does not mean that outbound innovation is intrinsically superior. Rather, in the empirical setting examined here, automotive projects are often organized around specifications and demands communicated by downstream actors. Under these conditions, the ability to transform internal expertise into externally useful solutions is closely connected to whether the project meets its intended objectives.

The findings also contribute to a more nuanced interpretation of openness in product development. A positive association between a knowledge flow and project performance does not imply that the flow is costless or universally beneficial. The weaker coefficient associated with inbound open innovation may reflect the time required to evaluate external inputs and reconcile them with knowledge already embedded in the team. In temporary project environments, these integration activities compete with schedule, budget, and scope pressures. The result therefore suggests that access to a broader knowledge base must be accompanied by mechanisms that help the team determine which information is relevant and how it can be incorporated without disrupting ongoing work.

Taken together, the results suggest a balanced managerial view of openness. Firms should avoid interpreting the weaker inbound coefficient as a reason to reduce external search indiscriminately. Instead, they should examine whether projects possess sufficient absorptive and coordination capacity to benefit from the knowledge they obtain. Similarly, the strong outbound effect should not be interpreted as a recommendation to disclose knowledge without considering appropriability concerns. Effective project management requires selecting which knowledge to acquire, which knowledge to retain, and which knowledge can be adapted or transferred to external actors. Leadership becomes important because these choices involve competing priorities and require collective judgment across organizational boundaries.

A further contribution of the project-level approach is that it directs attention to the boundary-spanning work performed by managers and team members. External relationships do not automatically become productive knowledge flows. Someone must identify relevant actors, interpret the information exchanged, communicate its implications, and coordinate subsequent actions. These activities can be distributed across the project team, but their effectiveness is likely to depend on organizational arrangements that support communication and accountability. Knowledge-oriented leadership provides one such arrangement by making knowledge processes visible within project management. This interpretation is consistent with the finding that leadership strengthens both relationships rather than replacing either inbound or outbound innovation as an independent source of performance.

The distinction identified in this study also has implications for how project performance is conceptualized. Performance in product development is not determined exclusively by the novelty of the knowledge available to a team. It also depends on whether that knowledge can be mobilized in a manner consistent with technical specifications, customer expectations, and project constraints. This perspective helps explain why knowledge acquisition and knowledge application may produce different empirical effects even when both are theoretically beneficial. The value of openness is therefore contingent on the fit between the direction of the knowledge flow and the immediate tasks confronting the project team. A team facing a poorly defined technological problem may benefit primarily from external search, whereas a team responding to clearly articulated requirements may derive greater value from adapting and deploying knowledge it already possesses.

These considerations also underline the importance of aligning openness with the specific decision context of each project.

For practitioners, the results suggest that managing development projects requires attention to the architecture of knowledge exchange as well as to conventional project controls. As firms rely more heavily on external relationships, managers must coordinate the boundaries through which knowledge enters and leaves the project.

Both directions deserve managerial attention, but they should not necessarily be managed identically. External inputs can broaden a team's knowledge base while also generating assimilation and coordination costs. Outbound activity can create a more immediate connection with customer needs when internal expertise is effectively directed toward external applications.

Project leaders should therefore complement responsibilities for scope, quality, time, and cost with practices that encourage communication, collaborative problem solving, and purposeful knowledge use. In particular, managers can help teams decide when exploration of new knowledge is necessary and when existing expertise should be adapted to meet partner or customer requirements.

The leadership implications are similarly concrete. Managers responsible for innovation projects can create regular opportunities for team members to discuss knowledge obtained from outside the project and to identify whether it should modify existing assumptions or practices. They can also establish clearer channels through which customer requirements and partner feedback reach technical specialists. Such practices do not require leaders to personally possess all relevant expertise. Instead, they require the ability to recognize where knowledge resides, encourage its circulation, and connect it with project decisions. This role may be particularly important in automotive supply networks, where technical specialization and interorganizational dependence make knowledge distributed across multiple actors.

For practitioners, the results suggest that managing open innovation in projects requires more than increasing the number of external relationships. Project managers should distinguish between activities intended to acquire unfamiliar knowledge and activities intended to apply existing knowledge to external needs, because each creates different coordination demands. Inbound collaboration may benefit from routines for screening information, assigning responsibility for interpretation, and documenting how external inputs affect technical decisions. Outbound collaboration may require closer interaction with customers and partners so that requirements are translated into actionable priorities. Treating both flows as a single undifferentiated collaboration strategy could obscure these differences.