Goal Failure Review: A Coaching Post-Mortem

A goal failure review examines a result that did not happen as intended. In coaching, a useful post-mortem preserves the original goal, reconstructs what occurred and helps the client choose a responsible response. It should not turn a missed outcome into a character judgment or relabel every loss as success.

The review also needs an appropriate scope. Individual coaching can examine the client's decisions and account of events. It does not replace a formal investigation, grievance process, operational incident review or professional advice required by the consequences.

Coachful is our #1 recommendation for organizing the agreed client reflection, evidence distinctions and follow-up actions. This is our editorial workflow recommendation, without an invented market ranking or a guarantee that reviewing failure improves results. Formal organizational records remain in their authorized systems.

Define what actually failed before explaining why

Recover the original goal, unit, scope and finish date. Four completed conversations, four invitations and four appointments are different targets. A draft delivered for approval differs from an approved publication. Without that distinction, the review can drift toward whichever measure makes the result look most favorable.

Record what completion required and who controlled it. Some outcomes depend on another person's choice or an external decision. That dependence does not make the target meaningless, but it affects what the client could promise and which explanation the evidence supports.

Keep an unsuitable goal visible rather than deleting it. The client may reasonably choose a different target after learning that the original was poorly specified or no longer useful. Record why it changed and what the original result was. A new definition cannot retroactively complete the old goal.

Choose the right review, with realistic evidence

After-action reviews provide one relevant learning tradition. U.S. Army guidance describes a facilitated discussion of observed performance and subsequent improvement, within its own training and operational context. A coaching conversation is not an Army assessment and should not import military authority or standards into the client's situation.

FEMA's guidance separately describes an after-action report as documentation of incident findings, including strengths, areas for improvement and potential actions. A reflective conversation, a formal report and completed corrective action are not the same thing.

Tannenbaum and Cerasoli's meta-analysis, published in the 2013 volume of Human Factors, found an average performance benefit across 46 samples involving 2,136 participants. The research covered varied team and individual debriefs and emphasized alignment of participants, purpose and measurement. That favorable average does not establish a guaranteed improvement for a particular coaching post-mortem or validate the original process below as treatment.

Use the relevant learning principle without overstating transfer: examine experience, preserve evidence and connect a response to what was found. A well-written document alone is not evidence that the response has been used.

Check immediate needs before reflective analysis

A missed goal may create a current obligation: tell someone a delivery will be late, correct a misleading statement, protect a customer or obtain qualified financial or legal advice. Identify those needs before asking the client to find a lesson.

Ask whether the client wants to examine the event now and at what pace. They can choose a pause, a factual starting point or a later conversation. Do not infer a nervous-system state from visible emotion or require a breathing, bodily or positive-reframing exercise as admission to the review.

Acknowledge the actual impact. Disappointment and careful reflection can coexist. The client need not call a setback a gift, and the coach need not promise to remove shame. Needs outside coaching competence require an appropriate support route.

Agree who receives the output. A sponsor might be entitled to an agreed progress summary without receiving private reflections or unnecessary third-party messages. Do not promise confidentiality that contradicts the actual agreement or applicable responsibilities.

Reconstruct decisions with the information available then

Build a short timeline using appropriate records: the goal, requests, confirmations, changes, decisions and the finish date. Mark what is supported, disputed or unknown. A memory can be included as a memory; it should not become an exact quotation without an adequate source.

Distinguish an event from an interpretation. A person canceled a meeting is different from a person does not value me. A request had no answer is different from a request was refused. Later knowledge does not automatically establish what the client could reasonably know earlier.

Include warnings that were available at the time. Avoid both hindsight certainty and a protective account in which no decision can be questioned. If an appointment was never confirmed, ask how it came to be counted as complete rather than blaming an unforeseeable external shock.

Use enough detail to inform the decision, not an exhaustive personal archive. Other people may have relevant perspectives that this coaching account does not contain. State that limitation instead of treating one person's reconstruction as a formal finding about everyone involved.

Test explanations instead of selecting a convenient cause

Separate a supported observation, a plausible contributor and an untested explanation. An unclear handover can plausibly contribute to delay, but one example does not prove it was the sole cause. Ask which evidence supports the connection and what alternative remains possible.

Consider the task, resources, authority, incentives, dependencies and the client's actions. Several can matter simultaneously. A lack of reserved time and an external cancellation may require different responses. An identity label such as lazy or incapable cannot replace examining those conditions.

The familiar five-whys idea can prompt further questions, but five is not a required depth. Each answer needs support. A chain ending in an elegant psychological story is not a verified root cause simply because the coach kept asking.

Accountability remains important. Avoiding unsupported blame does not mean every action was acceptable. The client may need to acknowledge a missed promise, stop pressuring another person or correct a record. Separate those responsibilities from a claim that they controlled the whole outcome.

