Issue Tree

The defining requirement of a good issue tree is that each level is MECE — Mutually Exclusive (the branches don't overlap) and Collectively Exhaustive (the branches cover every possibility). Mutually exclusive means no sub-problem is double-counted; collectively exhaustive means nothing is left out. For example, splitting 'customers' into 'new customers' and 'existing customers' is MECE — every customer is one or the other, with no overlap. Splitting them into 'young customers' and 'price-sensitive customers' is not MECE, because a customer can be both (not mutually exclusive) and some are neither (not exhaustive). MECE is what makes the tree trustworthy: if it's truly exhaustive, the answer must lie somewhere on it.

There are two flavors. A pure issue tree (or 'why tree') breaks the problem into open questions — 'Why are profits declining?' splits into 'Is it revenue or cost?' This is exploratory and used when you don't yet know the answer. A hypothesis tree starts from a proposed answer and breaks down what would have to be true for it to hold — 'We should enter the European market' splits into 'The market is large enough,' 'We can win share,' and 'It's profitable to do so.' Consultants often start with an issue tree to map the space, then switch to a hypothesis-driven approach to test the most likely answer first, which is faster than investigating every branch equally.

Start with the core question at the top, then ask 'what are the distinct components of this?' and split into 2–4 MECE branches. Repeat for each branch until the sub-problems are concrete enough to attach data or analysis to. A profitability tree is a specific, pre-built issue tree for profit problems (Profit = Revenue − Costs). For open-ended strategy questions you build a custom tree: 'How should the client increase revenue?' might split into 'increase volume' and 'increase price,' with volume splitting into 'existing customers buy more' and 'acquire new customers.' The goal is a structure you can verbalize, drive questions from, and prune as you learn which branch holds the answer.

Case interviews exist primarily to test structured thinking, and the issue tree is the clearest demonstration of it. A candidate who organizes an ambiguous prompt into a clean, MECE tree shows they can break down complexity, prioritize, and avoid the random 'spray of ideas' that signals weak problem-solving. Interviewers also watch whether you adapt the tree to the specific case rather than dumping a memorized framework — the strongest candidates build a bespoke, MECE structure tailored to the prompt and then hypothesis-drive into the most promising branch. The tree is both a thinking tool and a communication tool: it lets you and the interviewer see your logic on one page.