Boards keep betting the firm on a single hero, then acting surprised when succession turns into brinkmanship. Hybrid schedules spark culture fights, AI scrambles job design, and markets punish hesitation. In that environment, anchoring continuity to one leader is a fragile plan. The safer move is to make rotation a routine, not a rescue.
What the evidence actually says
Across thousands of real teams, shared leadership is not a feel-good slogan. It tracks with better results. A meta-analysis spanning 50 effect sizes and 3,198 teams found a positive relationship between shared leadership and team performance, with stronger effects in field settings than in classrooms or labs (D'Innocenzo, Mathieu and Kukenberger, 2016). How researchers measured leadership sharing mattered too. Social network approaches that capture how influence is distributed across a team showed larger effect sizes than simple aggregation of individual ratings, a cue that dense, distributed influence is more than a warm vibe. It is a structure you can design and see in the data (D'Innocenzo, Mathieu and Kukenberger, 2016).
A separate meta-analysis of 42 independent samples reported an overall correlation of rho = .34 between shared leadership and team effectiveness, with the relationship getting stronger as the work grew more complex (Wang, Waldman and Zhang, 2014). What teams share also matters. Distributing newer "genre" behaviors such as transformational leadership related more strongly to effectiveness than sharing traditional task- and relation-oriented behaviors, and shared leadership showed incremental effects beyond vertical leadership alone (Wang, Waldman and Zhang, 2014). In other words, in knowledge-heavy contexts where ambiguity is the work, distributing influence is not just compatible with performance. It strengthens it.
There is a boundary we should name. These are team-level findings, not CEO succession trials. Extending them to the top of the house is analogical, not causal. The direction of travel is clear, though. When influence is built as a network rather than a crown, performance tends to rise in the places that look most like modern companies: complex, interdependent, and fast-changing (D'Innocenzo, Mathieu and Kukenberger, 2016; Wang, Waldman and Zhang, 2014).
What that implies for the top job
If shared leadership works in teams, does it work in the corner office? The co-CEO record is thin but informative. A study of 87 companies with co-chief executives argues the arrangement is more viable than conventional wisdom allows, though it remains rare among roughly 2,200 firms in the S&P 1200 and Russell 1000 between 1996 and 2020, with fewer than 100 adopting it over that span (Feigen, Jenkins and Warendh, 2022). The authors do not sugarcoat the risks. They cite prominent failures, including Chipotle Mexican Grill, SAP, and Research In Motion/BlackBerry, particularly under stress (Feigen, Jenkins and Warendh, 2022). The core claim is pragmatic rather than ideological. Two leaders can run a firm when domains, decision rights, and rules of engagement are clear, and when the arrangement fits the context. That is not a blanket endorsement. It is a workable option with a design burden attached.
Taken together, the evidence points to a structured conclusion. Distributed power helps performance where the work is complex and interdependent, and shared top leadership can function, but only with explicit architecture and accountability. Treat the team-level results as a design principle and the co-CEO record as a caution light, not a green one.
How to rotate power without drama
The problem with succession is not the handoff. It is the concentration. A few design moves make rotation normal, not newsworthy.
-
Build influence as a network. The team meta-analysis found stronger performance links when shared leadership was measured using social network properties like density and decentralization (D'Innocenzo, Mathieu and Kukenberger, 2016). Translate that upstairs. Make lateral decision forums visible, publish who decides what, and audit how often critical choices rely on a single node. If a handful of names appear on every decision, you have a fragility, not a meritocracy.
-
Match rotation to complexity. Shared leadership’s benefits are larger when the work is complex (Wang, Waldman and Zhang, 2014). Prioritize rotating stewardship in knowledge-heavy portfolios such as product strategy, platforms, or data. In highly routine domains, use rotation to build bench strength, but do not expect the same performance lift.
-
Use time-bounded mandates. Treat senior roles as terms with explicit scopes and sunset dates. This is an interpretive design move drawn from the team-level evidence that distributed influence correlates with outcomes, not a tested CEO recipe. The point is to make rotation a scheduled event with continuity practices, not a scramble after a resignation.
-
Consider co-leadership as a transitional tool, not a doctrine. The field record shows co-CEOs are rare, sometimes fail, yet can work when carefully structured (Feigen, Jenkins and Warendh, 2022). If you use a co-CEO period to institutionalize a handoff, set a horizon, clarify domains, and define who breaks ties. Publish those rules to the board and the top team.
-
Measure outcomes, not personalities. Both meta-analyses highlight that what is shared and how it is enacted shape results (D'Innocenzo, Mathieu and Kukenberger, 2016; Wang, Waldman and Zhang, 2014). Carry that discipline into succession. Track a small set of outcome metrics for each rotating steward, keep the review cadence predictable, and separate role health from personal charisma. Rotation works only if accountability survives the handoff.
-
Create a visible queue. Succession fear thrives in opacity. Make the pool of future stewards explicit and rotate them through decision roles before they take the seat. This is an interpretive extension of the shared leadership logic: people make better decisions together when more of them have practiced deciding.
The practical read
The 2020s will not wait for perfect successors. The team-level evidence is robust that spreading leadership across a network correlates with better performance in real organizations, especially when the work is complex (D'Innocenzo, Mathieu and Kukenberger, 2016; Wang, Waldman and Zhang, 2014). The record at the very top is thinner and mixed, but it shows shared leadership can be made to function when the design is explicit and the context is right (Feigen, Jenkins and Warendh, 2022). Treat that as your brief.
If your plan is still to name a hero and hope, you are running a single point of failure. Make rotation a muscle. Build the network that decisions flow through. Use co-leadership as a tool with guardrails, not a creed. Then when the call comes to name a successor, the answer is already in motion. Succession becomes routine. The business keeps moving.
Sources
- D'Innocenzo, Mathieu and Kukenberger (2016). A Meta-Analysis of Different Forms of Shared Leadership–Team Performance Relations. Finds a positive shared-leadership to performance link across 3,198 teams, stronger in field settings and when leadership is distributed in dense networks. https://journals.sagepub.com/doi/10.1177/0149206314525205
- Wang, Waldman and Zhang (2014). A meta-analysis of shared leadership and team effectiveness. Reports an overall rho = .34 between shared leadership and team effectiveness, stronger in complex work and with incremental effects beyond vertical leadership. https://pubmed.ncbi.nlm.nih.gov/24188392/
- Feigen, Jenkins and Warendh (2022). Is It Time to Consider Co-CEOs? Argues co-CEOs are viable based on a study of 87 firms, notes rarity among roughly 2,200 companies and acknowledges notable failures, implying the model needs careful design. https://hbr.org/2022/07/is-it-time-to-consider-co-ceos