Guide
Use the Eisenhower matrix every day on a short list, not once on everything. Spend five minutes sorting about ten tasks on paper. Put the important-but-not-urgent work on your calendar first, give urgent-but-minor requests a fixed window, and drop the rest. Then compare the plan with what actually happened. A paper log is enough; on a Mac, Dayflow can keep that record automatically.
Monday morning you draw the four boxes. Urgent and important, important but not urgent, and so on. You sort twenty-odd tasks into them, and for an hour the page feels like control. The proposal draft sits in the “schedule” box, the expense report in “do now,” the newsletter in “drop.”
By Wednesday the sheet is under a coffee cup. Something came in from your manager, a client email needed an answer, and the day filled up the way it always does. The proposal is still in the schedule box, unscheduled. You are not sure whether the matrix failed or you did.
Neither, mostly. The matrix is a sorting idea, and sorting is the easy part. This article covers why the method tends to stall after a few days, six concrete things to try (most need only paper, a timer and a calendar), and how to check after a week whether any of it changed where your time went.
A note before we start: Dayflow, a Mac app we make, publishes this article. It comes up once, near the end. Everything else here needs no software.
The four boxes are simple:
| Urgent | Not urgent | |
|---|---|---|
| Important | Do it now | Schedule it |
| Not important | Contain or delegate it | Drop it |
The trouble is not in the table. It is in how days actually feel.
Urgency is loud and importance is quiet. An email with a red flag, a message that says “quick question,” a meeting invite for this afternoon: these all announce themselves. The work that matters most over a month, like planning, learning, or fixing the thing that keeps breaking, never sends a notification. Many people find that whatever is loudest gets their attention, regardless of which box it was sorted into.
“Important” drifts toward “stressful.” Without a written definition, it is easy to file something as important because thinking about it makes your stomach tighten. That puts half the list in the top-left box, and a matrix with everything in one box is just a to-do list with extra lines.
Sorting feels like doing. Moving tasks between boxes gives a small sense of progress. It is real work, but it is not the work. The matrix says “schedule it” for the second box, and scheduling is the step most often skipped.
Nothing checks the plan against reality. Most people sort in the morning and never look back. Without some record of where the day went, you cannot tell whether the matrix shifted your time or only your intentions. If you have ever ended a day wondering where your day went, you have felt this gap.
None of this is a character flaw. It is how attention works for a lot of people under normal workplace pressure. The fixes below aim at those four gaps.
What to do. Each morning, write down no more than ten things you might do today. Not everything you could ever do, just today’s candidates. Draw the four boxes on a sticky note or index card and sort them. Set a kitchen timer for five minutes so the sort cannot become its own project.
Keep the long backlog somewhere else. The matrix is a daily filter, not a storage system.
Illustration. A freelance designer writes eight items on an index card: client revisions due tomorrow, invoice follow-up, portfolio update, reply to a friend’s request for feedback, tidy the downloads folder, research a new tool, call the accountant, sketch ideas for a personal project. Revisions and the accountant call go top-left. Portfolio and personal sketches go top-right. The friend’s feedback request and the invoice follow-up go bottom-left. Downloads folder and tool research go bottom-right.
How to tell after a week. You should have five or so small cards. Look at them together. If the sort took under five minutes most days and the cards mostly hold today-sized items, the habit is set. If you kept writing fifteen items or more, the list is still the backlog, so split them.
What to do. Before your next sort, write two one-line tests and stick them where you sort.
Name your three goals on the same note. If you cannot name them, that is the first quadrant-two task.
Illustration. Someone has “reply to the all-hands thread” filed as urgent and important. Against the tests: it does not move any of their three goals, and nothing happens if it waits two days. It moves to the bottom-right box. Meanwhile “outline the Q1 budget,” sitting untouched, passes the importance test easily and gets promoted.
How to tell after a week. Count how many items landed in the top-left box each day. If it used to be most of the list and is now two or three, the definitions are working. If it is still crowded, tighten the tests. “Real consequences” might need to become “someone else is blocked.”
