05 · Executive Reporting & Governance¶
A status report written for a project team and one written for a steering committee are different documents, not the same document with fewer words. Executives need a decision-ready summary — what's the ask, what's the risk, what happens if we do nothing — while a team status report is appropriately full of the operational detail an executive should never have to wade through to find the one number that matters.
The exception report principle¶
The single biggest failure mode in executive reporting is symmetry: giving every project equal space regardless of whether it needs executive attention. An exception report inverts this — silence (or a single line) for projects on track, detail only for projects that need a decision.
| Project | Status | Executive attention needed? |
|---|---|---|
| Platform migration | Green, on track | No — one line: "On track, no action needed" |
| Vendor integration | Amber, vendor SLA breach | Yes — needs a decision on whether to invoke the penalty clause |
| Mobile relaunch | Green, ahead of schedule | No — one line |
| Compliance update | Red, blocked on legal sign-off | Yes — needs an escalation to unblock legal |
A one-page executive report from this portfolio devotes maybe 80% of its space to the two Amber/Red items and a single line each to the two Green ones — the opposite of a report that spends equal space on all four out of a sense of fairness to each PM.
The executive summary structure that works¶
| Section | Length | Content |
|---|---|---|
| Headline | 1 sentence | The single most important fact this period |
| Ask | 1–2 sentences | What decision or action is needed from this audience, specifically |
| Evidence | 2–4 bullets, quantified | The numbers that support the ask (CPI/SPI, cost, date) |
| Options (if a decision is needed) | 2–3 rows | What the executive is actually choosing between |
| What happens if no decision is made | 1 sentence | The cost of inaction, stated plainly |
Worked example: an executive ask, built from the structure¶
Headline: The vendor integration project will miss its go-live date by 5 weeks unless we act this week.
Ask: Approve invoking the contract's penalty clause and reallocating $40,000 to bring in a second integration team.
Evidence: CPI 0.82, SPI 0.79 (Red per standard thresholds); vendor has confirmed the delay is due to understaffing on their side, not scope ambiguity on ours; penalty clause entitles us to 1% of contract value per week late.
Options:
| Option | Cost | Effect |
|---|---|---|
| Do nothing, absorb the delay | $0 now | 5-week slip, no penalty recovered |
| Invoke penalty clause only | $0 (recovers ~$15,500, per contract terms) | Recovers cost, doesn't fix the delay |
| Invoke penalty + fund a second team | $40,000 | Recovers cost partially, cuts delay to ~2 weeks |
If no decision is made this week: the vendor's current staffing plan means the 5-week delay becomes locked in past the point a second team could still help.
This is what makes a report "decision-ready": an executive can read it in 90 seconds and know exactly what's being asked of them, compared to a narrative status update they'd have to interpret into a decision themselves.
Governance cadence by audience¶
| Audience | Cadence | Format | Focus |
|---|---|---|---|
| Project team | Daily/weekly | Detailed status, backlog, blockers | Operational |
| Program steering committee | Bi-weekly/monthly | Program rollup, cross-project dependencies | Tactical decisions |
| Executive committee | Monthly/quarterly | Exception report, portfolio health | Strategic decisions, capital allocation |
| Board | Quarterly | Portfolio-level trends only, major risks | Oversight, not operational decisions |
Reporting the same level of detail up every layer of this table is a common governance failure — a board does not need CPI/SPI numbers for individual projects, and giving it to them either buries the actual strategic question or invites the board into operational decisions that aren't theirs to make.
Governance escalation thresholds¶
| Trigger | Escalates to | Response time expected |
|---|---|---|
| CPI or SPI below 0.85 for 2+ consecutive periods | Steering committee | Within the next scheduled cycle |
| Projected budget overrun > 15% of total project budget | Executive committee | Within 1 week |
| Any safety, legal, or regulatory exposure | Executive committee + legal, immediately | Same day |
| Sponsor disengagement (module 07, Level 3 pattern) | PMO Director | Within 2 weeks |
Publishing these thresholds in advance — rather than deciding case by case whether something is "bad enough" to escalate — is what protects a PM from the two symmetric failure modes: escalating too much (noise, executives stop reading) and escalating too little (a real problem reaches the board only when it's already a crisis, per module 04).
How It Actually Works¶
The exception-report principle — report only what's off-track or newly decided, not full status — is an application of information theory's signal-to-noise concept: a channel (the executive's attention) has finite bandwidth, and every routine, on-track data point transmitted through it that requires no decision is pure noise consuming bandwidth that a genuine exception needs. Escalation thresholds function as a formal filter between the operational data layer (where every metric lives, updated continuously) and the governance layer (which should only see filtered signal) — the threshold's calibration matters exactly like Level 2 Module 9's RAG thresholds: set too sensitive, real signal drowns in false alarms and executives learn to ignore the channel entirely (the mechanism degrades back to full noise); set too insensitive, genuine problems pass through unfiltered until they're large enough to be obvious anyway, at which point the filtering added no value and the reporting channel failed at its one job.
Exercise¶
You manage the monthly executive rollup for a portfolio of 5 projects: 3 are Green, 1 is Amber (schedule slip, no cost impact, self-correcting per the PM), and 1 is Red (CPI 0.79, needs a $200,000 reserve draw to avoid a scope cut).
- Draft the one-page structure (headline/ask/evidence/options/cost-of- inaction) for the Red project only, inventing plausible specific numbers consistent with CPI 0.79.
- Write the one-line treatment each of the 3 Green projects and the 1 Amber project would get in an exception report, and justify why the Amber project doesn't need the full five-part structure.
- Using the escalation thresholds table, state which two thresholds (if any) this Red project has already crossed, and to which audience it should be escalated.