Decision Authority Matrix

A decision authority matrix is a simple map of which decisions belong to which role, and who is allowed to make each call when the usual owner is off. It stops absence from blocking work by settling, in advance, the one thing that stalls a team the most: not knowing who is permitted to decide.

Here is the pattern worth naming. When someone is out and the work stops, it is rarely because the question was hard. It is because the answer needed a yes from a specific person, and that person was asleep, on a plane, or on leave. The decision was small. The permission to make it was missing. A matrix exists to remove that gap before it costs you a week.

What a decision authority matrix is

At its core, the matrix answers three questions for every recurring decision: who owns it, who can make the call in the owner's absence, and where the limit sits. That last one is the piece teams forget. A backup who can approve anything is not a backup; they are a second single point of failure with less context.

Think of a real case. Sarah owns the payments service and is out for a week. A vendor change comes in. Can the on-call engineer approve it? Up to what size? Past what point does it wait for Sarah or go to her manager? A matrix answers all three before the vendor ever writes in, so the on-call engineer is not left guessing at the edge of their authority.

Without this, teams default to one of two bad habits. They escalate everything, which buries a manager and slows every small call. Or they let people improvise, which works until an improvised call turns out to be one the owner would never have made. The matrix replaces both with a clear line drawn in calm conditions.

How to build one that holds up

Start from the decisions, not the people. List the calls your team actually makes in a normal month, then assign each one an owner, a stand-in, and a boundary. Keep it to the decisions that recur; you cannot map every possibility, and trying to will produce a document nobody reads.

• Owner: The role that normally makes this call. Tie it to the role, not the individual, so it survives people changing seats.
• Stand-in: Who decides when the owner is off. Name a person, not a committee, so there is no ambiguity in the moment.
• Boundary: The size or type of call the stand-in can make alone, and the point past which it waits or escalates.
• Reason: One line on why the boundary sits where it does. This is what lets a stand-in judge an edge case the matrix did not foresee.

The reason column is what separates a matrix that holds up from one that breaks the first time reality does not match the grid. A stand-in who knows why a limit exists can extend the logic to a new situation. One who only has a number is stuck the moment the situation is slightly different.

The absence column most teams skip

Plenty of teams have a version of decision rights baked into job titles. Far fewer have written down what happens when the title-holder is unreachable. That absence column is the whole point, and it is the one most often left blank.

The gap gets worse across time zones. If your owner is in Tokyo and the question arrives during the Amsterdam morning, "wait for the owner" means a full working day lost, every time. A matrix that names a local stand-in and a clear boundary turns that lost day into a five-minute call. We cover the underlying cost of these gaps in working across time zones on an engineering team, and the delegation side in keeping a record of decisions across a team.

The absence column is also what makes it safe for people to actually be off. If everyone knows who covers your calls and how far that cover extends, you can close your laptop without leaving a landmine behind. Without it, being unavailable quietly becomes a privilege reserved for people whose work nobody depends on.

Turning the matrix into an answer

Here is the turn. A matrix on a wiki tells a stand-in they are allowed to decide. It does not tell them what the owner would actually have said. Those are different things, and the gap between them is where good stand-ins still hesitate and slow down.

This is where the mechanism matters. A system that generates an answer will produce something confident whether or not it reflects the owner's real thinking. A system that answers only from what the owner wrote down will give back the owner's actual reasoning where it exists, and stay quiet where it does not. The second is the only one you can trust with a decision.

StandIn works from the record you build. When Sarah writes down how she thinks about vendor changes, her boundary, and the point where a call has to wait for her, StandIn gives those answers back to whoever asks while she is off, in her words. When a question falls outside what she wrote, StandIn does not invent a ruling. It says the record does not cover it and points to the person who should decide. That keeps the fast calls fast and the risky ones honest. See how quickly this moves a team in our look at async status tools.

Common Questions

What is a decision authority matrix?

It is a map of which decisions belong to which role, who can make each call when the owner is away, and the limit on that stand-in's authority. Its job is to keep work moving during an absence without forcing every small call up the chain.

How detailed should the matrix be?

Cover the decisions that recur in a normal month, and no more. A matrix that tries to name every edge case becomes a document nobody reads. Map the common calls clearly, and give stand-ins the reasoning so they can handle the rare ones.

Who should be the stand-in for a decision?

Name a specific person with enough context to judge, not a committee. Where you can, pick someone in a working time zone close to where the questions arrive, so the coverage is real and not just formal.

How does this help when someone is on leave?

The matrix says who decides in their place and how far that reach goes. Paired with a written record, a system like StandIn can also give back the owner's own reasoning on the calls they mapped, and say plainly when a question falls outside it.

Absence stops blocking work the moment the answer to "who decides this" is already written down. A matrix draws that line in calm conditions, so the fast calls stay fast and the hard ones reach the right person without a week of waiting. See where your decisions actually slow down at the decision velocity tool.