Lessons Learned and Post-Project Reviews

The "Why": The Most Valuable Knowledge in Your Organisation Is Vanishing

Every completed project generates a unique body of knowledge — about what worked, what failed, which assumptions held, and which crumbled under contact with reality. This knowledge exists, briefly, in the minds of the people who lived through the project. Then those people move on to new assignments, leave the organisation, or simply forget.

Post-project reviews are the mechanism designed to capture this knowledge before it evaporates. Research consistently demonstrates that they are effective at disseminating good practices, correcting errors in people's understanding of how other functions within the organisation operate, and predicting how well alternative approaches would have performed. Yet these reviews are routinely curtailed, conducted superficially, or skipped entirely.

J.S. Busby's detailed study of post-project reviews across three engineering companies found that while the reviews delivered genuine value, they also exhibited several systematic weaknesses: diagnoses tended to be shallow, remedies were planned only at a superficial level, and participants made potentially misleading assumptions by treating specific events as evidence of general causes. The reviews were valuable — but far less valuable than they could have been.

Core Principle: Post-project reviews are important learning mechanisms whose value is consistently underestimated by individuals who do not appreciate the need to disseminate insights throughout the organisation.

---

The "What": Understanding Post-Project Reviews

A post-project review (also called a post-mortem, retrospective, or lessons-learned review) is a structured assessment exercise conducted after a project is complete. Its purpose is to identify what went right, what went wrong, and what should be done differently in the future.

Formal structured reviews are acknowledged as an essential component of project closure. They serve three functions simultaneously:

1. Internal Learning — Helping the project team identify successes and failures
2. External Reporting — Providing the project manager with information needed to prepare final feedback to clients, consultants, and contractors
3. Organisational Intelligence — Contributing documented lessons to the performing organisation's continuous improvement process

How to Conduct a Project Review

The recommended strategy involves two steps:

Step 1: Circulate a list of structured questions to all project team members and key stakeholders before the review meeting. This gives participants time for considered reflection rather than off-the-cuff reactions.

Step 2: Facilitate a meeting to formally discuss their responses.

Sample review questions include:

• What was the most satisfying thing about the project?
• What was the most frustrating thing about the project?
• Which of our processes worked well?
• Which processes did not work well?
• What would you do differently next time?
• Did we maintain a good relationship with our stakeholders?

Building the Review Foundation

Before the review meeting begins, two preparatory actions significantly improve the quality of discussion:

Build a project timeline — Map the project's milestones and key events chronologically. This enables you to show key decisions, movement of people in and out of the project, and budget approvals or rejections. The timeline becomes a shared reference that anchors discussion in facts rather than selective memory.

Define clear objectives — Assure people that this is not a witch hunt by defining very clear objectives, specifying which issues need to be addressed, and explaining how the results will be used. Without this assurance, participants will self-censor to avoid blame.

Prepare for Review

Build Project Timeline — (Milestones, Decisions, People)

Define Clear Objectives — (Not a Witch Hunt)

Circulate Questions — to All Stakeholders

Conduct Review Meeting

Develop Findings

Tentative Findings — (Major & Minor Events)

Preliminary Recommendations

State Findings Objectively

Build Action List

Document & Distribute — Lessons Learned

Relationship details

From	Relationship	To
Prepare for Review	leads to	Build Project Timeline — (Milestones, Decisions, People)
Prepare for Review	leads to	Define Clear Objectives — (Not a Witch Hunt)
Prepare for Review	leads to	Circulate Questions — to All Stakeholders
Build Project Timeline — (Milestones, Decisions, People)	leads to	Conduct Review Meeting
Define Clear Objectives — (Not a Witch Hunt)	leads to	Conduct Review Meeting
Circulate Questions — to All Stakeholders	leads to	Conduct Review Meeting
Conduct Review Meeting	leads to	Develop Findings
Develop Findings	leads to	Tentative Findings — (Major & Minor Events)
Develop Findings	leads to	Preliminary Recommendations
Tentative Findings — (Major & Minor Events)	leads to	State Findings Objectively
Preliminary Recommendations	leads to	State Findings Objectively
State Findings Objectively	leads to	Build Action List
Build Action List	leads to	Document & Distribute — Lessons Learned

---

The "How": How People Actually Learn in Reviews

Busby's research identified four distinct cognitive mechanisms through which participants learn during post-project reviews. Understanding these mechanisms helps facilitators design better reviews — and avoid the traps each mechanism carries.

