1. Outcome
Can the result be evaluated without guessing?
- 0: The request is activity, not an outcome.
- 1: The outcome is implied but not measurable.
- 2: The result and evidence of done are explicit.
Resources
Know what to delegate and what to keep.
Score a task for clarity, repeatability, context, risk, reviewability, and value before delegating it to AI or a person.
01
Avoid delegating ambiguous work that will return as rework.
02
Make the approval boundary explicit before execution.
03
Find high-value recurring work that is ready to systemize.
Use it now
The 12-point delegation test
Score each question 0 (no), 1 (partly), or 2 (yes). Then fix the lowest-scoring dimension before you hand off the work.
2. Context
Does the delegate have the information needed?
- 0: Essential context lives only in my head.
- 1: Some sources are available; important history is missing.
- 2: Sources, constraints, audience, and prior decisions are available.
3. Repeatability
Can the work follow a stable method?
- 0: Every case is novel and judgment-heavy.
- 1: Parts repeat, but exceptions are common.
- 2: The steps and exceptions are understood.
4. Risk
Can mistakes be contained?
- 0: A mistake is irreversible or materially harmful.
- 1: Risk is manageable with a narrow scope.
- 2: Work is reversible or stays behind approval.
5. Reviewability
Is there a clear checkpoint?
- 0: Quality cannot be checked before consequences occur.
- 1: Review is possible but subjective.
- 2: A person can review sources, output, and actions before release.
6. Value
Is the handoff worth maintaining?
- 0: Rare task with little time or attention saved.
- 1: Useful occasionally.
- 2: Recurring work that protects meaningful attention.
Copy, print, or paste this structure into the tool where your work already lives.
Where it helps
Use the template when the work needs structure.
New recurring task
Score the task before turning an occasional request into an automation.
Delegation failure
Use the missing dimension, whether context, outcome, risk, or review, to diagnose why the handoff failed.
AI workflow review
Re-score the job when access, consequences, or the quality bar changes.
How it works
Make the handoff explicit.
Good delegated work has a clear input, a useful artifact, and a moment where a person can review what happens next.
Rate each dimension from 0 to 2
Use zero for unclear or unsafe, one for partial, and two for explicit and ready.
Improve the lowest dimensions
Add context, narrow scope, define evidence of done, or create an approval step.
Delegate, pilot, or keep
A high score supports delegation; a mixed score suggests a limited pilot; high-risk ambiguity stays human.
Useful outputs
Work you can inspect, share, and act on.
The output is something you can review, not a vague promise that AI saved time.
01Readiness score
02Missing-context list
03Approval boundary
04Pilot scope
05Definition of done
06Review cadence
Delegation scorecard / review point
Active workflowHuman decision
01PreparedReadiness scoreReady
02CheckedAvoid delegating ambiguous work that will return as rework.Visible
03BoundaryConsequential actionWith you
The boundary
Assistance does not remove accountability.
A score is a decision aid, not a guarantee. High-consequence work still needs professional judgment, appropriate permissions, and a person accountable for the result.
Read about security and trustFrequently asked questions
The practical questions, answered.
01What score is ready for delegation?
A score of 9 to 12 out of 12 is usually ready for a controlled pilot. A score of 6 to 8 needs a narrower scope or better context. Below 6, redesign the task before delegating.
02Should high-risk work ever be delegated?
Preparation may be delegated, but irreversible actions and consequential judgment should remain behind a qualified human review.
03Can I use this for human delegation?
Yes. The same dimensions improve delegation to colleagues, contractors, and assistants.
Get access
Give Bumblebee one real job to handle.
Start with an output you can recognize, review, and improve.