A Delegation Framework for Operations Leaders

Level one is execute: do exactly what's specified, no judgment calls. Level two is recommend: research the options and bring a recommendation, but wait for a decision before acting. Level three is decide and inform: make the call, then tell me what you decided and why. Level four is decide and act: full authority, no report-back required unless something goes wrong.

Name the level out loud when you hand off a task, not just in your own head. A task assumed to be level one by the person doing it and level three by the person who delegated it is a near-guaranteed source of friction, usually surfacing as either overcaution or an unwelcome surprise.

Pick one of these when you hand off a task:

• Level one, execute: do exactly what was specified in the task, with no judgment calls of their own.
• Level two, recommend: research the options and bring a recommendation, then wait for a decision before acting.
• Level three, decide and inform: make the call, then report what was decided and why.
• Level four, decide and act: full authority, with no report-back unless something goes wrong.

Trust in a person overall and trust in their judgment on a specific type of task are different things, and conflating them leads to over-delegating in unfamiliar territory or under-delegating in an area someone has already proven themselves. Someone you trust completely with customer communication might need to start at level one on a new financial process until they've demonstrated judgment there specifically.

Move people up a level deliberately, based on demonstrated judgment on that specific category of task, not automatically after a fixed tenure.

Handing someone a task without specifying the level defaults, in practice, to whatever level makes the delegator most comfortable, which is usually lower than what the person receiving it assumed. This produces the frustrating pattern where someone thought they had real authority, made a reasonable call, and then got quietly overridden or second-guessed, which teaches them to stop exercising judgment at all going forward, even on tasks where you genuinely wanted them to.

For any task delegated at level three or four, write down what the person is now authorized to decide, any dollar or scope limits on that authority, and when they should escalate anyway despite having the authority to decide. A tool like Trainual is a reasonable place to keep this documented so it survives a role change or a new hire stepping into the same responsibility, rather than living only in an informal understanding between two specific people.

Periodically check whether the levels you've assigned across the team still match reality: someone who's proven themselves consistently at level three on a task might be ready for level four, and a level that's been quietly ignored by everyone involved probably needs to be renegotiated rather than left as a rule nobody follows. Track delegated tasks and their assigned levels in a workspace like ClickUp so this review has something concrete to look at instead of relying on memory.

A bad outcome at level four doesn't automatically mean the level was set too high. Separate two questions before reacting: was the decision reasonable given what the person knew at the time, and was the outcome simply unlucky despite good judgment. Pulling authority back after every bad outcome, regardless of whether the judgment behind it was sound, teaches people to avoid decisions entirely, which defeats the purpose of delegating real authority in the first place. Reserve pulling a level back for a genuine pattern of poor judgment, not a single result you didn't want.