1. Dialectic Argument

Participants commonly resort to a dialectic form of reasoning. One person offers an explanation, another counters with a contradictory explanation, and a third finds a synthesis that incorporates both. This process reflects the reality that any event has multiple sides and no single person holds enough information to see the complete picture.

Facilitator Implication: Structure discussions to deliberately invite opposing viewpoints. The synthesis that emerges from constructive disagreement is typically more accurate than any individual's initial explanation.

2. Event Rehearsal

Participants mentally replay sequences of events — recalling their interactions with clients, the succession of design changes, the chain of decisions that led to an outcome. This replay is a natural cognitive process: it helps people test whether event A actually caused event B.

Caveat: Participants tend to rehearse event sequences without attaching precise times to them. Precedence alone is only a partial indication of causality — and people are susceptible to inferring causality where none exists.

3. Mental Simulation

Participants work out what would have happened had practices been different. For example, reasoning about how using a different supplier would have changed the outcome. This counterfactual thinking is valuable because it broadens the sample of experiences from which conclusions can be drawn.

Caveat: Mental simulation uses causal reasoning (working forward from cause to effect), whereas deep diagnosis requires diagnostic reasoning (working backward from effect to cause). People instinctively prefer causal reasoning because it avoids the uncomfortable question of who did what wrong. The result is a lack of deep diagnosis — people jump from symptoms to remedies without tracing back to root causes.

4. Historical References

Participants occasionally reference past projects or organisational history to contextualise current events. Historical references serve three functions: providing evidence for explanations, demonstrating that the organisation has changed over time, and explaining why people behave in certain ways.

Caveat: Historical references were remarkably rare in Busby's study — only six appeared across 12 hours of review meetings. Without historical context, participants cannot determine whether problems are unique to this project, characteristic of a pattern, or systemic. This severely limits the quality of remedies.

How People Learn in Reviews

Dialectic — Argument

Event — Rehearsal

Mental — Simulation

Historical — References

Conflict without — resolution

False causality — from precedence

Causal reasoning — replaces diagnosis

Too rare — patterns — go undetected

Relationship details

From	Relationship	To
Dialectic — Argument	Risk	Conflict without — resolution
Event — Rehearsal	Risk	False causality — from precedence
Mental — Simulation	Risk	Causal reasoning — replaces diagnosis
Historical — References	Risk	Too rare — patterns — go undetected

---

The Pitfalls: Why Reviews Produce Shallow Results

Busby's research identified five systematic weaknesses that undermine the learning quality of post-project reviews. These are not occasional problems — they are structural tendencies embedded in how people think and interact.

1. Attribution Bias

Participants in reviews tend to overemphasise environmental factors and understate their own involvement when explaining results. Problems are attributed to forces beyond the participants' control — third parties, external conditions, client behaviour — while personal errors are minimised. In Busby's study, only two occasions across all reviews involved an individual admitting an error or a need to change their working approach.

Successful individuals and organisations exhibit the opposite tendency: an "internal locus of control" — the belief that events are within their control, which motivates them to ask what they could have done differently.

2. Excessive Concreteness

Review participants were too narrowly specific in their diagnoses. For instance, locating a piece of equipment in a place where it was hard to maintain was diagnosed simply as "a slip" — with no attempt to determine whether it reflected a broader pattern of installation and maintenance problems in equipment design.

