Turn a work retrospective into a next-period plan you can check

A useful review identifies what happened, which explanations need checking, and how to inspect one proposed change next time. Compare evidence with the goal, choose a worthwhile action, then specify ownership, dates, dependencies, acceptance, and follow-up.

This exercise turns fictional progress material into a fact-and-explanation table, one proposed improvement, and a complete report. Use the in-article record to retain your own initial version and append execution, evidence, and next-step decisions later. Action completion and observed effects stay separate. An action awaiting approval remains a proposal.

Reconcile the plan and actual progress

Original fictional teaching material: the overall configuration task contains 12 units. Eight in this week's plan were due by the agreed check-in; six are complete and two await interface information. The exact effect on Friday's demo and the full causes remain open. There are no measurements of customer waiting or efficiency. Today, tomorrow, and Friday are relative exercise times; replace them with actual dates when adapting the material.

Retain the original goal's timing and definition. If a goal was subsequently changed with approval, record the original, change, approval, and effective date before comparing results. A revised target should not conceal the original gap.

Choose one retrospective approach

KPT helps organize what to keep, problems to address, and changes to try. YWT draws learning from what you did before choosing what to do next. This example uses KPT to focus on work improvement; filling both tables is unnecessary.

Earlier dependency checks might expose gaps; they do not guarantee that information arrives on time. A Try is not an action already taken. One review cannot establish a universal best practice.

Make the proposed change inspectable

Proposed action: check required interface information and availability before the next configuration cycle starts.

Acceptance has two layers. Checking that the list is usable examines the action's output. Checking subsequent delivery examines its effect. Completing a list does not establish better efficiency. Changes in scope, complexity, or staffing also limit causal conclusions from a before-and-after comparison.

Open diagram at full size

Personal improvement and follow-up record; facts, agreement, and judgments are supplied by you

Full original, before fixing the initial version:

Period, original target version and timing:
Actual results, denominator, evidence and unfinished work:
Gap:
Possible causes, contrary evidence and unknowns:
Reason for choosing KPT or YWT:
Practice to retain or learning (status and evidence):
One proposed change, purpose and output:
Owner, date and time, dependencies and effort cap:
Separate output acceptance from effect observation:
Approval, implementation and observed-result status:
Follow-up timing and continue/adjust/stop conditions:



Proposed; agreement pending

Follow-up draft not yet appended:



Execution unknown

Effect unmeasured

Next step undecided

Timestamps are browser recording times in UTC, not actual agreement or review dates; state actual dates and evidence locations in your text. Completing an action does not establish an effect, and before/after change alone does not establish causation.

A complete review update

At this week's agreed check-in, eight configuration units were planned and six were complete; two await interface information. The 12-unit overall task is not this week's denominator. The gap is two units, while full causes and the exact Friday demo impact are still being checked.

We propose confirming required information, contacts, and availability before the next cycle to expose dependencies earlier. The owner, actual deadline, and effort cap need agreement; the change is not approved or performed. First check the list and follow-up arrangements for unknowns. At the next agreed check-in, inspect delivery status and added effort before deciding whether to continue, adjust, or stop. Current material does not establish an efficiency improvement.

A short report retains the gap, explanation status, proposed action, and approval state. Keep the detailed list for questions. If the purpose is simply progress awareness, start with the concise-update example without adding an unnecessary improvement proposal.

Complete a review of your own

Choose a real period and retain the target version, actual results, evidence, and unfinished work. Select one change worth the effort and identify observations that could support or challenge it. Paste a complete retrospective or plan above, select one action, and label agreement according to the actual response. Retain that initial version, then append later agreement, plan changes, observations, and unfinished work as follow-ups. Earlier entries stay intact. The practice pack remains available. The reporting tool's review mode lets you organize results, explanations, learning, and plans manually. Include the owner, effort and approval status, leaving unagreed details open.