Operating Model and Transformation Leadership Artifacts
Delivery Health and RAID Radar
SIMULATEDVerified Jul 2, 2026A steering board should not only report where workstreams stand. It should show where they are heading. This artifact compares reported status against actual health and trajectory so the one that reads green but is sinking cannot hide.
Same instrument · three industries pick a use case to reconfigure the run
Prefer to read? The two minute case study · problem → approach → metric → outcome
Problem
Traditional RAG reporting can hide deterioration when teams report a comfortable snapshot. Technology strategy professionals need a trajectory view that prices in risks, issues, assumptions, dependencies, and trend direction.
Approach
The radar pairs reported status with actual health, trend, and RAID detail. It calculates a portfolio health index and highlights workstreams that appear green while declining.
Why this way
This connects delivery governance to transparency, escalation discipline, risk management, and executive decision quality.
The metric
Portfolio health index; reported vs actual gaps; the sinking green flags.
The trade-off
A comfortable reported status versus the honest trajectory underneath it.
Outcome
The workstream that reads green but is sinking, surfaced before the next steering.
Actual health, trend priced in
Actual amber or red
Green on paper, worse in reality
Trend window: last 3 weeks
Reads green, actually sinking · what the RAG snapshot hides
Health × trajectory
Dot color = reported status; position = actual health × trend. A green dot in the shaded corner is the trap.
Disputes automation
Reported green, but eval pass rate slipped 14 pts and agent adoption stalled at 41%.
Risks · 1
- highQuality slide reaches production before it's caught in review
Issues · 2
- highGolden set pass rate 77% vs 90% gate
- medAdoption stalled at 41% of agents
Dependencies · 1
- highLabeled data for new categories (Ops)
Assumptions · 1
- New dispute categories reuse the existing eval harness, proving false
Auto drafted leadership status
Status. Disputes automation is at risk (green as reported, amber on trajectory, deteriorating).
What changed. Golden set pass rate fell from 91% to 77% after the new dispute categories landed; adoption flat at 41%.
Watch. The status is reported green on delivery milestones, but quality and adoption are both trending down, the milestone view is hiding it.
Ask. Fund a 1 sprint eval hardening spike and pause the category expansion, decision needed at this steering.
The reported vs actual gap
If you act on this · the call → expected lift → how you'd measure it
The call
Escalate the workstream whose trajectory is worse than its reported status.
Expected lift · illustrative
Reduces delivery surprises by surfacing declining health before it becomes a missed milestone.
How you'd measure it
Reported vs actual gap, portfolio health index, issue aging, dependency closure, steering decision cycle time.
Steering committee takeaway: Green with a downward trend is not green. Report trajectory or prepare to be surprised.
Resume echo, the weekly reality of multi portfolio EM work at AMEX.
How this is built
Stack: Next.js (static) + the shared design system; client side state only.
Data: authored sample portfolios (finserv + telecom), each workstream carrying reported vs actual RAG, a 3-week trend, and RAID items, shared with EL-10.
Health index = mean of actual health scores (green 100 / amber 60 / red 25); the radar plots actual health × trend with reported status as dot color, so a green dot in the deteriorating but looks better corner is the trap.
Narrative: deterministic template over the workstream's authored brief, disciplined into Status → What changed → Watch → Ask. A LIVE variant can generate the prose via LIVE_MODEL without changing the structure.
Limitations: this is a modeled delivery governance artifact. Real use would require current plan data, workstream updates, RAID ownership, dependency status, and leadership review cadence.