Choose recovery, prevention or a different goal deliberately

Recovery addresses a consequence that already exists. Prevention aims to change a contributing condition. A new experiment tests whether a proposed change is useful. Label them separately so one completed apology does not become proof that a future workflow is fixed.

Choose a response that fits the finding. More training will not necessarily repair missing approval authority. More invitations will not make another person available. A smaller goal about the client's own actions may be useful, provided it is not presented as completion of the original outcome goal.

Specify the owner, intended effect, effort, decision point and evidence. Check receiving work and permission where other people are involved. A goal is not made feasible by quietly assigning someone an unaccepted task.

Keep the next test limited enough to review honestly. If the client can spare fifty minutes, an action requiring at least a hundred needs a different decision. The answer may be fewer actions, more authorized resources, deferral or stopping.

Worked example: two conversations are not four

All people, exchanges and events below are fictional. Liam is considering a career move. On 3 June 2028, he writes a goal to complete four thirty-minute informational conversations by 30 June. He wants to learn about the work, not ask for job offers or guaranteed introductions.

The goal is an outcome involving other people's voluntary participation. Liam later begins treating appointments and unanswered invitations as progress equivalent to completed conversations. At the 30 June cutoff, the record is:

The original result is two of four, or 50%. Three people had at some point accepted an appointment, but that 75% confirmation figure is not goal attainment. Owen's silence does not establish that he read or rejected the invitation.

Build the actual review record

In the 3 July coaching review, Liam initially concludes that he is bad at networking. Coach Sera asks which events support that identity judgment. They establish that two people did converse with him, one canceled and one never confirmed. Those facts do not establish a universal inability.

Liam also acknowledges a concrete action he regrets. After Nina canceled, he replied that he needed the conversation to reach his target and asked her to fit it in anyway. That pressure was his choice. Her availability was not his to control.

The completed review record states:

Original result: two completed conversations against four by 30 June. Observed contributors: one canceled appointment, one unconfirmed invitation and my own confusion between appointments and completion. My responsibility: I counted states incorrectly and pressured Nina after cancellation. Unknown: why Owen has not answered and whether different timing would have changed participation. Recovery: correct my account and withdraw the pressure on Nina. Next question: can I make a small number of clear optional requests and record their actual states within the time available?

This record does not establish a single cause. Even with better state tracking, Nina could have canceled. The useful response concerns what Liam can change while preserving the other person's choice.

Complete a repair without demanding reassurance

Liam sends Nina a correction on 3 July:

I am sorry I responded to your cancellation by saying I needed you to help me reach my target. That was pressure I should not have placed on you. Please disregard that request. You do not need to rearrange a conversation or explain your availability. I will not follow up about another appointment unless you invite me to do so.

On 4 July, Nina replies: “Thank you for acknowledging it. I am not available for another conversation. Please do not follow up.” Liam acknowledges the boundary and records no further contact. An answer was not required to make his apology legitimate, and her reply is not treated as forgiveness or renewed participation.

Liam's original June goal remains missed. Completing this repair does not add a third conversation to the count.

Reject an infeasible correction and choose a bounded test

Liam first proposes sending twenty new requests. Even allowing only five minutes per request would require a hundred minutes, before replies or record updates. He has fifty minutes available for this experiment. Sera helps him notice the mismatch rather than endorse activity as proof of renewed commitment.

He chooses a new, explicitly different goal: send two clear optional requests by 6 July at 17:00 UTC and record the actual replies or uncertainty. This is an action goal under his control, not a promise of two completed interviews.

The accepted personal allocation is thirty minutes on 4 July and twenty on 6 July. The first block contains ten minutes selecting appropriate contacts, fifteen drafting and five sending. The second contains ten minutes reading and replying and ten updating the record. Review time with Sera is separate.

Neither Nina nor Owen is selected for renewed pressure. Liam chooses Anita and Jo, with whom an appropriate professional contact route already exists. He asks for voluntary participation and does not claim an entitlement to their time.

Show the actual requests and answers

Liam's request to Anita is:

Would you be open to a thirty-minute informational conversation about your work? I am exploring a possible career move and would value your perspective on the role. I can offer 18 July, 13:00–13:30 UTC. Please feel free to decline. This is not a request for a job referral or an introduction, and I will not treat no answer as agreement.

Anita replies on 5 July: “I can speak on 18 July from 13:00 to 13:30 UTC. I am happy to discuss the work informally, but cannot offer a referral or review your application.” Liam accepts those limits and confirms the time. He checks his own availability and reserves the conversation and ten minutes of later preparation. Those are future allocations, not completed work in this experiment.

Jo receives a separately addressed optional request for the same kind of conversation. On 6 July, Jo declines without giving a reason. Liam replies, “Understood, thank you. I will not follow up about a conversation.” No alternate appointment or hidden obligation is recorded.

By the 6 July, 17:00 UTC cutoff, Liam has sent both requests and updated the record: one future appointment accepted within stated limits, one request declined. Both requests are completed actions. Neither is a completed new conversation.

