A red pen circles a sentence on a strategy document, then stops. Nobody argues about the sentence itself. They argue about who was allowed to write it. In many organizations, power still looks like a title on an office door, yet consequential decisions increasingly depend on how information moves, who gets consulted, who can veto, and who carries the consequences. That makes leadership less like sitting at the top of a pyramid and more like designing a machine whose parts must know when to move, when to wait, and when to challenge one another.
A decision designer understands a slightly uncomfortable truth: poor decisions are often designed before anyone makes them. A meeting can guarantee that the loudest executive speaks first. A reporting structure can ensure bad news travels upward after being polished beyond recognition. A committee can invite twelve opinions while giving nobody genuine authority. Research on organizational design shows that structures ranging from centralized hierarchies to decentralized systems can produce different levels of decision reliability, depending on how responsibilities and information are arranged.
Picture a product team preparing a launch. Marketing knows customers are losing interest, engineering knows the platform is unstable, and finance knows the budget cannot absorb another delay. Each group holds one piece of reality. The executive who insists on making the final decision may feel powerful, but the organization has already performed a strange trick: it has gathered distributed intelligence only to compress it into one person’s viewpoint. Research on distributed decision making emphasizes precisely this problem, because information and authority are often spread across people and locations rather than sitting neatly inside one office.
A different leader, Nadine, changes the room without changing anyone’s title. Before a major decision, she asks engineering to state its strongest objection, finance to identify the hidden cost, and customer teams to explain what headquarters cannot see. Then she assigns a single decision owner and makes the disagreement visible before commitment. The room becomes less comfortable. Good. Organizational research has found that leadership and organizational design can shape information exchange, discussion, dissent, and collective commitment, meaning power is influenced not only by who leads but by the structure through which leadership operates.
That is where decision design becomes more interesting than empowerment slogans. Decentralization without guardrails can produce chaos, while centralization can create bottlenecks, so leaders must decide which choices belong close to the customer, which require enterprise oversight, and which demand genuine collective judgment. MIT research on decentralized decision making highlights purpose, data, policies, and resource allocation as important guardrails for giving teams room to act without losing alignment. A leader’s job therefore becomes architectural: define boundaries, distribute authority, expose information, and make accountability impossible to misunderstand.
Power looks different when leaders stop treating every decision as proof of their importance. A chief executive who personally approves everything may appear indispensable, yet an organization dependent on one person’s judgment is carrying a hidden fragility. Stronger leaders build decision systems that allow expertise to travel without asking permission at every intersection. The most consequential leadership skill may therefore be invisible from the outside, because it lives in meeting design, decision rights, information flows, and carefully chosen moments of intervention. When leaders rewrite how decisions happen, they are not merely changing processes. They are redrawing the map of power itself.
