I’ve spent a lot of time watching how decisions move through organizations.
I’ve seen it in large companies with three-year strategic plans and layers of matrixed leadership. I’ve also seen it in startups where everyone is moving quickly and the operating model is much less formal. The environments are different, but the organizations that work well have something in common: the work doesn’t stop because one person isn’t available to make the next decision.
That distinction has changed how I think about decision-making.
We tend to treat it as a leadership trait. We praise leaders who can make difficult decisions and criticize leaders who hesitate. But the more I look at how organizations actually operate, the less I think decision-making belongs to any one person.
Decision-making is an organizational capability.
The question isn’t whether a particular decision was right or wrong. We rarely have enough context to make that judgment from where we sit in the organization, and that isn’t the problem this essay is trying to solve.
The question is whether the organization has a structure that allows decisions to happen at the right level, with the right information, through the right people, so that work can continue.
An organization can communicate constantly and still be unable to move. There can be meetings, emails, calls, Slack messages, presentations, and recaps. Everyone can be talking about the issue and have access to the information, yet the same topic returns to the table the following week.
Sometimes the person with the authority to decide simply isn’t ready to make the call. That’s fine. The problem is when nobody knows that, because the people downstream are left guessing whether they should wait, escalate, or move forward. That uncertainty creates work of its own. Instead of advancing the work, people spend time trying to figure out where the decision stands and what they’re actually authorized to do.
Research into organizational decision-making has identified unclear decision rights, overlapping responsibilities, and unnecessary escalation as sources of organizational friction. McKinsey’s work on decision-making argues that organizations need clarity around who has the authority to make a decision and when that decision should be escalated.
When decision-making becomes a bottleneck, people adapt. Some find a way around it — they make the best decision they can, build the workaround, and keep moving. Sometimes that’s exactly what an empowered team needs to do. Sometimes it creates a decision that was never actually authorized. Others stop, because they don’t believe they have the authority to move, or because they don’t want to own a decision they weren’t authorized to make.
Neither response tells us much about the employee. One person has learned that the safest way to get work done is to make the decision and deal with the consequences later. Another has learned that the safest way to work is to wait. Both can exist in the same organization. When they do, productivity becomes uneven: one team keeps moving while another sits on a decision, and a project advances until it reaches the one issue nobody is authorized to resolve. The deadline moves, the conversation restarts, and the organization spends time managing the lack of a decision instead of doing the work.
If too many decisions have to travel upward, the organization creates a queue. That may be manageable when a company is small, but it gets harder as the organization grows. Large organizations don’t function because the CEO personally makes every consequential decision. They function because decision-making is distributed across the organization. Information moves upward so leadership can understand what’s happening. Strategy and direction move back down so the people doing the work understand where the organization is going. The decisions required to execute that strategy happen throughout the organization, and most of them should be resolved without requiring senior leadership intervention. There will always be decisions that need to move upward. The point is that not everything should.
Bain’s RAPID framework distinguishes between the people who recommend, agree, perform, provide input, and decide. The value of that distinction is that it separates participation from ownership — being involved in a decision doesn’t mean you have the authority to make it. That distinction sits at the center of decision architecture: where decisions belong, and the conditions people need to act within that boundary. That starts with trust, but trust alone isn’t enough. The person making the decision needs the authority to act, enough context to understand what they’re deciding, and the resources to carry it through.
If something falls within my scope, I should be able to resolve it. If it exceeds my scope, I should know where it goes next. If the next person can resolve it, the decision should stop there. If it genuinely requires executive attention, it should reach the executive with the information necessary to make the call. The goal isn’t to eliminate escalation — it’s to make escalation intentional.
Empowerment is more complicated than telling employees to “take ownership.” You can give someone responsibility for an outcome while withholding the authority required to influence it — that’s accountability without control. The same problem exists when authority is delegated but isn’t recognized by the rest of the organization: someone may have been told they can make the decision, but if everyone else keeps waiting for the CEO, the delegation hasn’t actually changed how the organization operates. McKinsey’s research on employee empowerment makes a similar point — giving people decision rights isn’t enough. They need the capability, tools, context, and guidance to exercise those rights effectively.
There’s another part of empowerment that gets discussed less often: the team has to trust the leader, too. If I’m trusted to make a decision, I need to know my leader has my back if the outcome isn’t what we expected. That doesn’t mean there are no consequences. It means a reasonable decision made within someone’s authority becomes a lesson when it doesn’t work, rather than an opportunity to find someone to blame. It also means giving the team credit when things do work — a leader who takes credit for everything reinforces dependency on the leader, while one who shares credit builds the organization’s capacity to operate without them.
Without that reciprocal trust, people protect themselves. They escalate decisions that should stay where they are, wait for permission, and spend more time figuring out what leadership wants than figuring out what the work requires. The organization gets slower without necessarily getting safer.
The leader’s job is to build an organization that can make decisions without needing the leader to personally make all of them. Information can roll upward without every decision rolling upward with it. Strategy can roll downward without every tactical decision needing to come back up for approval. People need to understand what they own, where their authority ends, and what happens when something exceeds it.
Documentation matters here, too. A decision needs to be visible enough that the organization doesn’t have to reconstruct the same conversation the following week. But documentation alone isn’t closure — people have to understand what was decided, who owns the next step, and what happens from there. That doesn’t mean every decision needs to be immediate. Some require more information, some require deliberation, and some carry consequences significant enough to warrant senior leadership. What matters isn’t speed for its own sake. It’s movement.
A well-designed organization will still have roadblocks, disagreements, and competing priorities. The difference is that one roadblock doesn’t stop the entire engine — there’s enough decision-making capacity distributed throughout the organization to keep other work moving. While one issue is being resolved, other teams can keep executing within their authority. When something genuinely requires escalation, it has a defined path to get there.
The measure isn’t whether every decision is fast or whether every decision produces the desired outcome. The measure is whether the organization has the capacity to keep moving while decisions are being made.
Organizations routinely ask employees for ownership, accountability, initiative, engagement, and productivity. Those expectations have to be supported by the operating system. You can’t ask people to take ownership while repeatedly preventing them from acting. You can’t hold someone accountable for an outcome without giving them the authority to influence it. And you can’t build a scalable organization around the assumption that one person will always be available to make the next decision.
What I keep coming back to is this: leadership isn’t about making every decision. It’s about creating the conditions in which people can make decisions at the appropriate level — with enough context to make them, enough authority to act on them, and enough support to own the outcome. Decision-making is an organizational capability, and the strength of that capability determines how much work an organization can actually move.