Who Owns What: Ownership Clarity for Distributed Teams (RACI and After)

Ownership problems present as dropped work and get diagnosed as communication problems. They are almost always a missing boundary: two capable people, each reasonably believing the other holds something, with nothing written that would settle it.

"Two people think the other one has it."

Two people, one gap

The failure is structural rather than personal, and it clusters at boundaries: between two teams, between two functions, between two countries. Inside a team, ownership is usually obvious from proximity. At a boundary there is no proximity, so the only thing that resolves it is a written statement, and boundaries are exactly where nobody has written one.

Distributed teams get a harder version. Even when ownership is clear, the owner may be asleep, so work stalls at the boundary anyway and the team cannot distinguish a routing failure from a time zone one.

What RACI does and does not do

RACI is genuinely useful for one thing: a structured conversation that forces a group to notice where they disagree about ownership. The exercise has value even when the artifact does not.

The artifact has three weaknesses. It is a matrix, which means finding your answer requires locating a row and a column, and people with a question do not do that. It describes involvement rather than decision rights, so "accountable" and "consulted" leave the actual question open. And it goes stale within a quarter, because people change roles faster than matrices get updated.

What people ask	What RACI tells them
"Who do I ask about this?"	Four names with four different letters
"Can I just decide this?"	Nothing specific
"Who signs off?"	Usually answered, if the matrix is current

After RACI: the one-line ownership list

Replace the matrix with a list. One line per area, in the channel topic or a pinned message, in this form: area, owner, what they can decide alone, and who to go to if they are away.

Three lines answer all three questions in the table above, in the order people ask them, in a place they will see without opening anything. And the decision limit is the part that prevents the most waiting, because it tells people when they do not need to ask at all.

Ownership that answers

A name only helps if the name is reachable. On a distributed team, "ask Ana" resolves the routing question and leaves the timing one, so work still waits a night.

Ownership written down becomes ownership that answers when the owner's record is answerable. With StandIn, Ana confirms a ninety-second brief at the end of her day, and while she is off, her StandIn answers questions from it in her words, with a source under every answer, clearly labelled, and never guesses. Someone in another zone asks about the payments work and gets Ana's answer during their own working day. If the question is a new decision rather than a retrieval, they are told so and pointed at Marek, who holds the authority for that.

The ownership list plus an answerable record covers both halves of the problem: who holds this, and how do I get an answer now. See how StandIn works.

Common Questions

Is RACI worth doing?

The exercise is worth doing once, because it surfaces disagreements about ownership that nobody had noticed. The matrix is not worth maintaining. Convert its conclusions into a one-line-per-area list and let the matrix go.

How do you make ownership clear on a distributed team?

One name per area, published where people already look, with a stated decision limit and a named fallback. The limit matters as much as the name, because it removes the cases where nobody needed to be asked.

What about work that genuinely spans two owners?

Name one owner anyway and say what the other is responsible for. Shared ownership of a decision means nobody makes it. Shared work with one decider is fine and common.

How often should the ownership list change?

Whenever someone joins, leaves, or changes area, which for most teams is every few weeks. That is a two-minute edit, which is the main argument for a list rather than a matrix.