What to do. Quadrant two is where the matrix earns its keep, and it only works if those tasks get actual time. Open your calendar at the start of the week, before you accept anything new, and block one or two sessions for important-but-not-urgent work. Treat them like a meeting with someone you respect. Give each block one named task, not a vague “deep work.”
If you have never tried fixed blocks, our guide to timeboxing goes deeper on sizing them.
Illustration. A product manager blocks Tuesday and Thursday from 9 to 10:30, labeled “Draft pricing proposal.” When a colleague asks to meet Tuesday at 9:30, the calendar already says no, so the request moves to the afternoon without any agonizing.
How to tell after a week. Count how many of your quadrant-two blocks actually happened as planned. Two out of two is great. One out of two is a real start. Zero means something is overriding them. Usually that is meetings, which suggests booking the blocks earlier in the day or earlier in the week, before the calendar fills.
What to do. The bottom-left box is where most days quietly go: small requests, replies, admin that someone else wants now. Many people have no one to delegate to, so “delegate” turns into “do it immediately.” Instead, give these tasks one fixed window per day, say thirty or forty-five minutes after lunch. Run a timer. When it rings, whatever is left waits until tomorrow’s window.
For things you could hand off, write a two-sentence script so delegating costs less than doing: “Could you take this one? Here is the link and what done looks like.”
If meetings are the main thing crowding this box, the meeting cost calculator is a quick way to put a number on a recurring one before you decide whether to keep attending.
Illustration. A developer notices that code review requests, Slack questions and calendar shuffling arrive all day and each one pulls them out of a task. They set 1:30 to 2:15 as the “small stuff” window, keep a sticky note of anything that arrives outside it, and clear the note in the window.
How to tell after a week. Ask whether the small requests got handled about as well as before. Usually they do, because few of them needed a reply within the hour. Then ask whether your mornings felt less chopped up. If people complained about slow replies, widen the window or add a second short one late in the day.
What to do. Tasks you drop have a habit of coming back, because nothing records the decision. Keep a short running list titled “Not doing, on purpose” with the date next to each item. When the item resurfaces in your head, check the list and leave it alone. Review the list once a month and delete anything that no longer nags.
Illustration. “Rebuild personal website (June 3).” “Learn a second note-taking app (June 3).” “Reorganize the shared drive (June 10).” When the urge to reorganize the drive hits on a slow Friday, it is already decided, and the decision costs nothing to keep.
How to tell after a week. Notice how often you caught yourself starting a dropped task. If the list made it easier to stop, keep it. If you kept doing the dropped tasks anyway, one of them may belong in quadrant two after all. Move it and give it a calendar block.
What to do. Once a week, hand your raw list to a friend, partner or colleague and ask them to sort it into the four boxes without your commentary. Then compare with your own sort. You do not have to accept their answer. The point is to see where your sense of urgency differs from an outsider’s.
Illustration. A teacher’s friend moves “redo the class newsletter template” from top-left to bottom-right and moves “plan next unit” from bottom-right to top-right. The teacher pauses on the second one. Planning felt optional because nobody was waiting on it, which is exactly what quadrant two looks like.
How to tell after a week. If you changed at least one item based on their sort, it was worth the ten minutes. If your sorts matched exactly, you probably have the definitions down and can skip this step.
What to do. This is the step most versions of the matrix leave out. At the end of each day, spend three minutes writing down roughly how your hours went, then tag each chunk with the box it belongs to. A plain notebook works. So does our daily work log template. At the end of the week, add up the rough hours per box.
If you would rather start with a quick estimate than a full log, the where does my week go calculator gives you a rough baseline to compare against.
Illustration. One person’s Friday tally reads: top-left about 9 hours, top-right 3, bottom-left 22, bottom-right 6. Their morning sorts looked fine all week. The log shows the bottom-left box ate most of the week. That points straight back to step 4, the container that was not holding.
How to tell after a week. You will have a small table you did not have before. The question is not whether the numbers look good. It is whether they surprise you. A surprise means the log is telling you something your memory was not.
