An honest assessment of whether your organization can responsibly scale work to autonomous agents -- and the action items to close the gaps
Why Readiness
The hard part of scaling autonomy is rarely model intelligence -- it is permission and coordination. Delegated systems accelerate the consequences of whatever organizational ambiguity already exists. Delegation Readiness judges whether the conditions for delegation are actually in place, and if they are not, it names the gaps and the work to close them. It is allowed to say "not ready." A green assessment nobody can fail is worthless -- discomfort is information.
The framework is grounded in the Agentic Operating Model: make stopping cheap first, then widen authority.
How to Access
1
Open Organization → Overview
The Delegation Readiness panel sits at the top of the Organization Overview tab.
2
Add Organization Context
The assessment is only as sharp as its inputs. Add context on authority, coordination, tacit knowledge, exposure posture, and systemic blockers.
3
Generate the Assessment
Run the agent. It reads your context, departments, and portfolio and returns a verdict, per-dimension scores, and prioritized action items (~30s).
4
Act and Re-run
Work the action items -- promote the big ones into the portfolio -- then re-generate. Each run is versioned, so you can watch readiness improve over time.
What You Get
Verdict
Ready / Conditional / Not Ready
A board-readable headline. "Conditional" names the conditions that must be met before widening autonomy; "Not ready" flags a critical gap.
Scorecard
Five Dimension Scores
Each dimension scored 0-100 with a rationale and the evidence (or absence of evidence) from your context.
Action Items
Owned, Conditional Gaps
Prioritized by severity, each with a suggested owner role and a clear "becomes true when" condition that defines done.
Promote
Gap → Portfolio Project
Promote any action item into a backlog project (often process-change work), so closing an org gap is tracked like any other delivery.
History
Versioned Over Time
Every generation is a new version. Compare runs to show that the gaps you closed actually moved the verdict.
Honesty
Allowed to Say No
When context is too thin to judge, the agent says so and makes "capture the missing context" the action item rather than inventing facts.
The Five Dimensions
1
Authority & Decision Rights
Is it clear who can act, override, and pause? Is the stop button owned, and is intervention a designed, low-cost action rather than an act of social defiance?
2
Coordination & Dependencies
Are handoffs, upstream/downstream dependencies, and exception paths explicit? Will speeding one step flood a downstream team?
3
Tacit Knowledge & Human Repair
Does the org understand the "tolerated vagueness" people currently absorb, and account for the expertise being displaced -- the translation debt?
4
Exposure & Commitment Posture
Is risk appetite explicit? Does the org grasp the commitments and P&L changes that shipping autonomous agents creates?
5
Portfolio Coherence
Is the agent portfolio coordinated -- clear ownership, departments, prioritization -- rather than fragmented parallel experimentation?
How the Verdict Is Decided
R
Ready
Overall strong (roughly 75+) with no critical gaps in authority or coordination.
C
Conditional
Workable, but with named conditions that must be met before widening autonomy.
N
Not Ready
A critical gap exists -- especially in authority or coordination -- or context is too thin to judge.