Framework library · Decision making and problem solving
Issue tree
An issue tree starts from one question and splits it into branches that are mutually exclusive and collectively exhaustive: no overlaps and no gaps. Each level breaks a branch down further until the questions are small enough to answer with one piece of analysis. The tree shows which analysis answers which part of the question, and which branches can be dropped early.
Use it when
- A broad question, such as how to grow revenue or why margins fell, needs breaking into pieces that can each be analysed.
- A team is about to start analysis and needs to agree which questions it will answer and who answers each.
- You suspect the team has jumped to one answer and want to lay out the alternatives.
Avoid it when
- The problem is a specific defect with a chain of causes. Five whys or a fishbone diagram is quicker.
- You are choosing between known options. Use a weighted decision matrix.
- The question itself is not agreed. A tree built on the wrong question is precise and useless. Agree the question with its owner first.
How to run it
Write one question at the root
Specific, with a measure and a date: "How can we raise revenue per tower by 15% by 2029?"
Split it so the branches add up
Follow the arithmetic where you can. Revenue per tower is tenants per tower times revenue per tenant, plus other revenue from the site. Splits that follow arithmetic have no overlaps or gaps by construction.
Test each level for overlaps and gaps
Check that no idea could sit in two branches and that every idea fits one of them. Fix the split before going deeper.
Go down until each branch is answerable
Stop when one piece of analysis can answer a branch, such as a count of towers inside operators' planned coverage gaps.
Size the branches and prune
Estimate each branch roughly and drop the small ones early. The tree is a plan of work, not an encyclopaedia.
Assign the analyses
Each leaf gets an owner and a date, and the findings roll back up the tree to answer the root question.
Work through it
Answer the questions below, or load the worked example to see a finished one. The drawing updates as you type. Export the result as a PowerPoint deck, a Word document, an Excel workbook, a PDF or plain text.
What you type stays in this browser, so you can close the page and come back to it. It is not sent to Blue Prysm or anyone else, and the exports are made here, on your device. Privacy policy.
Mistakes to avoid
- Splitting by department (sales, network, finance) instead of by the logic of the question, which builds in overlaps and gaps.
- Mixing levels, so a broad branch sits beside a detailed one and the tree cannot be read across.
- Building the whole tree before checking sizes. Most of the value sits in two or three branches; prune early.
- Treating the tree as the answer. Each leaf stays a question until its analysis is done.
Where it comes from
No single originator. Issue trees grew up in management consulting, notably at McKinsey, where Barbara Minto, who joined the firm in 1963, coined the MECE test (mutually exclusive, collectively exhaustive) for grouping ideas. She set out the logic in The Pyramid Principle, first published in the 1980s, and extended it to problem solving in The Minto Pyramid Principle: Logic in Writing, Thinking and Problem Solving (1996). Source.
Use it with
Further reading
Work through it with us
The frameworks here are free to use as they stand. If you would rather work through the question behind this one with us, these are the ways an engagement starts.