Suppose you find yourself in a situation at work where you’ve started many things but order isn’t everywhere. This happened because you were juggling too many tasks. It’s good—especially if you’re a manager or want to become one.
At some point, you’ll likely feel the need to get your affairs in order. Where will you begin?
The approach that is needed now differs from the one you use for everyday work. Simply writing down a plan with key points isn’t enough here.
Use a technique from child psychology to switch into superhero mode.
In my mother’s “board book,” I once came across two examples. They’re about how results depend on motivation.
In the first test group of 4–5-year-olds, they were simply asked to jump for distance as far as possible. Results were good.
For the second group, the task was set differently. They were invited to imagine that they were participating in the Olympic Games where athletes break records. The results were outstanding. Children tried much harder and jumped significantly farther on average than those in the first group did.
Another example involved solo assignments when children were tested individually at different times.
A child was asked to stand still for as long as possible without moving. On average, they lasted about one minute before losing focus and relaxing.
The second version asked the child to imagine they were a guard on watchtower duty, with the broom in their hand serving as a rifle. They were told not to move. Typically, test subjects managed four minutes without moving.
Before you start planning, imagine that your plan will determine how others work.
Seriously, don’t fall back on “we’ve always done it this way” or “it’ll do just fine.” If all managers ignored employees’ interests at some point, they’d be left alone. Managers optimize costs, and the system can slide into knee‑jerk workarounds. That works only so long as everyone is present. What if someone goes on vacation?
![In a room lit by natural light, a middle-aged man carefully explains something to a younger man while holding a document. (AI-generated image caption)|690x394]
down the hall:
- “Hey Boss, I need time off!”
- “When?”
- “Sorry… what day or date?”
It’s your turn to go on vacation. Even if everything is great in terms of responsibilities at work, you should occasionally ask yourself questions (write them down with the ending “in case I’m absent”):
-
What areas do I cover and who can step in?
-
Who are my direct reports and who will support their work?
-
Have I covered everything so that my boss won’t be stuck doing my work and making mistakes?
-
Which of the current (mine/team) tasks are most critical? Is there a documented process for solving them and am I using it now? What’s the cost of error on each one?
-
How likely is it that no one will continue my work, leaving me to clear up backlogs in addition to new urgent tasks, and then have to return to personal matters, essentially falling far behind?
-
Am I handing off work in a way others can understand? (code, description, plan, contacts, dependencies, current blockers, goal and where I am on the path toward it + next step)
The last point is very important. By approaching it carefully, you reduce problems for other tasks. That’s why I advocate emphasizing timely documentation. By “documenting,” I mean any record of the state that can be easily found and understood clearly. Don’t forget about small details either—people may not have access to your docs or a compatible format (for example, because you prefer Macs and Markdown while an analyst prefers Word).
Give extra attention to the penultimate question—it might prompt you toward organizing so you close tasks at ~80% completion in your own scoring system.
Example. You’re building a pipeline for automated code checking with AI on Gitlab’s Merge Request. You implemented it such that the AI’s response appears as plain text in runner logs, which is 80% done. Your next step is to convert that text into a comment attached to the bug tracker ticket. Stop. Think about how users will interact and about side effects. Likely, code review re‑runs multiple times by developers just to confirm the AI has been “warmed up” and the answer didn’t differ from earlier runs because of an incorrectly set temperature setting. Every such run triggers a message on that task, which in turn sends emails to all participants. I don’t think that’s right. Pause and close the task there, leaving a comment explaining your fix (in the ticket or docs—whatever you decide is best for your project).
So “tidy up by starting with the most important item” isn’t much of an answer. Dig deeper. Here’s my actionable guidance:
- Document early and often what was discussed; summarize meeting outcomes or note why a meeting didn’t happen.
- Add tags (if your tool supports them).
- Copy messages, screenshots, and images from messengers/emails to the task so they’re not lost later (instead of just linking, copy text/images).
- Attach files and screenshots; avoid losing sources in the future by preserving snapshots rather than links alone.
- Use OCR tools to pull text and audio from conference calls (use AI assistance).
This will help you reconstruct decision history over time.
Add permanent source references and documentation each time—don’t be afraid of repetition.
Here’s a quick checklist for the next step:
- Log the discussion points as soon as they’re made.
- End meetings with clear “what was decided.”
- Add tags where appropriate.
- Save copies of messages, screenshots, and images in the task (copy text/images; don’t just link).
- Attach files and screenshots; preserve them rather than relying on external links that may break.
- Use OCR/AI tools to extract notes from video calls.
This will make it easier to recover past decisions.