Issue Trees

Definition

An issue tree is a hierarchical decomposition of a hypothesis into the specific questions or sub-issues whose answers would prove or disprove the main claim. Unlike a logic tree (which is merely a structural grouping), an issue tree is decision-focused: each node represents a question that, once answered, either validates or invalidates a branch of the hypothesis.

In the Book

The book describes issue trees as "the evolved cousin of the logic tree." They bridge structure and hypothesis. For the Acme Widgets example, the team's initial hypothesis—"We can lower the marginal cost of thrum-mats with a new curing process"—becomes the root. Below it branch three major issues: Does the new process save money? Can we implement it? Will it maintain quality? Each major issue spawns subissues: implementation breaks into "Do we have the facilities?" and "Do our people have the skills?"

By laying out the tree visually, the team can determine what analyses matter and what dead ends to eliminate immediately. When the CFO example appeared in the text, the initial diagnosis (CFO problem) was replaced with a different hypothesis (CEO's lack of structure) once the issue tree revealed which issues didn't hold up.

Why It Matters

Issue trees are transportable: any domain where you're testing a central claim benefits from mapping its logical components first. They prevent wasted analysis on irrelevant branches, make the implicit logic of a proposal explicit (good for buy-in), and serve as a natural outline for a presentation. They also surface dependencies—what question must be resolved before you can answer another.