Project "Completed," but Benefits Can't Withstand Scrutiny? — Case Analysis of a Rebuilt Six Sigma Project Closure Review System in an Equipment Manufacturing Company

Companies implementing Six Sigma generally focus their efforts on two ends: rigorous gatekeeping during project initiation and phased reviews during execution. However, when it comes to "project closure," the gates often open wide. Review meetings turn into celebration sessions, with the project team presenting a 20-minute PowerPoint, and evaluators asking a few superficial questions, followed by applause and approval. The project team fills in the benefit figures, and the "results solidification" section often reads "control plan updated" as a formality. It wasn't until one day when the finance department refused to recognize the figures and the improvement effects began to wane that management realized: the "last mile" of project closure review had never been properly managed. This article uses a real case to dissect the four critical points behind the superficial project closure reviews and the effects of the rebuilt system.

1. Phenomenon: Eighteen Projects Closed, Finance Recognizes Only Half

An equipment manufacturing company has been implementing Six Sigma for five years, with over 40 Black Belts and 150 Green Belts trained. Each year, about 20 projects are initiated, and management expects "project completion to equal benefit realization." This year, 18 projects were closed, with six project closure review meetings held, all following a highly consistent process: the project team presented the entire DMAIC process in 20 minutes, evaluators asked two or three questions, and the project was approved with collective applause. The total annual benefits reported by the 18 projects amounted to 21 million yuan, which was included in the quality department's annual summary.

A year later, at the request of management, the finance department conducted a special review. The results were alarming: out of the 18 projects, the finance department could only confirm benefits of just over 9 million yuan, less than half of the reported amount. Even more disheartening, three projects showed significant regression within a year of closure—improvement measures were still in place, but the effects had vanished, with defect rates rebounding to 80% of pre-improvement levels. The quality director reviewed the 18 project closure reports and identified three frequent issues:

1. Diverse methods of benefit calculation, with some projects calculating capacity gains based on theoretical cycle times, others converting customer satisfaction improvements into monetary terms, and still others multiplying "expected reduction of 3 customer complaints" by the single complaint cost.
2. The "results solidification" section was often a single sentence, mostly stating "control plan updated." However, upon on-site inspection, two control plans were found to have never gone through the controlled distribution process, and the on-site execution documents were still the old versions.
3. The reports claimed "15 operators trained," but the training records lacked signature pages.

2. Diagnosis: Four Breakpoints Behind the Superficial Reviews

The quality department did not rush to hold the project teams accountable but instead reviewed all 34 projects that had been closed over the past two years. The conclusion was that the problem did not lie in the project teams being lazy, but in four breakpoints within the project closure review process itself.

1. Lack of Standards: There was no clear boundary between "completed" and "well done." Project closure equaled the completion of the report, with no quantitative acceptance criteria: to what extent must improvement goals be achieved to be considered qualified? How long must the effects be sustained to count? What criteria should be used to calculate benefits? Without these three essential tables, the review process relied solely on subjective judgment.
2. No Rules for Benefits: Benefits were self-reported by the project teams, with no financial participation in the closure review. There was no unified classification or calculation method for benefits, and no one verified them. The 12 million yuan gap between the reported 21 million yuan and the confirmed 9 million yuan was not due to intentional falsification by the project teams but rather a lack of scrutiny.
3. No Verification of Solidification: The entire closure review process took place in meeting rooms, with no evaluators visiting the production site. Whether the control plan was controlled and distributed, whether the poka-yoke devices were effective, and whether the SOPs were updated and training completed—these critical pieces of evidence that determine the long-term survival of improvements were not checked. If the report stated "solidified," it was accepted as such.
4. Decoupled Certification: The results of the closure review were completely decoupled from Black Belt certification, performance rewards, and mentor evaluations. Whether a project passed or failed the review had no impact on certification, bonuses, or mentor performance. This led to a situation where project teams treated closure as a "homework submission" rather than a "results delivery."

3. Rebuilding: Making the Closure Review a "Quality Gate"

After identifying the breakpoints, the company spent three months rebuilding the project closure review system. The core idea was simple: transform the closure review from a "procedural ritual" into a "quality gate." The specific approach involved a three-stage review process and three supporting mechanisms.

