Post-Investment Review¶
Retrospective — instantiates Escalation Exit Gate
Runs after a committed effort ends or pauses, converting the time and money already spent into reusable lessons and a dignified way to have stopped.
A Post-Investment Review is the structured look-back that runs when a committed effort ends, pauses, or reaches a decision point, and its job is to make the spending that already happened worth something. Its defining move is backward-facing conversion: it treats prior investment as two things at once — information to harvest, and a story that can be told without humiliation. It extracts the transferable lessons, reframes the money and time already gone as a cost paid rather than a debt owed to the future, and builds the exit narrative that lets stopping read as disciplined stewardship instead of failure. It informs decisions but does not make them; it is the retrospective, not the gate. Its entire value is in what the organization carries forward and how the people who ran the effort are treated afterward.
Example¶
A mid-size pharmaceutical company terminates a Phase II cardiovascular candidate after the trial misses its primary endpoint; roughly $180M and six years are gone. Eight weeks later, a post-investment review convenes. It extracts a specific, transferable lesson: the early biomarker the program had trusted for its internal go/no-go calls did not predict the clinical endpoint — and that lesson is written directly into the company's target-validation playbook so the next program does not repeat the mistake. It builds the face-saving exit: the program lead presents the termination at the R&D portfolio review as a well-run kill on pre-agreed evidence, the medicinal-chemistry work is redeployed to a sibling program, and the narrative to the board is "disciplined stop," not "wasted six years." And it names, on the record, the sunk-cost reasoning that had crept into the last renewal debate — "we're too far in to stop before Phase III" — so the reframe is explicit: the $180M is a cost already paid, not a reason to run a $400M Phase III to justify it.
Nothing that already happened changes. What changes is that a $180M failure becomes a cheaper future success, and the scientists who ran an honest program are not punished for reporting that it did not work.
How it works¶
- Run at or after the stop/pivot decision, not instead of it. The review harvests a commitment that is already ending; it does not adjudicate whether to end it.
- Extract lessons where the next effort will meet them. Findings are lodged in the live playbook, checklist, or process — not filed in a report no one reopens.
- Separate preventable error from irreducible uncertainty. Distinguishing bad luck from bad process is what keeps learning from curdling into blame.
- Reframe prior spend as sunk. State plainly that past cost is information, not a claim on future resources.
- Craft the exit narrative and stakeholder comms. Give the people involved a way to have stopped that reads as discipline, and a plan for who is told what.
- Preserve accountability alongside dignity. A dignified exit is not amnesty — preventable harm is surfaced, not hidden behind "learning."
Tuning parameters¶
- Timing — immediately at the gate versus after a cooling-off period. Immediate captures detail; delayed captures perspective and lowers defensiveness.
- Blame-versus-learning stance — how much psychological safety is engineered in. More safety surfaces more truth but risks letting real negligence pass as "a learning."
- Lesson-capture depth — a one-page headline versus a full teardown. Deeper is more reusable but more expensive and more likely to go unread.
- Lodging location — a searchable playbook the next team hits versus an archive. Where the lesson lives determines whether it is ever reused.
- Dignity-versus-accountability balance — how far the exit narrative protects the people involved. Too little and no one reports bad news; too much and no one is responsible.
When it helps, and when it misleads¶
Its strength is that it addresses the real blocker to stopping, which is usually social rather than evidentiary: it makes ending survivable, so people can report a weak commitment instead of hiding it, and it recycles expensive failure into cheaper future success. The disciplined-retrospective form it borrows is the After-Action Review — a structured look-back asking what was expected, what happened, why, and what to do differently.[1]
It misleads when it degrades into a documentation ritual disconnected from any decision — lessons written that no one reads — or when face-saving curdles into excuse-making, so that no one is accountable and the organization concludes, falsely, that nothing could have been done. The classic misuse is invoking "we're learning" to bury a preventable harm behind the language of growth. The guarding discipline is to lodge lessons where the next team will actually collide with them, and to keep dignity and accountability both real: bad news reported without humiliation, but negligence not laundered as learning.
How it implements the components¶
learning_extraction_record— it captures the transferable lessons from the spent effort and lodges them where the next effort will reuse them.face_saving_exit_path— it builds the exit narrative, stakeholder comms, and redeployment plan so that stopping is not treated as personal failure.sunk_cost_flag— applied in hindsight, it reframes prior spend as information, naming where continuation had been argued from cost already paid.
It does not carry decision authority or fire the stop/redirect (stop_or_pivot_rule, independent_review_gate) — those are Pivot Gate and Independent Project Audit; a retrospective informs the call but does not make it. Nor does it set forward release criteria (continuation_criteria), which is Funding Tranche Review and Stage-Gate Review.
Related¶
- Instantiates: Escalation Exit Gate — supplies the learning-and-dignity path that lets a commitment end without the ending being counted as failure.
- Consumes: the stop, pivot, or continuation decision produced by Independent Project Audit, Pivot Gate, or Stage-Gate Review — the review harvests whatever they conclude.
- Sibling mechanisms: Pivot Gate · Independent Project Audit · Funding Tranche Review · Stage-Gate Review · Project Kill Criteria · Stop-Loss Rule
Editorial Notes¶
Form Classification¶
Form family: Assessment, Review & Assurance
Rationale: Post-Investment Review operates as a bounded evaluation of existing evidence or work that produces a finding or disposition because it runs after a committed effort ends or pauses, converting the time and money already spent into reusable lessons and a dignified way to have stopped.
Independent corroboration: The frozen evidence defines Post-Investment Review as 'Runs after a committed effort ends or pauses, converting the time and money already spent into reusable lessons and a dignified way to have stopped', so its operative form is Assessment, Review & Assurance.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Organizational & Management Science
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Post-investment review is a project and portfolio governance practice for learning from completed commitments.
Related originating lineages:
- Economics & Finance — Finance contributes capital-allocation, sunk-cost, and realized-return analysis.
Review resolution: Both blind reviewers agree that organizational management is the primary origin. Reconciliation resolves origin mode disagreement. Formative alternate lineages are retained as economics_finance; later breadth of use is recorded separately as domain_reach=multi_domain, while origin_mode=cross_disciplinary_synthesis describes the relationship among origin lineages.
Review outcome: Reconciled after independent review; high confidence.
References¶
[1] United States Department of the Army. A Leader's Guide to After-Action Reviews. Training Circular 25-20. Washington, DC: Headquarters, Department of the Army (1993). Structures an After-Action Review around intended events, actual events, causal explanation, and improvements for future performance. registry ↩