One trusted task list: capture every work task into a single list
The fix for tasks scattered across flagged emails, chat saves, sticky notes and memory is not a better app. It is one list, in one place, that every work task enters the moment it appears, and that gets sorted once a day. A list earns the word trusted only when it is complete: the moment some commitments live in your head instead, your head keeps tracking all of them, including the ones that are written down. What follows is the migration runbook: audit where your tasks are hiding, move every input into the single list, run the daily sort, and know in advance the point at which a neglected list should be abandoned and rebuilt rather than patched.
Why completeness is the whole mechanism
The best known version of this idea is David Allen’s Getting Things Done, whose five steps the official GTD site gives as capture, clarify, organize, reflect and engage. Capture comes first and means consistently identifying everything that has your attention. The research underneath it is less famous and more useful. In experiments reported by Masicampo and Baumeister in 2011, unfulfilled goals kept intruding on unrelated work: participants showed intrusive thoughts while reading and performed worse on an unrelated task. Formulating a specific plan for each unfinished goal eliminated those effects. A captured task with a defined next step is exactly that, a specific plan, and the same studies add the condition that makes or breaks a task list: the quieting effect was strongest in the participants who actually executed their plans. A list full of items you do not believe you will do does not quiet anything.
The second leg is cognitive offloading. Research summarized in a 2025 Memory paper describes saving information to a computer or phone as offloading that reduces processing demands, and cites an experiment in which saving a first list of words improved memory for a second list: storing the old material freed resources for the new. Your list is that external store. It only earns the freed resources if it is the store you actually check, which is why the runbook below spends more time on inputs and sorting than on choosing the tool.
Step 1: audit where your tasks are hiding
Before choosing anything, spend twenty minutes listing every place a task can currently reach you. For most office workers the audit finds the same six hiding places. Each one loses tasks in its own way, and each one gets a capture rule in step 3.
| Hiding place | What gets lost there | Its job after the migration |
|---|---|---|
| Flagged or starred email | Tasks buried under newer mail; the flag means both “task” and “interesting” | An input only: anything flagged becomes a list item the same day |
| Chat saves and direct messages | Requests with no owner check; scrollback swallows them within a day | An input only: a request is captured before the conversation moves on |
| Sticky notes and notebook margins | Anything written more than a week ago; notes do not survive a desk tidy | Retired, or the notebook becomes the list itself (step 2) |
| Memory and verbal asks | Whatever was asked while you were carrying coffee; the highest loss rate of all six | Nothing. Verbal asks get written down before the conversation ends |
| The team project board | Your personal commitments, which the board never held in the first place | Stays the team’s record; your next actions from it live on your list |
| The calendar | Undated tasks parked on random days until the day is wrong | Time specific commitments only; everything else belongs on the list |
Step 2: pick the one list, and close the others
The list can be an app or a paper notebook. The selection test is not features; it is capture speed and presence. Can you add an item in under ten seconds, from your desk, from a corridor, from home? Will you open it every morning without being reminded? A task app passes the test for most people, and Todoist is the worked example here because its capture is built for exactly this: its Quick Add takes a task typed in natural language, dates, labels and reminders included, its voice capture turns a spoken ramble into structured tasks, and on a phone it offers a quick add tile and lock screen widgets, with the same list synced across phone, desktop and browser. The app is a free download, so the honest cost of the system is not the subscription decision. It is maintenance time: roughly ten minutes a day for the sort in step 4 and thirty minutes a week for the review, set against the cost of the dropped task you no longer drop. A paper notebook passes the same test if it lives on your desk, travels to every meeting, and is the only notebook. What fails the test is two lists, one digital and one paper, each holding half the truth.
Then close the other doors. Unflag the old emails once their tasks are captured. Delete the notes app list. Tell the two or three people who assign you the most work where requests now land: “send it however you like, it all ends up on my list the same day.” The closing matters because a trusted list is a monopoly. Every surviving side channel is a place a task can be that the list does not know about.
Step 3: one capture rule for every input
- Capture at the moment of contact. A task enters the list when it reaches you, in the meeting, in the corridor, in the minute after reading the message. Not that evening, reconstructed from memory: the evening reconstruction is where memory files its losses. If capture would derail the moment, a single holding note on the list’s inbox is enough; detail gets added at the sort.
- Write the next action, not the project. The GTD method is blunt on this point: individuals must define their own next actions for whatever has their attention. “Quarterly report” is a project and will sit there glowing. “Pull the sales figures for the quarterly report” is an action and can be done, scheduled or delegated. Capture can be rough; the sort in step 4 is where rough items get their verbs.
- Under two minutes, do it instead of listing it. Tracking a task costs more than the task at that size. The exception is a task someone is waiting to see acknowledged: reply first, then there is nothing to capture.
- One inbox inside the list. Every capture lands in the same intake view, unprocessed. Sorted projects and dated views are outputs of the sort, never places you capture into directly, because deciding where an item belongs at capture speed is how items get filed wrong and lost politely.
Step 4: the daily sort, ten minutes
- Empty the intake to zero. Every captured item gets one of five outcomes: do today, schedule this week, waiting on someone, someday, or delete. Deleting is a real outcome. An item you would not act on this month is not a task, it is inventory, and inventory is what makes a list stop matching reality.
- Check it against tomorrow. Open tomorrow’s calendar and confirm the list and the day agree: time specific commitments on the calendar, everything else on the list, nothing important living in both or neither. The official GTD framing calls this reflect, a review process that keeps your picture of your work current, and it is the step that keeps trust from decaying between weekly reviews.
- Pick the day’s short list from the one list. Choosing what to engage with, in GTD’s fifth step, is a selection from the full inventory, not a second list written on a sticky note. Star, pin or drag the chosen few inside the same system.
Once a week, give the whole list thirty minutes: every project still open has a next action, every waiting item has a date to chase it, every someday item is still wanted. Software holds a system like this well only when its inputs stay clear, current and complete, which is a fair summary of what the weekly review is for.
The kill criterion: when to abandon the list and rebuild
A list is a model of your work, and models go stale. Patch a stale list and you get a tidier stale list. Use three measurable triggers; any one of them means rebuild, not repair:
- The stale share. At the weekly review, a third or more of open items have no next action, no date, and have not been touched in two weeks.
- The shadow system. You catch yourself tracking a real commitment in memory, on paper, or in a chat thread because “the list is a mess right now.” The shadow system is proof the trust is already gone.
- The miss. A task that mattered was dropped, and it was dropped while living outside the list or buried inside it.
The rebuild is one sitting, about an hour. Start a fresh list, empty. Re-run the step 1 audit and capture everything currently live from every source, including a full read of the old list, keeping only items you can give a next action and a reason. Then delete the old list outright rather than merging it, because merging imports the staleness you are trying to escape. Expect to carry over less than half of it. That is not waste; that is the audit finding, in numbers, how far the old list had drifted from the work.
What success looks like: someone asks what you owe them and the answer comes from one screen, not from a search across four apps and a hopeful memory; nothing you have promised exists only in your head overnight; and the morning question changes from “what am I forgetting?” to “which of these first?”, which is a much cheaper question.
General information only: this article describes a personal work organization method. Employer rules on approved apps and data handling come first; if your workplace restricts task tools, the paper version of the same system is the compliant route.
— Sterling Cole, Money Writer — Earning Desk