You have read the article. You have run the workshop. Maybe you even printed the matrix and pinned it to the wall. And yet the same decision keeps coming back to your desk — re-opened, re-argued, quietly stuck. If that sounds familiar, you are not alone, and you are not doing it wrong. You picked a tool to fix a problem that the tool was never built to find.
RAPID, RACI, and DARE are genuinely useful frameworks. I reach for them in real engagements. But they all start from the same hidden assumption: that you already know which decision is broken and simply need to assign the roles around it. In most leadership teams I work with, that assumption is the actual problem. The decisions that hurt you are the ones nobody has named yet.
All three frameworks answer one question: on a given decision, who does what? They take a decision you have already identified as important and distribute the roles around it — who recommends, who advises, who can veto, who is finally accountable for the call. That is a real and valuable job. When a single high-stakes decision is being fought over in the room, a clean role map ends the fight.
Put simply, these are role-assignment tools. They are excellent at converting a messy meeting into a structured one. They make explicit what was implicit: that the head of product recommends, the CFO has a veto on spend, and the CEO decides. Ambiguity drops, and the decision moves.
But notice what they all take as given. They assume the decision is already on the table, already framed, and already understood to be the bottleneck. They start at step two. The framework hands you a clean process for a decision you have correctly diagnosed as broken — and stays silent on whether you diagnosed the right one.
RAPID, popularized in Bain's work and Rogers and Blenko's 2006 Harvard Business Review article "Who Has the D?", assigns five roles — Recommend, Agree, Perform, Input, Decide — with one accountable Decider and an explicit Agree role that carries veto or sign-off rights. Its real innovation is making the veto visible: the people who can block a decision are named before, not after, the call is made.
RACI is older and broader. Responsible, Accountable, Consulted, Informed is a responsibility-assignment matrix built for tasks, projects, and deliverables. It tells you who executes a piece of work and who is answerable for it. It was never designed to allocate decision authority, which is why teams who use it for judgment calls often find it strangely silent on the moment of actually deciding.
DARE, developed by McKinsey, sorts people into Deciders, Advisers, Recommenders, and Execution stakeholders — and reserves the vote for the Deciders. Its sharp edge is separating advice from authority: many people may legitimately have input, but input is not a vote. DARE exists precisely for the organization where too many voices believe they hold a decision right they do not actually have.
Here is the pattern I see across diagnostic work. A team adopts RACI, or RAPID, or DARE, runs it on the three or four decisions everyone already argues about, and feels relief for a quarter. Then the symptom migrates. A different decision starts re-opening. Escalations pile onto the same two executives. The framework was applied perfectly — and the system kept leaking somewhere else.
That happens because role clarity on one decision does not tell you anything about the load, the coupling, or the escalation behavior of the whole decision system. Leadership Architecture is the structural configuration of decision authority, role accountability, escalation pathways, and governance mechanisms through which leadership teams coordinate decisions and translate them into consistent organizational execution. A role map touches the first item on that list, on one decision, on one day.
The deeper question is fit: does your architecture's capability match the coordination demand your organization is actually placing on it? You can have flawless RAPID charts and still be drowning, because the volume and interdependence of decisions have outgrown the structure. Put simply: the frameworks tell you who decides; they do not tell you whether your decision system can carry the weight you are putting on it.
Each of these is a legitimate tool with a specific job. The trick is matching the tool to the situation — and knowing what none of them measure. Here is an accurate one-line read on each, and where it earns its place.
Popularized in the 2006 HBR article "Who Has the D?" Its strength is making veto rights explicit before the decision is made, so blockers cannot ambush the call after the fact. Best for high-stakes, contested, cross-functional decisions where the fight is about who gets the final word.
The familiar project-management matrix. It excels at clarifying execution: who does the work, who is answerable, who must be consulted or merely told. It was never built to allocate the authority to decide, so it tends to go quiet at the exact moment of judgment. Best for process and delivery clarity, not for resolving who holds a decision right.
From McKinsey's work on decision effectiveness ("The limits of RACI — and a better way to make decisions" and the 2022 piece "If we're all so busy, why isn't anything getting done?"). Its edge is cleanly separating advice from authority. Best when too many people believe they have a vote and decisions stall under the weight of false consensus.
Picture a leadership team deciding whether to enter a new market — a high-stakes, cross-functional call where finance, product, and operations all have a real stake, and one of them quietly believes it can block the move. RAPID is built for exactly this. Naming a single Decider ends the diffusion of accountability, and surfacing the Agree role forces the veto into the open before the decision rather than after it. When the problem is a contested, consequential, one-off decision, RAPID is the cleanest tool in the set.
Picture a 200-person manufacturing company rolling out a new quarterly planning process, where the recurring failure is not judgment but follow-through — tasks fall between roles and nobody is sure who owns the handoff. RACI shines here. Mapping Responsible and Accountable across each step removes the "I thought you had it" gaps in execution. Reach for RACI when the pain is operational ownership of work and deliverables, not the authority to make a strategic call.
Picture an executive team where every decision balloons into a ten-person debate because everyone assumes their input equals a vote. The meetings are long, the decisions are slow, and no one feels they can move without universal sign-off. DARE is the antidote. By reserving the vote for named Deciders and recasting everyone else as Advisers or Recommenders, it lets people be heard without letting input masquerade as authority. Use DARE when false consensus is the bottleneck.
Before you reach for any of these frameworks, do the one thing none of them do: look at the whole decision system, not a single decision. Here is what that looks like in practice this week.
No. DARE is McKinsey's decision framework, introduced in their work on decision effectiveness — notably "The limits of RACI — and a better way to make decisions" and the 2022 article "If we're all so busy, why isn't anything getting done?" It assigns Deciders, Advisers, Recommenders, and Execution stakeholders, reserving the actual vote for the Deciders.
You can, but it was not built for it. RACI is a responsibility-assignment matrix for tasks and deliverables — who does the work and who is answerable. It clarifies execution beautifully but tends to go silent at the moment of judgment. For genuine decision authority on contested calls, RAPID or DARE are better fitted to the job.
It depends on the failure you are seeing. RAPID fits high-stakes, contested single decisions. RACI fits task and process ownership. DARE fits teams where too many people assume their input is a vote. None of them tells you which decisions are actually broken — that requires measuring the decision system itself before you map roles.
RAPID, RACI, and DARE assign roles on one decision you have already flagged. Leadership Architecture measures the whole system first — your Leadership Architecture Index against your Decision Coordination Demand — so you learn which decisions are failing, why, and in what order to fix them. It sizes the problem; the frameworks then help you solve it.
Standardizing reduces confusion, but forcing one tool onto every situation creates its own friction. The healthier move is to diagnose first, then apply the framework that fits each class of decision — RAPID for contested calls, RACI for execution clarity, DARE for false-consensus stalls. Let the diagnosis, not the template, drive the choice.
Sources: Rogers, P. & Blenko, M. (2006). “Who Has the D? How Clear Decision Roles Enhance Organizational Performance.” Harvard Business Review. · McKinsey (2018). “The limits of RACI — and a better way to make decisions.”
Before you map roles, measure the system. The CEO Fit Diagnostic returns your Decision Coordination Demand, an Architecture Capability Estimate, and a Fit Score in minutes — so you know which decisions to fix first, and which framework fits. Take the diagnostic and start there.