# Decision Matrix: The Five Types, and How to Build One That Holds Up Canonical URL: https://headwayskills.com/knowledge/decision-making/decision-matrix/ Markdown URL: https://headwayskills.com/knowledge/decision-making/decision-matrix.md Entity type: Article Last updated: 2026-07-07 Language: en Primary audience: professionals improving decision-making at work Owner: Headway Skills Contact: https://headwayskills.com/contact/ ## Short answer A decision matrix scores your options against weighted criteria. The five types, how to build one in six steps, and the two mistakes that rig the result. ## Key facts - Title: Decision Matrix: The Five Types, and How to Build One That Holds Up - Category: Decision Making - Primary skill: Decision-Making - Related skills: Influence, Teamwork - Primary keyword: decision matrix - Source page: https://headwayskills.com/knowledge/decision-making/decision-matrix/ ## What this page covers - A decision matrix scores your options against weighted criteria. The five types, how to build one in six steps, and the two mistakes that rig the result. - Practical guidance for decision matrix - How this topic connects to Decision-Making ## Detailed explanation You have three or four defensible options, half a dozen things that matter, and no way to say out loud why one of them wins. A decision matrix fixes that. It is a grid: your options as rows, the factors that matter as columns, a weight on each factor showing how much it counts, and a score in every cell. Multiply score by weight, total each row, and the highest total is your answer — along with a visible record of how you got there. What most guides leave out is that "decision matrix" names at least five different grids, and reaching for the wrong one is the first mistake. ## The five grids that all get called a decision matrix ASQ, the American Society for Quality, catalogs this tool under several names at once — Pugh matrix, problem selection matrix, selection grid. The Wikipedia entry adds grid analysis, opportunity analysis, criteria rating form, and multi-attribute utility theory. The naming is a mess; the underlying tools are not. Five distinct grids are in common use, and they answer genuinely different questions. Pick by the question you are holding, not by the phrase someone used in the meeting. ### The unweighted decision matrix Options down the side, criteria across the top, a score in every cell on one fixed scale — usually 1 to 5 or 1 to 10 — and a straight sum across each row. What sets it apart is that every criterion counts the same, which is both why it is fast and why it is usually the wrong choice. If cost matters three times as much as delivery speed on this particular decision, an unweighted grid insists they matter equally, and the total it hands back is not an answer to your question. Use it when the criteria really are comparable in importance, or as a [five-minute first pass](/knowledge/decision-making/pros-and-cons-list/) to see whether any option is obviously out. ### The weighted decision matrix The version nearly every search result means. Each criterion carries a weight for how much it counts, expressed either on a 1-5 scale or as percentages summing to 100. You score each option, multiply by the weight, and sum the row. Its distinguishing feature is that it makes you state, in numbers and in advance, what actually matters most. That is where most of the value sits. In practice the clarity arrives during the argument about weights, not at the total — deciding that reliability is worth twice what price is worth settles more than the arithmetic afterward ever does. It is also the variant most easily manipulated, which the section below covers. ### The Pugh matrix Named for Stuart Pugh, professor and head of design at the University of Strathclyde, who built it to choose between design alternatives; ASQ still lists the method under his name. The distinguishing move is the baseline. Rather than scoring every option in the abstract, you nominate one reference option — usually the incumbent or the status quo — and mark each alternative against it on every criterion: **+** for better, **S** for the same, **–** for worse. Then you count. For anyone junior, this is the most useful variant here, because relative judgments are far easier to defend than absolute ones. "Their support response is better than what we have now" survives being challenged across a table. "Their support response is a 7" does not, because the next question is always what a 7 means. Pugh matrices exist in weighted and unweighted forms; start unweighted. ### The Eisenhower matrix A 2x2 grid plotting [urgency against importance](/knowledge/time-management/prioritize-tasks/), sorting items into four quadrants — do, schedule, delegate, delete. It is named after Dwight D. Eisenhower, the 34th U.S. president. It gets filed under decision matrix constantly, and it is structurally a different instrument. The scoring grids above rank a few options against shared criteria; this one triages many items against two fixed axes. If the problem is "which of these three vendors," this is the wrong grid. If the problem is "there are forty things on my list," it is the right one, and no amount of weighting will make a scoring matrix do that job. ### The risk-reward grid Also a 2x2, but with the axes set to whatever trade-off you care about — payoff against risk, or impact against effort. Options are placed as points on the grid rather than scored to a total. Its distinguishing feature is that it deliberately refuses to name a winner. You get the shape of the trade-off instead. Two options with identical weighted totals can sit in completely different places here — one cheap and modest, the other expensive and transformative — and a single number would have hidden that. It suits early triage and portfolio conversations, and it is the wrong tool when someone is [waiting on one recommendation](/knowledge/influence/pitching-ideas/). ## How to build a decision matrix in six steps The build itself is a solved problem. Asana, monday.com, Lucid, Wrike, airfocus and SafetyCulture all publish very nearly the same sequence: list your options as rows, list your criteria as columns, assign a weight to each criterion, score every option against every criterion on one fixed scale, multiply each score by its weight, then sum each row. Two details do most of the real work. The first is how many criteria you allow. Standard guidance across these guides is to consolidate down to roughly five to eight, because past about ten each additional criterion dilutes the weight of every other one and the totals flatten toward a tie. When a matrix comes back with every option within a point of the others, the cause is usually too many criteria rather than genuinely equivalent options. The second is sequence — and it