Templates › Post-implementation review report
Post-implementation review report
Template id: post-implementation-review-report
Responsible skill: value-realisation:post-implementation-review. Benefit evidence comes from the benefits-register kept through value-realisation:benefit-realisation-plan, and the estimated-versus-actual return from value-realisation:roi-model.
How to use this template: prepare sections 1 to 4 as an evidence pack and circulate it before the review meeting, so the meeting discusses causes rather than figures. Complete sections 5 to 8 in or after the meeting. Report at team, role or solution level only, never for named individuals. A benefit without adequate evidence is recorded as "not demonstrated", never as achieved. Mark anything not yet supplied as [To confirm: ...].
1. Review context
| Field | Value |
|---|---|
| Solution, rollout or programme reviewed | [To confirm] |
| Approved case reference and approval date | [To confirm] |
| Go-live date | [To confirm] |
| Review date and period covered | [To confirm] |
| Risk tier | [To confirm] |
| Review chair (role) | [To confirm] |
| Participants (roles) | [To confirm: benefit owners, solution owner, finance, CoE, user support] |
| Decision the review supports | [To confirm: continue, extend, change, retire, release funding, none] |
2. Case as approved
| Element | What was approved |
|---|---|
| Problem and outcome | [To confirm] |
| Benefits (register IDs) | [To confirm] |
| One-off cost | [To confirm] |
| Run cost per year | [To confirm] |
| Adoption level assumed | [To confirm] |
| Quality threshold at release | [To confirm] |
| Key risks and controls | [To confirm] |
| Key assumptions | [To confirm] |
3. Evidence and variance by element
Variance class: achieved, delayed, partial, not achieved, not demonstrated.
| Element | Planned | Actual (range) | Evidence and confidence | Variance class |
|---|---|---|---|---|
| Benefits (one row per benefit ID) | [To confirm] | |||
| One-off cost | [To confirm] | |||
| Run cost | [To confirm] | |||
| Adoption (team or role level) | [To confirm] | |||
| Live quality against threshold and guards | [To confirm] | |||
| Risk (incidents, near misses, control effectiveness) | [To confirm] | |||
| Timing of milestones | [To confirm] |
4. Estimated versus actual return
| Line | Estimated at approval (range) | Actual to date (range) | Main reason for difference |
|---|---|---|---|
| Cashable benefit | [To confirm] | ||
| Non-cashable value | [To confirm] | ||
| Cost avoidance | [To confirm] | ||
| Total cost | [To confirm] | ||
| Non-financial benefits (own units) | [To confirm] |
5. Causes of variance
For each material variance, record the cause the organisation could change, and whether it lay in the case (estimate) or in delivery and realisation.
| Element or benefit | Cause | Case or delivery | Evidence |
|---|---|---|---|
| [To confirm] |
6. Actions for remaining benefits
| Benefit ID | Action (re-profile, corrective action, retire) | Owner (role) | Due date | New measurement point |
|---|---|---|---|---|
| [To confirm] |
7. Lessons
| Lesson | Type (estimation or delivery) | Routed to (role or practice owner) | Date routed |
|---|---|---|---|
| [To confirm] |
8. Conclusion and decisions
| Field | Value |
|---|---|
| Overall finding | [To confirm: case delivered, partly delivered, not delivered, not yet demonstrable] |
| Decisions taken | [To confirm] |
| Referrals (scale-or-stop, retirement, responsible AI, security and privacy) | [To confirm] |
| Next review date, if any | [To confirm] |
| Summary approved for wider sharing by (role) | [To confirm] |
9. Caveats
This review compares outcomes with the approved case; it cannot prove that AI alone caused every change observed. Confounding changes in the period: [To confirm]. Measures based on self-report are perceptions and are shown alongside measures from the work itself.
Credit: Template from CopilotForge by Simon Foote (https://github.com/SimonFoote-greensparx/CopilotForge), licensed under CC BY 4.0. Keep this credit when you share or adapt the template; in your own completed document you may keep it as a footer or remove it.