What Is a Logic Tree? — Breaking a Problem Down Branch by Branch
A logic tree takes one big question (the trunk) and splits it, branch by branch, into pieces small enough to act on (the leaves). If MECE is the quality rule — is each split without overlaps or gaps? — a logic tree is the tool that makes the decomposition visible so you can think through it. Done well, it turns a vague problem into moves you can actually assign and measure. This guide shows the three types of tree, how to build one, and where to stop branching — with a concrete example.
What a logic tree is — thinking in trunk, branches, leaves
The logic tree is a staple of consulting problem-solving, and a close relative of the 'a claim above is supported by reasons below' structure in Barbara Minto's Pyramid Principle. You split top-down: from the trunk (the original question) to branches (sub-questions), then to leaves (concrete factors or moves).
MECE and logic trees are often mentioned together, but they play different roles. The logic tree is the tool that draws the decomposition; MECE is the rule that tells you whether each split is a good one. In a good logic tree, the branches at every level are MECE — mutually exclusive, collectively exhaustive.
Three kinds of logic tree — What, Why, How
The same 'tree' comes in three types depending on which question you ask. Pick the one that fits your goal.
- What-tree (component breakdown) — split a whole into its parts, e.g. 'revenue' into price × customers × purchase frequency. Use it to understand the current structure.
- Why-tree (root-cause) — repeat 'why?' to dig down to causes: churn is high → why? → … Use it to pin down the cause.
- How-tree (issue tree) — expand with 'how?' into moves: 'grow new sign-ups' → how? → … Use it to get to action.
In practice, a two-part move works well: use a Why-tree to find the real cause, then a How-tree to turn it into moves. Choosing the right question first is what decides the quality of the whole tree.
How to build one — set the trunk, split one axis per level
The key trick: at each level, split by a single, consistent axis. Take the strategy 'launch a B2B coffee subscription' (the output of the earlier SWOT and Cross-SWOT) as the trunk and break it into moves with a How-tree:
- Customer acquisition — how prospects find and sign up
- Product & pricing — what's in the plan and what it costs
- Delivery operations — how coffee reaches the office reliably
These three don't overlap and together cover the whole launch (each level is MECE). Then split each branch one more level — 'customer acquisition' into 'get found / complete the sign-up', for example. If you mixed axes — 'acquisition, pricing, and social media' — you'd get overlaps and gaps, because social media is part of acquisition, not a peer of it.
When to stop — stop once you reach a move
Not branching too far matters just as much. Stop when each leaf is 'something you can assign to someone and measure'. Most real problems need only two or three levels; splitting finer than that tends to become analysis for its own sake (analysis paralysis).
You can check each level in two directions. Bottom-up, 'so what?' — do the branches, bundled, add up to the parent's claim? Top-down, 'why so?' — does splitting the parent produce the branches? If both hold, the tree hangs together logically.
From logic tree to OKR — making leaves measurable (in the app)
A logic tree is still a structure, not a plan. Each leaf becomes a place to attach a goal: turn 'customer acquisition' into an Objective with measurable Key Results, and so on. Because the tree is exhaustive, you know your goals cover the whole strategy; because it's exclusive, no two goals fight over the same work.
Our strategy cockpit runs this whole cascade — 3C → SWOT → Cross-SWOT → MECE → OKR — on one screen, with each step feeding the next. The MECE view is where you build the logic tree; it flags duplicate siblings and hands the branches to the OKR step. Everything you type is stored on your device only (nothing is sent to our servers), and the AI drafting assistant is free to try on SWOT — bring your own API key (BYOK); you pay only your provider's usage. Unlocking AI across every framework is a one-time $9.90.
FAQ
What's the difference between a logic tree and MECE?
A logic tree is the diagram (the tool) that breaks a problem down branch by branch; MECE is the rule that tells you whether each split is 'without overlaps or gaps'. In a good logic tree, the branches at every level are MECE.
When do I use a What-, Why-, or How-tree?
Use a What-tree (component breakdown) to understand the current structure, a Why-tree to find a problem's root cause, and a How-tree (issue tree) to turn it into moves. In practice, chaining a Why-tree into a How-tree — cause, then action — works well.
How deep should I split?
Until each leaf is 'something you can assign to someone and measure'. Most problems need only two or three levels; going finer tends to become analysis for its own sake. Check up and down with 'so what? / why so?' — if the logic holds together, you can stop.
What is the So-what?/Why-so? check?
It checks each level two ways. Bottom-up, "So what?" — do the branches, bundled together, add up to the parent claim? Top-down, "Why so?" — does splitting the parent produce these branches? If both hold, the tree is logically connected.
How does a logic tree relate to Barbara Minto's Pyramid Principle?
A logic tree is a close relative of the "a claim above is supported by evidence below" logical structure in Barbara Minto's Pyramid Principle. Breaking a trunk into branches into leaves, top to bottom, is the same logic viewed from the other direction.
Other guides
- How to Do a SWOT Analysis — Complete Guide with Cross-SWOT (TOWS)
- How to Write OKRs — A Practical Guide with Examples
- How to Do a 3C Analysis — Customer, Competitor, Company
- What Is MECE? — Decompose Any Problem Without Gaps or Overlaps
- How to Use the TOWS Matrix (Cross-SWOT) — Turn Four Quadrants into Strategy
- Strategy Frameworks Explained — Use 3C, SWOT, TOWS, MECE and OKRs Together
- KPI vs OKR — the difference, when to use each, and how to connect them
- Turn Strategy into Execution — Running PDCA and Tracking the Daily Numbers (with Baton Board)
- Strategy Frameworks by Industry — 3C→SWOT→OKR Worked Examples for Four Business Types
- The Limits of SWOT Analysis — and How to Complement Them
- What Is PEST Analysis? Read the Macro-Environment and Feed SWOT
- What Is Porter's Five Forces? Read an Industry's Profit Structure
- What Is Value Chain Analysis? Break Down Internal Activities and Feed SWOT
- What Is the Ansoff Matrix? Choose a Growth Path by Product × Market and Feed OKR
- What Is VRIO Analysis? Evaluate Internal Resources with Four Questions and Pick SWOT's Strengths
- What Is STP Marketing? Narrow the Market in Three Steps to Decide Who to Serve
- How to Choose a Strategy Framework — Which One, and When
- What Is the BCG Matrix (PPM)? Allocate Resources Across Businesses in Four Quadrants
- What Is Jobs to Be Done (JTBD)? Finding the "Job" Your Customer Needs Done
Try it in your browser — free, no sign-up
3C → SWOT → Cross-SWOT → MECE → OKR on one screen. Build your logic tree in the MECE view and hand it to your OKRs. Strategy data stays on your device; the AI draft assistant is free on SWOT (with your own API key — BYOK).
Open the app →