Three-Stage Review, Layered Checks

1. Initial Document Review: Led by the quality department, this stage involved a detailed check against the project closure documentation list: whether the evidence chain for each DMAIC phase was complete, whether the benefit calculation used a unified "benefit calculation table," and whether the results solidification list was itemized. Incomplete documentation was returned and not allowed to proceed to the next stage.
2. On-Site Verification: The review committee visited the production site to verify the improvements: comparing data system records before and after the improvements, checking whether the control plan was controlled and distributed, whether the on-site execution version was the latest, and verifying the effectiveness of poka-yoke devices and the completion of SOP updates and training signatures. Projects that failed on-site verification were returned for rectification.
3. Final Defense Review: The project team presented to the review committee, with finance personnel confirming the benefits on-site. Evaluators focused on the sustainability plan and the horizontal deployment plan. Only projects that passed all three stages were considered officially closed.

Three Supporting Mechanisms, Plugging the Loopholes

1. Financial Pre-Review: Financial personnel were involved in benefit calculation from the project initiation stage. At closure, benefits must be confirmed with a financial signature. Direct benefits were calculated according to the financial accounting criteria, and indirect benefits were listed separately and not mixed.
2. Reorganized Review Committee: The committee was composed of Black Belt Masters, financial managers, process engineers, and production supervisors, with customer representatives invited when necessary. This changed the previous situation where the quality department evaluated its own projects.
3. Linked Review Results: Projects that failed the closure review had their Black Belt certification delayed for three months and were required to rectify. The review pass rate was linked to the performance of Black Belt mentors, with mentors whose projects had a pass rate below 70% for two consecutive years being suspended from mentoring.

4. Effect: Reviewing "Critically," Realizing "Solidly"

In the first year after the rebuild, the effects were immediate. The 18 projects that year reported benefits of 19 million yuan, 200 million yuan less than the previous year—because the project teams learned to "calculate according to the criteria" and no longer overestimated. The finance department ultimately confirmed 16 million yuan, with the verification rate increasing from less than half to over 80%. More importantly, the regression rate changed: three out of the 18 projects from the previous year showed significant regression, while the projects closed after the rebuild maintained their effects at over 90% a year later. The tone of the review meetings also changed: from "celebration sessions" to "critical sessions," with project teams conducting three rounds of self-checks before submitting their closure materials, knowing they couldn't avoid the on-site verification.

The supporting case was equally convincing. An electronics company that adopted this approach implemented a tiered review system: Black Belt projects went through the full three-stage review, while Green Belt projects used a simplified version—initial document review and on-site verification, skipping the final defense review, with Black Belt Masters overseeing. The accuracy of benefit reporting for Green Belt projects significantly improved, and the review cost increased by less than 20%. This demonstrates that project closure review is not about being overly heavy but about the "gate" mindset: each level of project has clear acceptance criteria, and each benefit is scrutinized.

5. Key Implementation Points: Six Experiences

1. Establish Standards Before Evaluating Projects: The benefit calculation table, results solidification list, and review scoring table should be in place first. No standards, no review meetings.
2. Finance Must Be Involved: Benefits without financial confirmation are not counted. This is the most effective way to prevent "paper benefits."
3. On-Site Verification Cannot Be Skipped: The review process must include at least one on-site visit. Evidence is on-site, not in the PPT.
4. Link Review Results to Consequences: Tie the review results to certification, performance, and mentor evaluations to give the review process teeth.
5. Tiered Review to Control Costs: Full process for Black Belt projects, simplified version for Green Belt projects, ensuring the review cost matches the project level.
6. Closure Is Not the End: Set a three to six-month benefit tracking period. Projects that show regression during the tracking period must be re-evaluated, and horizontal deployment plans must be followed up.

The project closure review is the "last mile" of the Six Sigma mechanism and the final gate for benefits to move from "paper" to "account." Choosing the right projects, executing them correctly, and closing them rigorously—these three steps must be tightly connected for Six Sigma to truly form a closed loop. Completing a project is not difficult; the challenge lies in ensuring it can withstand scrutiny after completion.

A superficial project closure review means Six Sigma benefits are "paper wealth"—managing the last gate ensures that improvements truly take root.

Knowledge code: 6.1.3

Version: v20260826

Author: Quality Think Tank Quality Think Tank is dedicated to providing systematic professional knowledge, methodologies, and practical tools for quality management practitioners, helping companies continuously improve their quality capabilities.