Decision rights — the formal specification of who can decide what, under what conditions, with what level of consultation or approval — are among the most powerful governance tools in an organization''s architecture. They are also among the least explicitly designed.
Most organizations have implicit decision rights. They emerged from personalities, from history, from convenience, from who happened to be in the room when a particular type of decision was first made. Over time these implicit rights solidified into patterns. Senior leaders started being consulted on things because it became the norm, not because the norm was ever evaluated as correct. Authority for categories of decision consolidated at certain levels because that is where it had always been, not because that was where it should be.
The consequences of implicit and poorly designed decision rights are consistent and recognizable: bottlenecks at senior levels where more decisions accumulate than can be processed efficiently, confusion at operational levels about what each person is actually authorized to do, and the constant escalation of decisions that could and should be resolved lower in the organization. Organizational speed is reduced. The people doing operational work develop learned helplessness about decision-making — they refer up because that is what the pattern rewards, not because the decision actually requires senior involvement.
I have worked inside and alongside organizations where this pattern was severe enough that the speed difference between senior-reviewed decisions and delegated ones was measured in weeks, not days. The people waiting for decisions were not idle; they were doing other work and context-switching when approvals arrived. The senior leaders reviewing decisions were not adding value proportional to the cost of the bottleneck. The pattern had simply never been designed, so it had never been designed well.
The RACI Framework and Its Limitations
RACI — Responsible, Accountable, Consulted, Informed — is the most commonly used framework for specifying decision rights. It has genuine utility. It forces organizations to name who owns each role in a decision process, and it distinguishes between the person doing the work (Responsible), the person who owns the outcome (Accountable), the people whose input is required (Consulted), and the people who need to know the outcome (Informed).
The limitations of RACI matter in practice. First, RACI conflates accountability with authority: the person who is Accountable for an outcome may or may not have the authority to commit resources to achieve it, and RACI does not distinguish between these two. Second, RACI treats Consulted as a uniform category, but there is a significant difference between consultation that can change the decision and consultation that is advisory. Third, RACI does not address the conditions under which authority can be exercised without consultation — which is usually the critical question for operational speed.
These limitations do not make RACI useless. They mean that RACI is an incomplete specification of decision rights, and organizations that treat it as complete will still experience the bottlenecks and confusion that decision rights frameworks are supposed to eliminate.
The Five-Dimension Decision Rights Framework
A more complete specification of decision rights requires five dimensions, each of which answers a distinct question about how a category of decision should work.
Dimension 1: Authority to decide. Who is authorized to make this decision and bring it to closure? This is the most basic dimension. The authority holder is the person whose determination is binding — not the person who implements the decision, not the person who approves it, but the person who decides. In well-designed decision rights, the authority holder is the person at the lowest level in the organization who has the judgment, information, and accountability to make the decision well. The impulse to place authority higher than necessary — because senior leaders have more experience, or because the organization defaults to seniority — is one of the primary generators of bottlenecks.
Dimension 2: Authority to commit resources. Who is authorized to commit the financial and operational resources required to implement this decision? This is frequently distinct from the authority to decide. A department manager may have the authority to decide how a project should be structured but not the authority to commit budget to it. Conflating these two creates confusion: the person with decision authority is told they own the decision, then discovers they cannot act on it without seeking separate approval for the resources it requires. Explicit resource commitment thresholds, tied to specific roles, are necessary to make decision authority functional.
Dimension 3: Authority to represent externally. Who is authorized to communicate this category of decision, and the commitments it creates, to parties outside the organization — partners, customers, regulators, the public? This dimension matters most for decisions that create external obligations. In many organizations, it is unstated and defaults to seniority, which means that operational-level people who make decisions routinely are not authorized to communicate them externally, creating an unnecessary layer of communication latency and a source of confusion about who speaks for the organization on what.
Dimension 4: Obligation to consult. Who must be consulted before the decision is made, and what does consultation mean? The key distinction here is between consultation that can change the decision — genuine input that the authority holder is obligated to consider — and notification that is called consultation because it feels less unilateral. The second is not consultation; it is notification before the fact. Calling it consultation creates the expectation of influence that is then disappointed, which is worse for organizational relationships than honest notification would be. Explicit specification of who must be consulted, by when in the decision process, and what the obligation to consider their input actually means, eliminates this source of friction.
Dimension 5: Obligation to inform. Who must be notified of the decision, and by when? Notification is distinct from consultation — the decision is already made, and the obligation is to communicate it to people who need to know it. This dimension is often underspecified, which produces the situation where people who should have known about a decision discover it late, have to adjust their work, and experience the organization as poorly coordinated. Explicit specification of who has a right to timely notification, and the timeframe within which notification is required, addresses this.
Mapping Current Decision Rights
Before redesigning decision rights, it is necessary to understand what the current implicit rights actually are. The mapping process has three steps.
Identify the decision categories. Create a list of the significant decision categories that recur in the organization — not every individual decision, but the categories of decisions. Procurement above and below thresholds. Personnel decisions (hiring, performance management, termination). Product or service scope changes. Customer commitments beyond standard terms. Budget reallocations. Policy exceptions. For each organization this list will be somewhat different, and the process of creating it surfaces the first set of insights: decision categories that have been handled inconsistently because they were never explicitly categorized.
Map the current practice. For each decision category, describe how decisions actually get made — not how they are supposed to get made, but the actual practice. Who initiates? Who is consulted? Who decides? Who can commit resources? How long does it typically take? Where does it stall? This mapping is done through observation and interviews, not through reading policy documents, because the policy documents typically describe the intended process rather than the actual one.
