Three Things Leaders Can Do to Build Stronger Networks on Their Teams

Most formal leaders try to solve team performance problems by leading more. More direction, more check-ins, more oversight. The research says the opposite works better: lead in ways that make it possible for others to lead alongside you.

We’ve known for a while that shared leadership predicts team effectiveness beyond what vertical leadership alone explains. What’s newer and more useful is the specificity we found in four meta-analyses covering over 6,000 teams. Edelmann and colleagues’ 2024 study of 31 organizational teams applied Necessary Condition Analysis and found that certain peer leadership behaviors aren’t just correlated with team effectiveness. They’re prerequisites for it. Without minimum levels of task, motivational, and critical innovation leadership from within the team, high performance becomes statistically unachievable. It doesn’t matter how strong the formal leader is.

We don’t need to debate whether peer leadership matters. We need to figure out what formal leaders can do to build it.

Empower rather than direct

The most replicated finding in this literature is also the simplest: empowering leadership by the formal leader is the strongest predictor of peer leadership emergence. Edelmann, Boen, and Fransen (2020) found that empowering behaviors predicted higher peer leadership quality across task, motivational, social, and external roles. Carson, Tesluk, and Marrone (2007) found the same pattern in 59 consulting teams: shared purpose, social support, and voice predicted shared leadership emergence, which in turn predicted client-rated performance.

Humility amplifies empowerment. Chiu, Owens, and Tesluk (2016) studied 62 professional work teams and found that leaders who admitted their own limitations and spotlighted others’ strengths created the conditions for team members to claim leadership for themselves. The effect was strongest when team members had genuine task competence. Humility without substance is just self-deprecation. Humility, paired with genuine redistribution of expertise, is a structural invitation.

And it feeds back. A longitudinal study of 721 team members across 169 teams found that shared leadership increased leaders’ trust in their teams, which in turn increased their empowering behavior, which in turn increased shared leadership (Hemshorn de Sanchez et al., 2025). Fransen and colleagues (2020) found the same reinforcing pattern in 64 teams: the higher the peer leadership quality, the more the formal leader was perceived as a better leader. Distributing leadership doesn’t dilute authority. It compounds it.

In practice, this means delegating decision authority on specific domains, soliciting input before deciding, asking questions rather than providing answers, and publicly acknowledging team members’ expertise. These are learnable behaviors, not personality traits.

Map the network and assign roles deliberately

Most formal leaders operate blind to the actual influence structure on their teams. Rob Cross’s organizational network research across 300+ organizations found that, without structured analysis, leaders identified only about half of their teams’ central connectors. We routinely over-rely on the most visible people and overlook the ones quietly holding the network together.

Fransen and colleagues developed Shared Leadership Mapping, a social network methodology where team members rate each other’s leadership quality across defined roles. The outputs reveal who leads, who is overlooked, and where gaps exist. This mapping feeds into the 5R Shared Leadership Program, which has been tested in multiple RCTs and shown consistent improvements in team identification and functioning (Mertens et al., 2020; Fransen et al., 2022).

Deliberate structure matters because simple decentralization doesn’t work on its own. Mehra, Smith, Dixon, and Robertson (2006) found that only “distributed-coordinated” leadership structures outperformed centralized ones. Fragmented distribution, where multiple people lead without coordination, performed no better than having a single leader. D’Innocenzo and colleagues’ 2016 meta-analysis confirmed it: network-based measures of shared leadership showed significantly higher effect sizes than simple aggregation measures. The structure of distribution matters as much as the fact of it.

Even a simplified version works. Ask team members to identify who they see as leading in task coordination, motivation, social connection, and external representation. Use the data to formally recognize peer leaders, clarify responsibilities, and fill gaps. Revisit the mapping quarterly. Edelmann and colleagues (2023) distilled four implementation principles: involve the formal leader throughout, use bottom-up selection, provide clear role definitions, and ensure each role is filled by multiple people.

Build equitable relational infrastructure

This is where the research connects to something we’ve been arguing in our relationship infrastructure work. Peer leadership can’t emerge without trust and psychological safety across the team. Drescher and colleagues (2014) tracked 142 groups over four months and found that trust both resulted from shared leadership and was the mechanism through which it drove performance. Trust is the substrate.

But trust doesn’t develop equally across a team unless the conditions are deliberately designed for equity. Wang, Jiang, Liu, and Ma (2017) found that leader-member exchange differentiation, the degree to which a formal leader has stronger relationships with some team members than others, negatively predicted shared leadership emergence. When leaders develop deep rapport with a few favorites while remaining transactional with everyone else, they create in-groups and out-groups that suppress peer leadership among those they don’t invest in.

We’ve argued before that relationship infrastructure is foundational, and its design determines whose relationships actually develop. The LMX differentiation finding is the empirical confirmation. Leaders who gravitate toward a favored few — the same lunch partners, the same people consulted before decisions, the same names on high-visibility assignments — are structurally suppressing peer leadership for everyone else. They’re building a seawall into their own rising tide.

In practice, this means distributing high-visibility assignments equitably, soliciting input from quieter members, and creating structured opportunities for team members to develop familiarity with each other’s expertise. Monitor whether relationship quality across the team is distributed or concentrated.

What’s still missing

Three gaps matter for implementation. Most studies are cross-sectional, so we don’t know how long it takes to build durable peer leadership networks. The literature barely addresses hybrid environments, despite Yang and colleagues’ (2022) finding that remote work weakens weak ties. And equity measurement is almost entirely absent. We don’t know whether these programs equalize influence or entrench existing patterns.

The model is clear, empowering, building deliberate structure, and investing in relationships equitably. The implementation challenge is translating that into daily behavior that persists under pressure. That’s always the gap between knowing and doing, and it’s the one that matters most.

Key sources: Edelmann et al. (2024), Behavioral Sciences; Wang et al. (2014), Journal of Applied Psychology; Carson et al. (2007), Academy of Management Journal; Chiu et al. (2016), Journal of Applied Psychology; Drescher et al. (2014), Journal of Applied Psychology; Mehra et al. (2006), The Leadership Quarterly; Wang et al. (2017), Human Performance; Hemshorn de Sanchez et al. (2025), Journal of Business and Psychology.

Contact us for a complete list of works cited.