LEARN. PRACTICE. GROW.Compare measures with matching scope
A report shows two incidents in twenty changes before a response and three in one hundred afterward. Count increased, but observed rate fell from 10% to 3%. This supports investigating improvement; it proves neither causality nor permanence. Check period, severity, change mix, and denominator definition. A new control may reveal incidents previously unrecorded. For a recovery test, retain version, volume, configuration, result, and limitations. Twenty failure-free low-load rehearsals do not demonstrate zero risk during peak processing. Report available evidence together with what remains unmeasured.
Connect signals to actions and context
A queue forecast crosses an approved trigger, but the dashboard data is thirty minutes old. Verify freshness through an appropriate source and apply the plan within existing authority; neither ignore the alert nor declare certain outage from stale data. When volume doubles or the design changes, the old residual assessment may no longer represent the service. Keep the baseline for comparison and update the forecast with the identified change. In the register, connect risk, action, evidence, residual exposure, and the next review point. Closing the installation ticket proves task execution, not automatic effectiveness.
Report reserve and exposure together
In the fictional exercise, authorized reserve was 60 thousand, 35 thousand has been consumed, and the current forecast for remaining responses is 40 thousand. There is 25 thousand left and a 15-thousand gap to fund or address through another authorized decision. Consumption alone does not show whether the project is safe. Show forecast scope, committed responses, and excluded uncertainties. Fewer red risks may reflect justified closure or changed criteria; explain the movement. Management needs exposure and the requested decision; APS needs signals, limits, and execution. Both should use the same evidence.
Close by criterion and transfer with acceptance
An accepted delivery can close that delivery’s lateness risk while a separate certificate-renewal risk remains. Do not retain every risk routinely or delete those surviving the project. For APS handover, confirm an accepted owner, access, runbook, contacts, triggers, and next review date. Sending an email does not prove the team can execute the action. If access is missing one week before expiry, keep exposure open and escalate the gap before project closure. Lessons should change subsequent work: for example, demonstrate recovery using the operational account before acceptance rather than merely writing that communication should improve.
Workshop: prepare a five-line report with incident rate, sample comparability, remaining reserve, forecast gap, and requested decision. Add a verifiable criterion for APS to accept a renewal risk that persists after the project.
Common pitfalls
Comparing counts without denominators, claiming causality without evidence, erasing the baseline when updating a forecast, or closing exposure solely because the project ended.
Related topics: Priority, expected cost, and confidence · Choose and execute responses · Monitor, close, and hand over to operations
Demonstrate execution and effectiveness separately. Explain trends, gaps, and remaining exposure; close by criterion or transfer with documented acceptance and ability to act.
Reference: How risky is your project and what are you doing about it? · PMI-RMP five-domain ECO, updated-2024 public document