The Eisenhower matrix does not work, and what to do instead

Filed under Guide

By Gerald · 27 August 2026

Two sticky notes, one orange and one yellow, placed on a plain wall

I have drawn the four-box grid on a whiteboard more times than I can count, usually during a team offsite, always with real enthusiasm. Within a week, almost every task in the room had drifted into the top-left box. Urgent and important. The other three boxes sat empty, like furniture nobody uses.

That is not a personal failing. It is what the matrix does to real work.

The short version: the Eisenhower matrix asks you to sort tasks by urgency and importance, but importance is not a property a task has on its own. It only exists relative to a goal you were supposed to have already picked, and the matrix never asks you to pick one. Replace the two axes with one question and a hard cap on active work, and the sorting problem mostly disappears.

A grid with two axes gives you the feeling of a decision without the decision itself.

What the matrix is meant to do

The idea traces back to a 1954 speech where Dwight Eisenhower quoted an unnamed college president: "I have two kinds of problems, the urgent and the important. The urgent are not important, and the important are never urgent." Eisenhower used that distinction to explain how he ran his own schedule. He did not draw a four-quadrant grid.

The grid came later. Stephen Covey formalized it in "The 7 Habits of Highly Effective People" in 1989, adding the labels most people now recognize: do first, schedule, delegate, delete. It is a genuinely useful teaching tool for one specific lesson, which is that busy and important are not synonyms. That lesson is worth learning once. It does not need a grid you redraw every Monday.

The pitch is simple. Plot each task on two axes, urgent and important, and the box tells you what to do with it. Do the top-left box now. Schedule the top-right box. Delegate or shorten the bottom-left box. Delete the bottom-right box.

On paper it sounds like a filter. In practice it is closer to a mirror. It shows you what you already believe about your workload, sorted into four bins, without changing anything about how you decide what belongs where.

Why almost everything lands in one quadrant

Empty wooden pigeonhole mail slots partly obscured by a green plant — Sorting into a fixed number of slots works when the categories are stable. Tasks rarely are.

Try the exercise honestly with a real list of open tasks. Client work is important because it pays the bills, and urgent because someone is waiting on it. The email from your manager is urgent because it is sitting in your inbox, and you convince yourself it is important because ignoring your manager feels risky. The thing you have been avoiding for three weeks is now urgent purely because you delayed it, which retroactively makes it feel important too.

The two axes are supposed to be independent. In real inboxes they are not. Urgency contaminates the importance judgment almost every time, because a deadline creates a felt sense of stakes that has nothing to do with actual consequence.

The result is a grid where one box holds 80 percent of the list and the other three hold almost nothing. At that point the tool has stopped doing any sorting work. You are back to a single undifferentiated pile, just with a fancier name for it.

The missing step: deciding what matters first

Here is the part the matrix skips. Importance is not something a task carries by itself. A task is important only in relation to something you have already decided you care about this month, this quarter, or this year. Without that reference point, "important" collapses into "feels significant right now," which is exactly the trap the matrix was supposed to prevent.

The fix is not a better grid. It is doing the work the grid assumes you already did: naming the two or three things that actually matter right now, before you touch a single task. If you have not named them, no amount of quadrant-sorting will produce a real answer, because there is nothing to sort against.

This is also why the matrix feels most useless during the busiest weeks. When everything is on fire, you have the least mental space available to step back and ask what you are actually optimizing for, which is precisely the moment the grid needs that answer most.

One question instead of two axes

Once you have named what matters, prioritization stops being a two-axis plot and becomes one question per task: does this move one of the two or three things I actually care about right now, or not?

That's it. Not urgent-and-important, not delegate-or-delete. A yes or a no, applied to a short list, decided at the moment you are choosing what to work on next rather than during a weekly sorting ritual that then goes stale by Wednesday.

Tasks that pass the test go into active work. Tasks that fail it go somewhere else, not because they are worthless, but because they are not what this week is for. Some of them will matter later. Most of them were never going to get done regardless of which quadrant they sat in.

This is deliberately less sophisticated than a four-box grid. That is the point. A method you actually use beats a method that produces a satisfying diagram once and then gets abandoned.

Capping what is in progress

The other half of the fix has nothing to do with sorting and everything to do with limiting. A quadrant has no ceiling. You can pile fifty tasks into "urgent and important" and the grid will hold every one of them without complaint, which is exactly the problem: it never forces a decision about capacity.

A board with a hard cap does. I run FlowBoard with three columns, and the middle one, Executing, is the one that matters here: it is meant to hold only what you have actually committed to working on now, small enough to see at a glance without scrolling. When it fills up, the honest move is to finish something before adding anything else, not to widen the column. Flow does not enforce a numeric limit on that column for you, so the discipline is still yours to keep, but the shape of a short middle column makes it obvious when you have quietly stopped being selective.

That constraint does more prioritization work than a second axis ever did. A quadrant tells you where a task theoretically belongs. A cap tells you when you are lying to yourself about how much you can actually do this week.

When the matrix is still the right tool

I want to give the matrix its due. As a one-time teaching exercise, it is genuinely good at making a specific point: busy is not the same as important, and a lot of what feels urgent is really just loud. If you have never internalized that distinction, drawing the grid once, honestly, with real tasks, can be a useful jolt.

It also still works reasonably well for big, slow-moving decisions, like quarterly planning or deciding which of three initiatives a team should staff first, where the number of items is small and the time horizon is long enough that urgency and importance genuinely do separate. The failure mode shows up specifically at the pace of daily task management, where the list refreshes constantly and there is no time to re-litigate two axes for every new item.

Use it as a one-time lesson or a slow strategic exercise. Do not use it as your Tuesday morning task list.

Frequently asked questions

What are the four quadrants of the Eisenhower matrix? The four quadrants are urgent and important (do it now), important but not urgent (schedule it), urgent but not important (delegate it or shorten it), and neither urgent nor important (drop it). The labels come from Stephen Covey's 1989 formalization of a distinction Eisenhower described in a 1954 speech, which he attributed to an unnamed college president rather than claiming credit himself.

Why does everything feel urgent and important? Because urgency and importance are not actually independent in most people's inboxes. A deadline creates a felt sense of stakes, and that feeling gets mistaken for genuine importance. Without a prior decision about what actually matters this month or quarter, almost every task will read as both urgent and important, which is why the top-left box fills up while the other three sit empty.

What is a simpler alternative to the Eisenhower matrix? Name two or three things that actually matter right now, then apply one question to each task: does this move one of those things forward or not. That single yes-or-no filter, applied consistently, does the same job as the four-box grid without the false precision of two separate axes.

How many tasks should be in progress at once? Small enough that you can see the whole list at a glance and recall every item without re-reading titles. There is no universal number, but if you need to scroll or you keep forgetting what is on the list, it is too long. A capped board column enforces this better than a quadrant ever will.

Does prioritisation work if my deadlines come from other people? Yes, but the "important" judgment still has to be yours. Someone else's deadline determines urgency. It does not automatically determine whether the task moves something you actually care about. Keep both questions separate, and be honest that a deadline alone does not make a task important.

My verdict

The Eisenhower matrix teaches one real lesson well and then overstays its welcome. Learn the lesson once. Then replace the grid with a named priority and a hard cap on active work, and check the list against both every time you sit down to decide what to do next.