What to Review First: Promises or Unfinished Personal Tasks?
If you are deciding what to inspect first in a weekly review, start with promises to other people. Check what you agreed to deliver, when it is due, and whether anything has changed. Then review unfinished personal tasks. This order is a practical way to surface time-sensitive expectations early; it does not mean every promise should be executed before every personal task. Once you have reviewed both groups, choose what to do by deadline, consequence, dependency, and available time. The distinction matters: a weekly review is a pass through your commitments and task records to make them accurate. It is not a ranking of everything you will do next week. The Getting Things Done (GTD) method describes review as one of its five steps, after clarifying and organizing, and recommends reviewing relevant system contents ([GTD’s overview of its five steps](https://gettingthingsdone.com/what-is-gtd/)). That is a useful basis for checking the whole picture before acting on it.
Why begin with promises to other people?
An unfinished personal task may matter a great deal, but a promise to another person can carry a date or expectation that exists outside your own list. If you inspect it first, you can quickly identify a due date that is near, a delivery that is blocked, or an agreement that needs to be renegotiated. This is a decision rule for the review order, not a claim that all external commitments are more important than personal work.
For this purpose, define a promise broadly but concretely: an explicit agreement, a deliverable someone is waiting for, or a follow-up you told someone you would make. Capture the person or group, the promised outcome, the date or timing, and the next action. “Send the draft to Lee by Thursday” is reviewable. “Keep Lee happy” is too vague to guide a next step.
GTD’s official weekly-review material presents review as a complete process rather than a single scan of due dates. Its June 2025 page describes an 11-step Weekly Review ([GTD, “Episode #314: Weekly Review Best Practices”](https://gettingthingsdone.com/2025/06/weekly-review-best-practices/)). The source does not prescribe “promises first” as a universal sequence. The order here is editorial judgment built around the practical value of seeing other people’s time-bound expectations early.
What does “inspect promises first” look like?
Use a short pass through messages, meeting notes, and your task system—whatever records where you made agreements. Do not try to solve every item as you find it. Confirm what the commitment is, whether it remains active, and what information would make it actionable.
For each promise, ask:
- What exactly did I agree to provide or do? - Who is expecting it, and by when? - Is it still valid, or has the other person changed the request? - Is it on track, blocked, at risk, or already complete? - What is the next visible action: draft, ask a question, schedule a meeting, or send an update?
If a due date is uncertain, mark it as uncertain and verify it; do not silently invent a date. If the agreement no longer fits, record the conversation needed to clarify or renegotiate it. A review should expose the state of the commitment, not turn a guess into a new promise.
Then review unfinished personal tasks
Next, scan personal tasks that remain open. The aim is to decide what each one means now, not to carry every old item into another week automatically. Is it still wanted? Is it waiting on something? Does it belong to a larger project? Is it a specific next action, or only a broad intention?
GTD’s official overview separates clarifying from organizing: decide what an item means, including the next action if it is actionable, then place reminders in appropriate locations. It also describes reviewing relevant system contents as a recurring step ([GTD’s five-step overview](https://gettingthingsdone.com/what-is-gtd/)). For an unfinished personal task, that suggests a useful cleanup sequence: make it actionable, move it to a waiting or later list if appropriate, keep it as a project with a next step, or remove it if it no longer matters. These are review decisions; none requires doing the task during the review.
Avoid treating “unfinished” as proof that an item deserves another week. A task may be stale, superseded, or dependent on a decision you have not made. If you keep it, make the next action clear. “Sort the tax folder” could become “Find the missing receipt from March.” If you cannot identify a next action, note the question or choice that is blocking it.
Review order is not execution priority
After both passes, compare tasks for execution. A promise due tomorrow might rise to the top; a personal task connected to a fixed appointment or major dependency might outrank a promise due later. A task that can be finished in a few minutes may still be less important than a longer action whose delay blocks someone else.
Use four prompts to make the next-week choice:
1. **Timing:** What has a real due date or a near-term window? 2. **Consequence:** What becomes harder, costlier, or less useful if delayed? 3. **Dependency:** Is someone else waiting, or does this unlock another step? 4. **Capacity:** What can reasonably fit alongside fixed commitments and ordinary work?
This is an editorial decision aid, not a scoring formula. If two items compete, write down the trade-off rather than relying on the order in which they appeared in your list. Weekly-review guidance from Todoist likewise frames a review as reflection on the past week and preparation for the next, and says its checklist structure can be adapted to the user’s needs ([Todoist, “The Weekly Review: A Productivity Ritual to Get More Done”](https://www.todoist.com/productivity-methods/weekly-review)). That supports tailoring a review process, but it does not establish a single priority rule for every reader.
A worked example: a due promise and a personal task
Suppose you promised to send a colleague a two-page draft by Wednesday, while “replace the hallway light” has remained on your personal list for two weeks. During the review, check the agreement first: confirm Wednesday is still the expected date, note that you have a draft to finish, and identify the next action. Then review the light task: perhaps the bulb is already purchased, so the next step is to install it; perhaps you no longer need to replace it.
These details are illustrative, not reported results. When choosing what to execute, the draft may come first because another person is waiting and the date is near. But if the hallway light must be replaced before a scheduled inspection, the personal task could have the stronger immediate constraint. The review order helped you notice both facts; it did not settle the priority by itself.
When should a different review order make sense?
Start with a personal task when there is an urgent safety or household issue, a fixed deadline, or a task that unlocks several commitments. Likewise, if promises are already tracked reliably in one place and personal tasks are scattered or easy to forget, begin wherever a quick scan is most likely to reveal missing information. The point is to inspect both categories every week, not to follow a rigid ritual while important records remain out of view.
For a simple recurring routine, try this sequence:
1. Scan promises to other people; verify outcomes, dates, status, and next actions. 2. Scan unfinished personal tasks; clarify, reschedule, delegate where appropriate, or remove stale items. 3. Look at the coming week’s fixed dates and dependencies. 4. Select actions by timing, consequence, dependencies, and realistic capacity. 5. Send any needed updates or clarification messages after the review, once you know what you can actually commit to.
Keep the two questions separate: “What must I make accurate in my system?” and “What should I work on first?” Promises to other people are a sensible first stop for many weekly reviews because their dates and dependencies may be less visible if you only scan your own task list. Reviewing personal tasks immediately afterward prevents that first pass from becoming the whole review. The final action order depends on the specific week.