Overly specific diagnoses produce overly specific remedies that address symptoms rather than systemic causes. Very few participants asked generalising questions like "Is this a case of a bigger problem?" or "Are we missing something larger?"

3. Shallow Diagnosis

Participants preferred causal reasoning (cause → effect) over diagnostic reasoning (effect → cause). No one during the reviews asked a truly diagnostic "why" — they only asked clarifying "whys" (meaning "in what way was it poor?" rather than "what were the underlying causes?").

The explanation is partly cognitive (people find forward reasoning easier) and partly social (asking "why did you do that?" feels like an accusation, threatening working relationships).

4. Lack of Historical Context

With only six historical references across 12 hours of meetings, participants had no basis for determining whether problems were one-off events or recurring patterns. Without this context, every problem appears unique, and remedies cannot be calibrated to the actual frequency or severity of the underlying issue.

5. Superficial Remedies

Remedies proposed during the reviews were not analysed for side-effects, were only briefly contested, and were not planned through implementation. Two of the four reviews were rushed toward the end, giving remedies only superficial treatment. Even in the more thorough reviews, there was only an acknowledgement that someone should "go away and plan the remedies in more detail."

The Uncomfortable Truth: Organisational culture often prioritises being constructive — "come to me with solutions, not problems" — which actively discourages the deep causal investigation that effective reviews require. People avoid exploring root causes not because they lack analytical skill, but because doing so risks surfacing uncomfortable truths about colleagues they must continue working with.

---

Gathering Lessons Learned: Sources and Techniques

Valuable knowledge can be captured from a broad range of project participants:

Techniques for gathering lessons learned include meetings, surveys, interviews, observations, audits, and informal discussions. The most effective approaches combine structured methods (surveys, audits) with unstructured ones (informal discussions) — since people often share their most candid assessments outside of formal settings.

The Lessons Learned Register

All captured lessons should be recorded in a structured register:

Key Point: Lessons learned typically highlight issues associated with risk management, human resources management, contract management, and communications management. However, they should also capture insights across all project knowledge areas.

---

Six Recommendations for Better Reviews

Based on the research evidence, the following practices significantly improve the quality and impact of post-project reviews:

1. Encourage deep diagnosis. Use cause-and-effect diagrams (fishbone diagrams) and other analytical tools. Push past the first explanation to find root causes.

2. Encourage attention to history. Actively ask whether similar things have occurred historically. Without this question, every problem looks unique and every remedy looks adequate.

3. Encourage examination of the bigger system. Look beyond the immediate confines of the project. A "communications problem" between two people is a starting point for diagnosis, not a conclusion — examine how different assumptions arise and why they persist.

4. Discourage glib categorisation. Labelling something as "a communications problem" and moving on is not diagnosis. It is evasion. Categories are starting points, not finishing points.

5. Plan remedies properly. Examine side-effects. Think through implementation. If this requires a second meeting, hold a second meeting. Unplanned remedies breed cynicism — people learn quickly that review recommendations never materialise.

6. Invite key outsiders. Inviting managers from new projects to attend the review is one of the most effective dissemination strategies available. In Busby's study, outsiders gained a profound understanding of what had succeeded and failed — seeing not just the conclusions but the reasoning behind them. Written summaries, by contrast, tend to reflect one person's perspective and lack the contextual richness of observed dialogue.

