Shared Leadership
Shared Leadership
What is shared leadership?
- Influence shared according to situation and expertise, not position
- Works only with clear authority boundaries, information sharing, and trust
- Spreading accountability too thin slows the organization down
In a team that adopts shared leadership, the first thing to change is not the org chart but the order in which people speak in meetings. Before, the team lead set the direction and everyone else filled in the gaps; now the person closest to the issue proposes first and the team lead wraps up last. The approval line does not disappear. The path an agenda item takes upward is reversed. So the first thing to adjust is not the grade structure but the meeting management rules that record who owns each agenda item and who speaks in what order. Once people start saying “this one is led by so-and-so” instead of citing job titles, you are halfway there.
It is easily confused with delegation, but the direction differs. Delegation is a vertical transfer, in which a team leader hands part of their authority to a specific person. Shared leadership is a horizontal process in which members influence one another. So this concept is treated not as a way to erase the formal leader, but as a complement that raises performance when it coexists with vertical leadership. Its scope also needs to be defined. It pays off less in repetitive work with fixed procedures than in planning, development, and new business, where interdependence is high and there is no single right answer. Team size tops out at single digits, where everyone can see each other's work. Layer it onto an organization of more than 20 and you get more voices but broken coordination.
The most common mistake in practice is reading shared leadership as decision by consensus. When “everyone is a leader” hardens into “nothing moves without unanimous agreement,” a decision that should take two days drags on for two weeks because of a single objection. In an incident, the cost is even higher. If a server outage or a field defect hits and no one has been designated to declare a rollback or recall, each side waits for the other to decide, and the recovery delay turns directly into losses. Put two modes in writing: distributed in normal times, single command in an emergency. And name the person with final accountability, so that when results are bad, it cannot be brushed off as “something we all decided together.”
Hwi KimCEO of HRDIST & Talent Development Systems Consultant
John KimJohn Kim, author of B2B Marketing Design and CEO of Nanum ConsultingAcademic discussion took off when Craig Pearce and Jay Conger, in their 2003 edited volume, conceptualized leadership not as an individual attribute but as a dynamic process of mutual influence within a group.
Consider a hypothetical case. Company A makes industrial sensors, with 140 employees and KRW 32 billion (≈ US$23,941,300) in annual revenue. It formed a seven-person task force for new product development, but every decision, from spec changes to parts orders, landed on one person, the head of development. Approval waits averaged 6 days, the launch date slipped twice, and before a third delay the company changed how the team operates.
- At the kickoff workshop, the head of development said firmly, "Manager Kim, finalize the technical specs without my sign-off," and named leads for four areas on the spot: circuits, firmware, mass production, and customer support.
- They set the amount a lead could decide alone at KRW 5 million (≈ US$3,740) per item, and left anything above that, along with launch-date changes, to the division head. The 11 items that caused disputes because they were ambiguous were put in a table and posted on the TF room wall.
- Senior Manager Park from the quality team raised the issue: "The development team can’t even see mass-production yield data, so how are they supposed to make judgments?" The yield and defect data that had been emailed twice a week was opened to the whole TF on a real-time dashboard.
- In a 40-minute coordination meeting every Thursday, they covered only items where areas collided. If an item still wasn't resolved, the division head decided it the same day and left a one-line note explaining the reasoning.
- They added a separate clause: for a field defect or line stoppage, sharing stops and the division head takes single command. When a connector defect hit in February, the clause was triggered and a recall decision was made within 8 hours.
- Four months later, approval wait time had dropped from an average of 6 days to 1.2 days and the weekly TF meeting had shrunk from 5 hours to 2. Mass-production approval came 3 weeks ahead of schedule, with no further delays.