Review the test without rewriting the missed goal

At the 10 July review, the new action goal is complete. The original June outcome is still two of four. The future Anita conversation is scheduled for 18 July, and there is no evidence yet about whether it will occur or what Liam will learn.

The request experiment used its planned fifty minutes: thirty for selecting, drafting and sending, then twenty for replies and records. A declined request was handled within that allocation. Liam does not count the entire second block as more outreach.

The completed client-and-coach work for the review, repair and new experiment is:

That account totals 245 client-and-coach person-minutes. It does not purport to include the earlier June activity, the counterparts' voluntary time or the future July conversation. The original two conversations actually took place; omitting their time from this later review account does not erase them or make them free.

Liam remains disappointed about the missed June target. He says the corrected states and the limited experiment were useful, while Sera notes that usefulness does not prove the review caused Anita's acceptance. Another person could still cancel, and Jo's refusal remains an ordinary valid outcome.

Compare a different kind of missed goal

Suppose a client promised to submit a completed report by Friday and had all the necessary authority and inputs. The report was not sent. That review differs from Liam's partly dependent outcome. Examine the actual work, priorities, available time and decision to delay rather than automatically attributing the miss to other people's choices.

Conversely, a client can complete every appropriate action and still lose a competitive bid or receive a declined invitation. Do not infer a personal defect from the outcome. The review should preserve both effort and result without pretending that either tells the whole story.

If the goal itself has become unsuitable, use an explicit goal recalibration conversation. Keep the old record and the reason for change. A different goal can be a sound next decision without serving as a retroactive success label.

Use a concise record and an actual follow-up

A useful record contains the intended result, actual state, impact, timeline, supported observations, possible contributors, uncertainties, response and next review. Use clear state labels such as sent, accepted, scheduled, completed and declined where they matter.

At follow-up, ask whether the action occurred and whether the relevant condition changed. A completed checklist does not by itself establish that a handover is usable. A sent apology does not prove the recipient's view changed. A completed new experiment does not alter the old result.

If no justified lesson emerges, record that uncertainty. Further information is worth obtaining when it could change an important decision. Do not collect more sensitive material or invent a positive insight merely to finish the worksheet.

Organize the client reflection in Coachful

A worksheet can hold the timeline and evidence distinctions. Liam's completed personal response is: “Two conversations happened. I treated other states as completion and pressured Nina. I corrected that message, respected the refusal and sent two new optional requests. One future appointment is agreed; it has not happened.”

Sera's Coach review is private to the authorized coach team. Saving Reviewed or Follow up does not deliver feedback in the client's response PDF. Sera sends the actual comment separately through Liam's private client conversation: “Your account keeps the original result, the repair and the new experiment distinct. At the next review, bring the actual state of the future appointment rather than counting its confirmation as completion.” Liam acknowledges that he can maintain those distinctions.

Agree the audience and necessary information before collecting material. Some specialty workflows allow client-explicit sharing of journal pages, so do not promise that every journal page is always unreadable by coaches. A sponsor-facing summary should include only the information agreed for that audience.

Recaps send automatically by default. Set Approve before sending and check session-type overrides before relying on a held recap. Review the actual content and send it or keep it private; this cannot recall a message already sent. The demonstration screens show client preparation and follow-up, not a formal investigation or automatic proof that a goal was achieved.

The weekly accountability guide develops ongoing review. Explore Coachful's features for the supporting client workflow.

Practical steps and troubleshooting

1. Preserve the original target: define its unit, scope, completion criterion and finish date.
2. Address urgent consequences: agree readiness, responsibilities and the appropriate review route.
3. Reconstruct the sequence: separate records, recollections, interpretations and unknowns.
4. Examine contributors: test explanations and preserve both external dependencies and the client's actions.
5. Choose a feasible response: distinguish repair, prevention and a new experiment with accepted effort and authority.
6. Review actual use: check what happened without rewriting the original result or guaranteeing the next one.

If the timeline changes, correct it rather than defend the first story. If the proposed fix does not address the finding, reconsider it. If an action depends on someone else, obtain real agreement. If a formal concern or significant distress exceeds coaching scope, use the appropriate professional or organizational route.

Frequently asked questions

When should the review happen?

Preserve relevant information promptly, address urgent needs and agree a useful time. There is no universal deadline that overrides readiness or an appropriate formal process.

Does avoiding blame remove accountability?

No. Examine actions and consequences accurately while avoiding unsupported identity judgments. Formal responsibilities and separate processes remain where appropriate.

Must I ask why five times?

No. Further inquiry needs evidence. A repeated question can deepen speculation instead of finding a cause.

Can the next goal be easier or different?

Yes, if the change is deliberate and justified. Preserve the old result and explain why the new target serves the next decision.

What if the review produces no clear lesson?

Record the limits and decide whether more information could change a useful action. An honest uncertainty is preferable to an invented success story.