Six Recommendations

1. Deep Diagnosis — (Use Fishbone Diagrams)

2. Historical Attention — (Ask: Has This Happened Before?)

3. System-Level Thinking — (Look Beyond Project Boundaries)

4. No Glib Categories — (Categories ≠ Conclusions)

5. Plan Remedies Properly — (Side-Effects, Implementation)

6. Invite Outsiders — (Best Dissemination Method)

Higher-Quality — Organisational Learning

Relationship details

From	Relationship	To
Six Recommendations	leads to	1. Deep Diagnosis — (Use Fishbone Diagrams)
Six Recommendations	leads to	2. Historical Attention — (Ask: Has This Happened Before?)
Six Recommendations	leads to	3. System-Level Thinking — (Look Beyond Project Boundaries)
Six Recommendations	leads to	4. No Glib Categories — (Categories ≠ Conclusions)
Six Recommendations	leads to	5. Plan Remedies Properly — (Side-Effects, Implementation)
Six Recommendations	leads to	6. Invite Outsiders — (Best Dissemination Method)
1. Deep Diagnosis — (Use Fishbone Diagrams)	leads to	Higher-Quality — Organisational Learning
2. Historical Attention — (Ask: Has This Happened Before?)	leads to	Higher-Quality — Organisational Learning
3. System-Level Thinking — (Look Beyond Project Boundaries)	leads to	Higher-Quality — Organisational Learning
4. No Glib Categories — (Categories ≠ Conclusions)	leads to	Higher-Quality — Organisational Learning
5. Plan Remedies Properly — (Side-Effects, Implementation)	leads to	Higher-Quality — Organisational Learning
6. Invite Outsiders — (Best Dissemination Method)	leads to	Higher-Quality — Organisational Learning

---

The Dissemination Problem

Even when reviews produce valuable insights, those insights frequently fail to reach the people who need them. Busby's research identified a critical disconnect: most review participants — the people who worked on the project under scrutiny — would say things like "I already knew X." But the knowledge needs to reach people on future projects, not just the people who just finished this one.

Dissemination matters because organisations are rarely structured so that the same person always does the same type of work. What one person learns on Project A must reach the people who will face similar challenges on Projects B, C, and D. Without deliberate dissemination, this transfer does not happen, and repeated errors become a characteristic of organisational life.

The reviews in Busby's study served several dissemination functions even when participants did not fully recognise it. They gave people a chance to demonstrate concern with organisational objectives, helped correct misconceptions formed during normal project activity, suggested available practices that had not been considered, gave people a chance to explain and justify their actions, and promoted collective commitment to remedies.

Without Dissemination

Project A — (Lessons Learned)

Project B

Project C

Project D

Team Dispersal

No Documentation

No Formal Channel

With Structured Dissemination

Project A — (Lessons Learned)

Lessons Learned — Repository

Future Project — Managers Review

Project B

Project C

Project D

Relationship details

From	Relationship	To
Project A — (Lessons Learned)	leads to	Project B
Project B	leads to	faded
Project A — (Lessons Learned)	leads to	Project C
Project C	leads to	faded
Project A — (Lessons Learned)	leads to	Project D
Project D	leads to	faded
Project A — (Lessons Learned)	leads to	Team Dispersal
Project A — (Lessons Learned)	leads to	No Documentation
Project A — (Lessons Learned)	leads to	No Formal Channel
Project A — (Lessons Learned)	leads to	Lessons Learned — Repository
Lessons Learned — Repository	leads to	clear
Lessons Learned — Repository	leads to	Future Project — Managers Review
Future Project — Managers Review	leads to	clear
Future Project — Managers Review	leads to	Project B
Project B	leads to	clear
Future Project — Managers Review	leads to	Project C
Project C	leads to	clear
Future Project — Managers Review	leads to	Project D
Project D	leads to	clear

---

Key Takeaways

• Post-project reviews are valuable but consistently underutilised — they are often curtailed, conducted superficially, or skipped entirely
• People learn in reviews through four mechanisms: dialectic argument, event rehearsal, mental simulation, and historical references — each carries specific cognitive risks
• Five systematic weaknesses undermine review quality: attribution bias, excessive concreteness, shallow diagnosis, lack of historical context, and superficial remedies
• The organisational norm of "being constructive" actively discourages the deep diagnostic investigation that reviews require
• Lessons learned should be captured from all project participants — not just the project team — using both structured and unstructured techniques
• Inviting managers from future projects to attend reviews is one of the single most effective dissemination strategies available
• Always include the project team in the review process — they can provide insight into how recommended changes would actually affect project outcomes