Project Closure & Lessons Learned
Key Takeaways
- Closure reviews whether objectives in the charter were met—primary Y, consequential metrics, scope, timeline, and benefit claims.
- Formal closure includes handing sustainment to process owners with a live control plan, not merely stopping team meetings.
- Project documentation must be stored in the organization’s system of record so results and methods remain findable.
- Lessons learned capture what worked, what failed, and what to change in future DMAIC projects and tollgates.
- Closure should inform the organization of further opportunities—replication, related Y’s, and portfolio ideas—without reopening uncontrolled scope.
Project Closure & Lessons Learned
Quick Answer: Close a DMAIC project by comparing results to the charter, confirming Control handoff, storing documentation where the organization can find it, running lessons learned, and communicating further opportunities—then release the team with explicit residual-risk owners.
Why Closure Is Part of the BoK
ASQ CSSGB BoK II.C.8 is at Apply level: Green Belts must execute closure, not only “stop working.” Many projects die in a gray zone—Improve finished, meetings faded, control charts unwatched. Closure is the governance act that:
- Declares whether the charter promise was kept
- Transfers accountability to operations
- Preserves knowledge
- Feeds the improvement portfolio with next opportunities
Without closure, benefits claims stay anecdotal and replication is guesswork.
Review Objectives vs. Charter
Start with the approved charter (and any approved change logs). Score the project against commitments:
| Charter element | Closure questions |
|---|---|
| Problem statement | Is the problem reduced in the real process, not only in a pilot week? |
| Goal / primary Y | Did we hit the target (or agreed revised target)? With what statistical/practical confidence? |
| Consequential metrics | Did we harm service, cost, safety, or quality elsewhere? |
| Scope | Did we stay in bounds? If expanded, was it approved? |
| Timeline | Actual vs. plan; major drivers of slip |
| Business case | Hard/soft savings validated by finance method? Ongoing tracking owner? |
| Deliverables | Control plan, SOPs, training, MSA, storyboard complete? |
Outcome categories (be honest)
| Outcome | Meaning |
|---|---|
| Met | Goals achieved and sustained through Control handoff |
| Partially met | Material progress; residual gap documented with next actions |
| Not met | Target missed; root causes of project failure documented |
| Stopped / killed | Leadership ended work; document why (feasibility, strategy shift) |
A “not met” project closed with integrity is more valuable than a “met” claim with no data. Partial success still warrants closure so the organization stops paying for an endless team.
Worked example — charter vs. results:
- Charter goal: Reduce coding defect rate from 6.4% to ≤3.0% within 6 months
- Result: 2.7% for three consecutive months post-Improve; p-chart stable
- Consequential: Average days-to-pay improved 0.4 days (no harm); auditor exceptions flat
- Savings: $142k annualized hard savings validated by Controller method v3
- Decision: Met — close after Control audit #1 and documentation archive
Closure Activities Checklist
Use a structured closeout so nothing critical is left in the Green Belt’s inbox.
1. Results package
- Final before/after for primary and consequential metrics
- Time windows, sample sizes, operational definitions
- Pilot vs. full-scale distinction clearly labeled
2. Control handoff confirmation
- Process owner sign-off on control plan and response plan
- SPC or KPI monitoring live in the operational system
- Training records complete; job aids accessible
- Escalation path tested at least once (tabletop or real signal)
3. Administrative close
- Open actions closed or reassigned with dates
- Budget/code closed if used
- Team recognition and release of borrowed resources
- Risk log: residual risks have operational owners
4. Stakeholder communication
- Champion briefing with outcome category and benefit status
- Thank-you and feedback to SMEs and process staff
- Customer-facing notes only if changes affect them
5. Documentation storage
- See next section—non-negotiable for II.C.8
Documentation Storage
Where documents live matters as much as what they contain. Store the project record in the organization’s system of record (QMS, project PMO library, controlled SharePoint with retention rules)—not only personal drives.