Identify the misalignments. Compare the current practice to what the decision rights should be given the strategic and operational needs of the organization. The misalignments fall into predictable categories: decisions being made at levels higher than necessary (bottlenecks), decisions being made without required consultation (surprises and coordination failures), decisions being made without clear authority at the operational level (confusion and escalation), and resource commitment authority mismatched with decision authority (decisions that cannot be acted on).
Redesigning for Throughput Without Accountability Gaps
The design objective is maximum decision velocity at the lowest level of the organization that can make the decision with appropriate quality and accountability. This sounds simple; it is organizationally difficult because it requires senior leaders to actively distribute authority that they currently hold.
The redistribution is not primarily about trust. It is about the structure of the decision. Decisions that require integration across multiple senior functions belong at senior levels. Decisions that can be made with the information and authority available at operational levels should be made there. The question is not whether to trust operational-level people with the decision; it is whether the decision, properly specified, can be made well at the operational level.
The practical instrument for the redesign is the decision rights map: a structured document that specifies, for each decision category, the five dimensions. The map is organized by decision category, not by role, because the purpose is to specify how each type of decision works, not to describe each person''s responsibilities.
The design principles that prevent accountability gaps are specific:
Authority and accountability must be linked. The person with the authority to decide must also carry the accountability for the outcome. Separating these — giving one person the authority and holding another person accountable — produces the worst of both worlds: the authority holder has no stake in quality outcomes, and the accountable holder has no leverage to produce them.
Resource commitment authority must be sufficient for the decision. If a decision is delegated to an operational level, the resource commitment authority at that level must cover what is needed to implement the decision. A delegation that stops at the authority to decide without covering the authority to commit resources is an incomplete delegation.
Escalation paths must be explicit. Not every decision fits neatly into pre-specified categories. The redesign should specify what happens when a decision is at the boundary of a category, or when a decision at a delegated level has unexpected complexity. The escalation path — who to consult, who to escalate to, under what conditions — prevents ambiguous situations from defaulting back to centralized decision-making by habit.
The Political Dynamics of Decision Rights Redesign
Decision rights redesign is a governance problem with political dimensions that are real and worth acknowledging.
Authority is not neutral. Senior leaders who currently hold authority for categories of decision — formally or by default — have a stake in that authority. Some of that stake is legitimate: the decisions that consolidate at senior levels often do so because experience and organizational visibility genuinely improve decision quality. Some of it is not: authority is a form of organizational status, and the redistribution of authority requires the people who currently hold it to accept a change in their role that they may not welcome.
The political dynamics do not make redesign impossible. They make it a leadership challenge rather than a purely analytical one. The leaders who successfully redistribute decision authority typically do so by being explicit about the logic — this decision is better made at this level because of these specific reasons — rather than by framing it as an efficiency measure. The efficiency framing is accurate, but it does not address the stake that senior leaders have in the current distribution.
The redesign also creates new obligations for the people receiving authority. Delegation without support is not effective delegation. People who are receiving decision authority for the first time in a category need clarity about what the authority actually entails, what the boundaries are, what the escalation path is, and what quality looks like. Organizations that distribute authority without providing this infrastructure often find that the authority consolidates back upward within months because the people who received it were not set up to exercise it well.
After the Redesign: Making It Operational
A decision rights redesign that produces a document is not yet a decision rights redesign. The document specifies the target state. Making it operational requires several additional steps.
Communication. The redesigned rights need to be communicated to everyone they affect. This means not just distributing a document but explaining why the changes were made, what each person''s new authorities and obligations are, and what the escalation paths are for situations that do not fit clearly.
Practice. Decision rights change through practice, not through documentation. This means creating opportunities for people to exercise newly delegated authority, and resisting the organizational pull toward escalation that will exist in the early period after the redesign. Senior leaders who receive escalations that should have been handled at delegated levels need to redirect them rather than decide them.
Monitoring. Track decision velocity and escalation rates in the periods following the redesign. If escalation rates are not declining for the categories where authority was redistributed, the redistribution is not taking hold. Investigate why — whether it is lack of clarity about the new rights, lack of confidence in the people exercising them, or cultural resistance to the change — and address the root cause.
Revision. Decision rights need maintenance. As the organization grows, as the environment changes, as new decision categories emerge, the map needs to be updated. A redesign that produces a static document will gradually become as disconnected from actual practice as the implicit rights it replaced.
Conclusion
Decision rights are organizational architecture. How authority is distributed determines how fast decisions get made, whether the people with the most relevant information are making the decisions or merely advising them, and whether the organization can execute at the speed its environment requires.
Most organizations have this architecture by default rather than by design. The default produces predictable failure modes: senior bottlenecks, operational confusion, and the slow erosion of the organizational capacity to act quickly.
The alternative is explicit design: mapping the current state, identifying the misalignments, rebuilding the rights against the five-dimension framework, and making the new rights operational through practice and maintenance. This is governance work. It is slower and more complex than solving a process problem or deploying a tool. It is also more durable, because it addresses the underlying structure rather than the symptoms.
Organizations that invest in explicit decision rights design consistently find that it returns more throughput per unit of effort than most of the operational interventions they apply to the same problems. The decisions that were taking weeks take days. The escalations that were consuming senior bandwidth get resolved at the level where they belong. The people doing operational work develop the confidence to act without constant upward reference.
These are not small outcomes. They compound over time.