A 2×2 has four cells. Add a third axis — eight. By the time you reach eight binary questions, you are looking at two hundred and fifty‑six distinct outcomes. The order in which you ask them decides whether that tree is a tool, or a trap.
Almost every prioritization conversation begins here. Pick two axes — cost and importance are the classics — draw a cross, and you have a vocabulary.
Four cells. A team can hold all four in their head at once. This is why the 2×2 is so seductive — and why we reach for it again and again.
A 2×2 cannot hold a third axis. We could draw a cube — three dimensions, eight corners — but a cube on a slide is already a stretch. So we draw a tree instead. Press Add a question and watch what happens.
Every dimension you add does not add options — it multiplies them. This is the part that breaks intuition. Three questions: 8. Four: 16. Eight: 256. Each row in the table below is one binary question.
Hover a row in the table to see only that many outcomes light up. Two‑hundred‑and‑fifty‑six is what eight binary questions actually buys you. Nobody can hold that in their head.
These are the eight axes most leadership teams actually argue about. Edit them to match your team’s vocabulary. Each row contributes a factor of two; the bottom of the list is where the multiplication finally lands.
Click a verdict to assemble it. This is what one cell of the 256-cell hypercube actually says.
A decision tree is not just a list of questions — it is an ordered list. The first question you ask is the most powerful one, because each answer eliminates half of everything that lies beneath it. Ask the trivial question first, and you waste that cut on noise. Ask the right question first, and 128 outcomes vanish in a single breath.
Both cut the tree in half — the math is the same. But which half. In Approach A, the 128 you keep are still tangled with the question you actually care about; you have to keep asking. In Approach B, you have already isolated the half of the world that matters — everything downstream is now a refinement, not a discovery.
This is the same eight questions, asked in two orders. The bars show how many of the 256 outcomes are still worth thinking about after each question.
The lesson is not that decision trees are bad. It is that the first question is doing most of the work — and almost nobody spends any time choosing it.
Drag the arrows to move questions up and down. The tree below redraws each time. Your job: find the order in which the early questions feel load‑bearing — the order where, halfway through, you already know roughly what you’re going to do.
Each new criterion doubles the outcome space. Eight criteria is two‑hundred‑and‑fifty‑six. Twelve is four‑thousand. Adding “just one more axis” is never just one more.
Every question cuts half the tree. The first question cuts the most absolute volume. Spend disproportionate effort on choosing it.
A three‑question tree asked in the right order beats an eight‑question tree asked in the wrong one. Prune before you grow.