Minimum archive set
| Artifact | Why |
|---|---|
| Final charter + change log | What we promised |
| Storyboard / final report | Narrative of method and results |
| Data dictionary & operational definitions | Reproducibility |
| Key analyses (or pointers) | Defend conclusions |
| Control plan + SOPs + training | Sustainment |
| Benefit validation worksheet | Finance audit trail |
| Lessons learned | Organizational learning |
| Tollgate decision log | Governance history |
| Risk register (final) | Residual risk ownership |
Storage practices
- Use naming conventions and metadata (site, process, date, belt name, Y metric)
- Set access so process owners can read control docs without the belt present
- Follow retention and confidentiality rules (especially healthcare, finance, personal data)
- Link from the control plan to the archive location
- Prefer controlled documents for SOPs; keep analytical workbooks as referenced records
If the next Green Belt cannot find the MSA definition of the defect code six months later, closure failed even if the metric looked green on day 90.
Lessons Learned
A lessons learned session is a structured reflection—ideally with the core team, Champion, and process owner—captured in writing.
Facilitation prompts
| Theme | Questions |
|---|---|
| What went well | Tools, stakeholder moves, data strategies worth repeating |
| What went poorly | Delays, rework, political blockers, tool misuse |
| Surprises | Unexpected X’s, consequential metric moves |
| Method | Was DMAIC fit right? Would Kaizen or DFSS have been better? |
| Tollgates | Which gate added value? Which was rubber-stamp? |
| People | Skills gaps, RACI clarity, sponsorship quality |
| Recommendations | Changes to templates, training, selection filters, IT support |
Rules for useful lessons
- Be specific (“Extract approval took 19 days because Form 12 lacked a backup approver”), not vague (“communication was hard”)
- Separate fact from recommendation
- Avoid personal blame; focus on system fixes
- Feed lessons into deployment standards (charter template, MSA checklist, tollgate agenda)
Lessons learned that never leave a sticky note pile do not meet the intent of closure.
Informing the Organization of Further Opportunities
Closure is not the end of improvement—it is a portfolio feed. Communicate opportunities that appeared during the project but stayed out of scope:
| Opportunity type | Example |
|---|---|
| Replication | Same defect mode at Plant B and C |
| Upstream / downstream | Vendor master data quality feeding coding defects |
| Related Y | First-pass yield on a sibling process |
| System enablers | Permanent data mart for defect codes |
| Quick wins leftover | 5S or standard work items parked during DMAIC |
| DFSS / redesign | Process architecturally incapable beyond current gains |
How to communicate without scope creep
- List opportunities in the final report with estimated impact/feasibility sketches
- Present them at the closure tollgate as candidates, not silent work
- Route into the formal project selection process (II.A.1)—do not auto-start a second project without charter
- Assign a parking-lot owner (Champion or deployment lead), not the exhausted belt by default
This fulfills II.C.8’s expectation to inform the organization of further opportunities while protecting DMAIC discipline.
Closure Tollgate Agenda (Template)
- Charter commitments vs. results (primary + consequential)
- Control system demo / evidence of sustainment
- Benefits status and tracking owner
- Documentation archive location and access
- Lessons learned highlights (3–5 actionable)
- Residual risks and owners
- Further opportunities for portfolio intake
- Formal close decision and team release
Common Closure Failures
- Meeting fade-out — No formal close decision
- Metric mirage — Goal “met” on a two-week honeymoon sample
- Control on paper only — Plan written, charts not reviewed
- Laptop archive — Knowledge leaves with the belt
- No lessons — Same MSA failure repeats next quarter
- Opportunity silence — Replication value dies in a drawer
- Celebrating before finance validation — Credibility burn for the deployment
Green Belt Closure Checklist
- Charter comparison completed and outcome category assigned
- Process owner accepted control plan and monitoring cadence
- Documentation stored in organizational repository with links
- Lessons learned recorded and routed to deployment improvements
- Further opportunities listed for selection/replication
- Residual risks owned outside the project team
- Champion formal close decision recorded
- Team released and recognized
Apply closure rigorously and the project’s value compounds: results stick, methods travel, and the next charter starts smarter. That is the practical meaning of CSSGB II.C.8.
At project close, the Green Belt shows a two-week post-pilot dip in defects to the charter target, but the control plan is unsigned, charts are not in the ops dashboard, and consequential metrics were never rechecked. What is the most appropriate closure assessment?
Which action best fulfills the closure expectation to store documentation and inform the organization of further opportunities?