Step 7 depends on remembering to write the log, and on remembering the day accurately when you do. Both are hard on a busy day. By evening, a lot of the bottom-left work has blurred together, and the log becomes a guess.
This is the one place software can make a real difference. Dayflow is a free, open-source Mac app that records your screen at one frame every 10 seconds and has AI write the day into a readable timeline. There is no timer to start and no day to reconstruct. At the end of the week you can read what was actually on screen and sort those blocks into the four boxes yourself.
Some specifics, from Dayflow’s own documentation as reviewed on 2026-09-06:
If you want to compare other approaches first, our roundup of Mac time trackers covers the main options.
Dayflow publishes this article, so weigh this section with that in mind.
Be clear about what it does not do, because several of these matter for this method.
The Eisenhower matrix is a fine idea that usually fails at the edges: sorting too much at once, fuzzy definitions, quadrant-two work that never reaches the calendar, small requests that spread across the whole day, and no check against what actually happened.
Start small. Sort ten items tomorrow morning in five minutes, using a written test for “important.” Block one session this week for something in the top-right box. Give small requests a single window. At the end of the week, look at where the hours really went, from a notebook or an automatic record, and change one thing. That loop matters more than how neat the boxes look.
The matrix sorts tasks on two questions: is it urgent, and is it important? Urgent and important tasks get done now. Important but not urgent tasks get a scheduled time. Urgent but not important tasks get delegated, batched or kept small. Tasks that are neither get dropped. The whole value is in the second box, the work that matters but never shouts for attention.
Daily, and briefly. A five-minute sort at the start of the day on a short list of around ten items works better for many people than a big weekly sort of everything. Priorities shift as emails and requests arrive, so a matrix drawn on Monday is often stale by Wednesday. Add a short weekly look back to see which box your time actually went to.
Quadrant 2 holds work that matters but has no deadline forcing it: planning, learning a skill, fixing a recurring problem, writing documentation, exercise, a hard conversation you keep postponing, the project that would make next quarter easier. These tasks rarely feel pressing on any given day, which is why they slide. The practical move is to put them on your calendar before anything else gets there.
Write a one-line test before you sort. For example: a task is important if it moves one of my three current goals forward or if skipping it causes real harm to someone. Anything that fails the test is not important, however loud it is. Without a written test, importance tends to drift toward whatever feels most stressful in the moment, and the matrix stops helping.
That usually means the definitions are doing no work. Ask what actually happens if each task waits two days. If the honest answer is mild inconvenience, it is not urgent. Another check is to have a friend or colleague sort your list without your commentary. An outside reader often sees quickly which items are anxiety and which are deadlines.
Yes. For solo workers, the delegate box becomes a contain box. Give urgent but minor tasks a fixed window, such as thirty minutes after lunch, and do them together with a timer running. You can also automate them, reply with a template, or ask the requester whether a later date works. The goal is the same: stop these tasks from spreading across the whole day.
After a week, look at how your hours actually split across the four boxes, not at how tidy the sorting looked. If the important but not urgent box got real blocks of time, it is working. If the week went mostly to urgent but minor requests, adjust the definitions or the containers. A simple end-of-day log, on paper or automatic, makes this check possible.
No. Dayflow records what was on your Mac's screen and has AI write a timeline of the day. It is not a place to keep or sort a task list. For the sorting itself, a sheet of paper with four boxes, a notebook or the task list you already use is the better tool. Dayflow is useful afterward, to check where the time actually went.
No. Dayflow does not provide website blocking. If your main problem is drifting into sites you have already decided to drop, a dedicated website blocker is the better choice. Dayflow can show you afterward that the drift happened and roughly when, but it will not stop it in the moment.
Yes. Dayflow is $0 and MIT licensed, and it runs on macOS 14 or later. An optional Pro plan at $20 a month has Dayflow manage cloud analysis. You can instead run local AI on your Mac through Ollama or LM Studio, which keeps screenshot analysis on-device, or connect your own provider, whose charges and privacy terms apply.
Dayflow is $0 and open source. Full recordings stay on your Mac, with local AI available for on-device analysis.
Download for MacOS macOS 14+ · MIT licensed · no account required