Issue Tree
An Issue Tree is a diagram that breaks a large problem or question into smaller parts that do not overlap and together cover the whole. The technique comes from management-consulting practice and is closely tied to the MECE principle (mutually exclusive, collectively exhaustive) associated with Barbara Minto and McKinsey. A problem tree branches by asking why something happens, a solution tree by asking how it could be fixed.
Facilitation script
- 1
Agree the root: phrase the problem or question precisely and write it at the left or top of the board.
8 min - 2
Individually sketch a possible first-level split, then compare two or three versions and choose the one that separates the problem most cleanly.
12 min - 3
Develop the second and third levels together or in subgroups, one branch each. Keep entries as categories.
20 min - 4
MECE check: walk the tree, merge overlapping branches and add anything that is missing.
8 min - 5
Prioritise branches using available data and judgement; mark the two or three that matter most.
8 min - 6
Start a solution tree for the top branch: solution categories first, then ideas within each.
15 min - 7
Assign owners and next steps for analysis or follow-up.
5 min
Tips
The first split decides the quality of the whole tree, so try two or three different ways of cutting the problem before committing, for example by process step, by customer segment or by an arithmetic breakdown such as volume times price.
Build the tree on sticky notes so branches can be moved when an overlap shows up.
Aim for MECE at the top levels and accept a looser fit further down; chasing perfect logic on the fourth level costs more time than it returns.
Common pitfalls
Accepting the first split that comes to mind, which often mirrors the organisation chart and hides causes that cut across departments
Mixing causes and solutions in one tree, so branches stop being comparable and the logic cannot be checked
Letting branches overlap, which leads to double counting and to two people analysing the same thing
Writing specific hypotheses at the top levels, which narrows the tree to what the group already believes
Breaking every branch down to the same depth instead of prioritising, which turns a 60-minute session into an unfinished wall chart
Variations
Hypothesis tree: when the group already has a likely answer, put that hypothesis at the root and branch into the conditions that would all have to be true, then test them. Remote: use a mind-mapping tool or a shared board with connector lines. Larger groups can agree the first level together, then split into subgroups that each develop one branch and return to check for overlaps. A problem tree can be followed by 5 Whys on a single priority branch when a deeper causal chain is needed.
Where it fits
When to use it
A team faces a broad problem and people are proposing solutions for different parts of it without a shared picture
A metric has moved and you need to locate where in the business or process the change comes from
Analysis work has to be split between several people without gaps or duplicated effort
A group needs to explain to decision-makers why it is focusing on one part of a problem and not others
An ideation session keeps producing the same kind of idea and you want to lay out the full range of solution categories first
When not to use it
Causes interact and feed back on each other; a tree forces them into separate branches, so use a causal loop diagram or the Iceberg Model
You already have a pile of observations or ideas to make sense of; an affinity map groups existing material bottom-up, while an issue tree structures top-down
The problem is a single incident with one causal chain; the 5 Whys is quicker
The group does not know the domain well enough to propose a sensible breakdown; gather facts or interview experts first
The question is about values or direction and not about parts of a whole; a facilitated dialogue will serve better than decomposition
Related methods
Frequently asked questions
What is Issue Tree?▾
An Issue Tree is a diagram that breaks a big problem or question into smaller parts, level by level. Each level should consist of parts that do not overlap and together cover everything above them. Teams use it to structure a problem before analysing it and to decide where to focus.
What does MECE mean in an issue tree?▾
MECE stands for mutually exclusive, collectively exhaustive. Mutually exclusive means no cause or item belongs in two branches; collectively exhaustive means nothing relevant is left out. The principle is associated with Barbara Minto and consulting practice at McKinsey.
What is the difference between a problem tree and a solution tree?▾
A problem tree starts from a problem and branches by asking why it occurs, which produces categories of cause. A solution tree starts from a chosen sub-problem and branches by asking how it could be solved, which produces categories of solution and then ideas. Many teams build the problem tree first and a solution tree for the branch they prioritise.
How is an issue tree different from an Ishikawa diagram?▾
An Ishikawa or fishbone diagram collects possible causes under standard headings such as people, methods and machines. An issue tree has no fixed headings and requires each level to be a clean, complete split of the level above. The fishbone is quicker for brainstorming causes; the tree is stricter and better for dividing analysis work.
How many levels should an issue tree have?▾
Two to four levels are usual. Stop when a branch is concrete enough to investigate with data or to act on. Going deeper on every branch is rarely worth the time; prioritise first and only extend the branches that matter.
Plan your next workshop with AI
Workshop Weaver helps you combine methods like Issue Tree into a complete, timed agenda in minutes.
Try it freeMethod descriptions on Workshop Weaver are original content written by our team, based on established facilitation practices. This method was inspired by work from Management-consulting practice (MECE principle, Barbara Minto).