Impact vs Effort Matrix

What this tool does

CM8-256 answers one question: what should we do next, when everything on the list looks worth doing? You list the items, score each on two axes from 1 to 5, and the tool places it in a quadrant, ranks the board by priority score, and prints a running order you can hand to a team. The value is in asking the same two questions of every item — including the one the loudest person in the room brought — and then living with the answers.

Everything runs inside this single file: no account, no upload and no network request of any kind, so a candid assessment of what your organisation should stop doing never leaves this computer.

The two axes, and how to score them honestly

Impact is how much better things get if the item is done: time saved, errors avoided, risk removed, money earned, frustration ended. Use the same yardstick every time. A 3 should mean a clear benefit somebody would thank you for; a 5 should be rare and should change something structural. If half the board scores 5 the scale has collapsed — rescore, or raise the threshold in Settings.

Effort is your effort, not the ideal-world estimate. This is where most matrices go wrong. An honest effort score covers everything between now and done: the specification nobody has written, the two people who have to agree, the approval that takes three weeks, the supplier who does not answer, the retraining, and the day it goes wrong and has to be redone. A change that is four hours of actual work but needs sign-off from three departments is not a 1. Score what it will cost you to get this over the line in the organisation as it actually behaves, not the one you wish you worked in. Teams that do this are startled by how many of their quick wins move a quadrant to the right — and that movement is the most useful thing the exercise produces.

The four quadrants

The threshold in Settings — 3 by default — divides both axes. High is at or above it; low is below it.

Quick wins (high impact, low effort). Do them this week — not next quarter, not when the project finishes. Their real value is often credibility rather than the saving: three visible improvements delivered quickly is what buys permission to spend six months on a major project.

Major projects (high impact, high effort). These deserve to happen and will not happen on enthusiasm. Each needs a sponsor with authority to remove obstacles, a plan, a date and a budget — and a business case of its own, because a row on a board is not one. The common failure is starting three at once. Pick one, resource it properly, and let the others wait here in plain sight.

Fill-ins (low impact, low effort). Genuinely useful, genuinely minor. Use them to fill gaps — the afternoon a machine is down, the week a project is blocked. A fill-in must never displace a quick win. They are attractive precisely because they are easy, which is how a team ends up busy, cheerful and going nowhere.

Thankless tasks (low impact, high effort). The hardest discipline in the method is actually dropping these. They survive because someone senior asked, because removing them feels like failure, or because half of one is already done. Say out loud that it is not going to happen, mark it dropped, and record why — a dropped item with a reason does not come back next quarter as a fresh idea.

One honest exception: some thankless tasks are mandatory. Compliance work, statutory records, mandated inspections and contractual obligations often score low on impact and high on effort, and must be done anyway. Mark those agreed, with an owner and a date, and stop scoring them — ranking a fixed obligation against optional work is a category error.

The Eisenhower preset

The axis preset in Settings switches the labels to the classic urgency-and-importance grid, and the quadrants become Do now, Schedule, Delegate and Drop. Be clear about what it does: the arithmetic is identical. Same thresholds, same quadrant rules, same priority score. It relabels the axes and renames the quadrants — nothing you could check with a calculator changes.

To use it, read the first column as importance and the second as how soon it must be done: 1 means today, 5 means no deadline pressure. Scored that way the quadrants land where the classic grid puts them — important and immediate is Do now; important with time in hand is Schedule; unimportant but pressing is Delegate; the rest is Drop.

The original's warning still applies, and is the whole reason the grid exists: urgency is the enemy of importance. Urgent work announces itself; important work waits quietly and never becomes urgent until it is too late to do well. A board full of Do-now items with an empty Schedule quadrant is a reactive board, not a prioritised one.

Cost, benefit and payback

Where you have real figures, enter them: the cost of doing the item and the benefit you expect in a full year. The tool divides one by the other for a payback in years, and totals the benefit of everything agreed or in progress.

Where you do not have figures, leave them blank. A blank is honest; an invented number is not, and it will be quoted back at you with a confidence it never earned. Plenty of good improvements have benefits that are real and not measurable in money — fewer errors, less frustration, a job that is safer or less unpleasant. Those belong on the board with the money columns empty. Enter cost and benefit together or not at all: a payback built from half a pair means nothing.

Running a 30-minute prioritisation session

The method works best out loud, with the people who will do the work.

• Beforehand. Load every item with a name and a one-line description. Score nothing yet.
• First 5 minutes. Agree what a 5 means on each axis, using an example everybody recognises. Without that anchor you will spend the session comparing optimism.
• Next 15 minutes. Score each item in about a minute, without debate. Where a score is contested, take the more pessimistic effort score and move on.
• Last 10 minutes. Read the quadrants together. Give every quick win an owner and a date, pick one major project to plan properly, then take the thankless quadrant item by item and either drop it or say plainly why it must stay.

Repeat quarterly, not weekly. This is a planning instrument, not a task tracker.

Keeping the board short

A backlog of sixty items is a museum: nobody reads it, nothing is ever removed, and its length gets used as evidence the team is overloaded rather than unfocused. Twenty items is a working board. Deferring is not deciding — if an item has survived three review cycles without moving, the organisation has already decided not to do it. Mark it dropped and let the board tell the truth.

Where it sits alongside other methods

This matrix sequences work; it does not choose between competing solutions to one problem. To separate what must ship from what would be nice within a fixed scope, MoSCoW is the better instrument. To pick one option from several against criteria that pull in different directions — three machines, four suppliers, two premises — use a weighted decision matrix, which scores every option against every criterion instead of reducing each to two numbers. Impact versus effort is for when the items are not alternatives at all: they are all worth doing, and the only question is the order.

The formulas

Payback appears only when both cost and benefit are present and the benefit is above zero; otherwise the cell reads as a dash. Note what the score cannot do: it is a ratio, so 4 over 2 and 2 over 1 both score 2.00 although one delivers twice the benefit. Read it alongside the quadrant, never instead of it.

FAQ

Everything is scoring 4 or 5 for impact. The scale has collapsed, usually because people are scoring how much they want the item rather than what it delivers. Rescore against a fixed anchor, or raise the threshold to 4 so the quadrants separate something again.

Can I keep several boards in one file? Yes — give each a board name spelled identically on every item, then use the search box.

Two items have the same score. The matrix cannot separate them and should not pretend to. Break the tie on something real: which has an owner, which has a date, which the team will finish.

Why does an item show no payback? Cost or benefit is blank, or the benefit is zero — a payback from half a pair would be a confident wrong number.

Does switching the preset change my scores? No, only the labels. Reread the second column though, because the two presets ask you to score it in opposite directions.

Saving your work

Items, settings and the report header are written to this browser's local storage as you type. That storage belongs to one browser on one computer: another browser, a private window or a clean-up tool that clears site data will not have it.

Treat Export .json as the real save — one file containing everything, which Import .json restores anywhere. Export CSV gives you the board for spreadsheet work. Reset asks twice, then erases everything this tool has stored. There is no undo.

Accuracy & disclaimer

Every number here derives from two judgements made before the work is done, by people with an interest in the answer. Impact is an expectation; effort is an estimate, and estimates of effort are wrong more often than they are right. An item can change quadrant on one point of either score, so treat a position near the threshold as "roughly here", not as a verdict.

What the matrix offers is consistency rather than accuracy — the same two questions asked of everything, so a list of opinions is at least sequenced the same way each time. That is better than sequencing them inconsistently, and not the same as being right. Nothing here is a business case or a budget approval.