deserves its own section, because getting it backward is how this tool fails. ## Where a decision matrix quietly goes wrong Set the criteria and their weights before you look at how the options score. That single rule is most of the discipline. If you already know which option you want and adjust the weights until it wins, you have not evaluated anything; you have built what Wrike accurately calls a [confirmation-bias machine](/knowledge/self-awareness/unconscious-bias/) and handed it the authority of arithmetic. It is nearly impossible to detect from outside, because a rigged grid and an honest one look identical on the page. The second failure mode is quieter. A matrix containing only measurable factors biases the result toward the most quantifiable option rather than the best one. Cost, delivery time and headcount score cleanly. Maintainability, team morale and strategic fit do not, so they get dropped — which does not give them no weight, it gives them a weight of zero. The arithmetic stays faithful. It totals exactly what you chose to put in front of it. One more thing worth being honest about: there is no evidence that matrix-assisted decisions turn out better. The claims made for the tool — removes emotion, reduces subjectivity — are claims about mechanism, not measured outcomes. What a matrix reliably delivers is reasoning you can show and repeat. That is a genuine benefit, and a smaller one than the templates imply. All three problems are the same problem underneath: the grid computes, and you decide what goes into it. Whether you would catch yourself steering the weights toward a conclusion you had already reached is a question about your own habits rather than about the tool, and it is worth [checking your decision habits](https://assessment.headwayskills.com/) before the next call that counts. ## The skills that decide what the grid cannot Look at what is left over once the arithmetic is done. Something has to establish whether this decision is even yours to make, whether these are the right criteria, who else belongs in the scoring, and how to carry a total into a room that may not want to hear it. None of that lives in the grid. All of it is practiced behavior rather than temperament. **Decision-Making** is what sits underneath any version of this tool. Know your decision authority and your organization's guardrails before you build anything, because a beautifully weighted matrix on a decision that was never yours is wasted work. Slow down when you notice you are rushed or emotional. Use hard facts where they exist and accept "good enough" rather than retuning weights for a third evening. And get another opinion, deliberately from someone likely to disagree — it is the cheapest protection there is against anchoring on whichever option you saw first, or against defending the incumbent because of what has already been spent on it. **Influence** is where the matrix actually pays off, because the finish line is not the highest total, it is someone else agreeing. A completed grid is a demonstration: it makes your reasoning visible, and it lets you present the runner-up with its drawbacks intact rather than delivering a one-sided pitch, which is what makes the case credible instead of promotional. It also forces the preparation that junior people skip — working out who genuinely decides, and what your criteria are worth to them rather than to you. **Teamwork** takes over the moment more than one person is scoring. Group matrices fail in predictable ways: have people score independently before comparing, so the most senior voice in the room does not set everyone else's numbers, and ask the quiet ones directly for theirs. Disagree about a weight without making it about the person. Then, once the total is in, commit to it loyally even if you argued the other way — a team that relitigates settled calls pays for every decision twice. The Work Skills Test will tell you [which one gives way first](https://assessment.headwayskills.com/) when the pressure is on. These three sit among **twelve** work skills it measures, and knowing their order for you is considerably more useful than another template. ## Where this leaves you Some of this is probably not new. Most people who go looking for a grid have already been weighing the same factors in their head for days — the tool mainly gets them onto paper where they can be argued with. You have just read through five variants and three failure modes to the end, which is more scrutiny than most people apply to a tool before using it on something that matters. Judgment is not a fixed quantity you either have or lack; it is built from decisions you have already made and reviewed honestly, and it can be built further without becoming a different person in the process. It is worth building now for a plain reason: decisions get less reversible as your remit grows, and the habits you form on small calls are the ones that will be running when the calls stop being small. ## See where your judgment holds The grid you can set up in ten minutes. What it cannot tell you is how sound the person filling it in currently is. That is what the **free** Work Skills Test measures — a short self-assessment across all twelve work skills, decision-making, influence and teamwork among them, returning them ordered so you know which single one to work on first instead of trying to improve everything at once. One sitting, nothing to prepare. **[Take the skills test](https://assessment.headwayskills.com/)** *Takes 7 minutes, and your results come back straight away.* ## Who this is for - Professionals building practical workplace skills - Readers looking for specific, usable work advice - Managers, educators, and coaches supporting career readiness ## Common questions ### What is this guide about? A decision matrix scores your options against weighted criteria. The five types, how to build one in six steps, and the two mistakes that rig the result. ### Which Headway skill does this connect to? This guide connects primarily to Decision-Making. It also relates to Influence, Teamwork. ### What is the recommended next step? Use the free Work Skills Test to reflect on which work skill to improve next. ## Related pages - https://headwayskills.com/knowledge.md - https://headwayskills.com/knowledge/decision-making.md - https://headwayskills.com/knowledge/influence.md - https://headwayskills.com/knowledge/teamwork.md - https://headwayskills.com/work-skills-test.md ## Citation guidance Use the canonical page when citing this content: https://headwayskills.com/knowledge/decision-making/decision-matrix/ Preferred summary: "A decision matrix scores your options against weighted criteria. The five types, how to build one in six steps, and the two mistakes that rig the result." ## Change log - 2026-07-07: Content collection version published.