Why a named method
Most teams don't lack intelligence, information, or motivation — they lack a shared discipline for how a decision should move from question to conclusion. Without one, discussions default to their natural physics: the first framing anchors, the loudest voice carries, the last argument wins, and the reasoning evaporates when the meeting ends. Those failure patterns are so regular we've named them.
The Argumentree Method is the counter-discipline: five elements that any consequential decision walks through, in order, on an argument tree. It descends from a long lineage — Toulmin's model of claims, evidence and warrants; the IBIS tradition of issue-based information systems — and it belongs to the decision-making cluster this library is built around — but the point is not theory. The point is that a team that can say "we're on element three" has a shared map of where the decision is, what's been done properly, and what hasn't.
Element 1 — Frame the Question
What it means: state the decision as one explicit claim that can be supported, attacked, and eventually judged — "We should enter the enterprise segment next year", not "enterprise strategy discussion". The frame includes scope, horizon, and what's deliberately excluded.
Why it matters: the frame is the first thing that can be wrong, and the most expensive. An excellent answer to the wrong question fails completely — which is why decision quality puts the appropriate frame first among its six elements, bound by the chain rule: overall quality equals the weakest link.
Common mistakes: letting the first speaker's framing stand unexamined (the Anchoring Trap); framing as a binary when the option space is wide (the False Dichotomy Trap); topics instead of decidable claims. In Argumentree: the frame is the tree's root claim — written, visible, and challengeable like any other node. "Is this the right question?" is a normal move, not a derailment.
Element 2 — Surface Arguments
What it means: gather the options and their pro and con arguments — independently, in writing, from everyone with relevant knowledge — before group discussion can shape them. Every argument carries its evidence; every option is a sibling, not an amendment.
Why it matters: a decision with one real option is not a decision, and a case gathered through open discussion inherits the room's dynamics — anchoring, conformity, airtime. Independence first is the condition collective intelligence research keeps finding at the bottom of every group that outperforms its members.
Common mistakes: brainstorming aloud before writing (the confident early voice sets the set); collecting only from the usual contributors (the Echo Chamber Trap); treating con arguments as disloyalty rather than contribution. In Argumentree: contribution is written and asynchronous — nobody outtalks anyone — and AI extraction can seed the tree from existing documents and discussions, so surfacing starts from what the organization already knows. Surfacing every side is pluralism made operational rather than aspirational: the tree is not complete until the strongest case against is on it, whoever in the room holds it.
Element 3 — Challenge Assumptions
What it means: every significant claim faces structured examination — questioned, answered, followed up, resolved — before it gets to carry weight. Not adversarial culture; a bounded protocol for testing what the case rests on.
Why it matters: reasoning that hasn't survived challenge is indistinguishable from rationalization. The sound-reasoning element is the one no framework supplies — quadrants and scans generate inputs, but only examination turns inputs into a case.
Common mistakes: challenges directed at paraphrases instead of the actual argument (the Strawman Trap); claims accepted or dismissed on their author's rank (the Appeal to Authority Trap); challenge reserved for other people's ideas. In Argumentree: structured challenge chains attach to the exact node they question — four turns, on the record, with the objector guaranteed a follow-up — and steelmanning is the house norm. When a challenge ends in genuine disagreement rather than a verdict, a structured compromise chain negotiates the middle ground on the same node, and a review chain re-examines settled arguments later — so the challenge history stays attached to the argument it tested. This is also where listening stops being a virtue and becomes a process property: a chain guarantees the challenged argument an answer — active listening enforced by the workflow — so no contribution can be skimmed past unread.
Element 4 — Weigh Evidence
What it means: the surviving arguments are rated on their merits — by everyone, before results are visible — and the ratings aggregate up the tree into a visible verdict on the question.
Why it matters: unstructured rooms weigh arguments by accident of delivery — volume, confidence, recency, rank. Merit-based, independent rating is the correction: the aggregate reflects the group's judgment of each claim, not the meeting's acoustics.
Common mistakes: the last argument outweighing the accumulated case (the Recency Trap); assertiveness carrying weak claims (the Loudest Voice Trap); weighing by memory at the end of a long meeting. In Argumentree: all arguments stay visible simultaneously — the case is a structure, not a stream — and ratings attach to nodes, blind to author, volume, and timestamp.
Element 5 — Reach Conclusion
What it means: decide against the whole visible case — and record the decision with its reasoning on the tree. If the call goes against the weighted verdict, the decider's node is where that gets explained. Then the record persists.
Why it matters: a decision nobody executes was a conversation, and a decision nobody can reconstruct gets relitigated. The recorded conclusion is what makes the decision durable — reviewable against outcomes, inheritable by people who weren't in the room, and safe from being re-argued from scratch every quarter.
Common mistakes: consensus theater (closing on fatigue rather than judgment); conclusions recorded without reasons; records that die in slide decks. In Argumentree: the conclusion is a node with the why attached, the audit trail keeps the whole case, and the next decision starts from what this one learned. And the calendar’s timeline slider replays how the tree grew — when each argument landed, and which recorded events influenced or steered the debate — so the decision can be re-examined months later exactly as it unfolded.
The method in practice — one decision, five elements
Illustrative walkthrough, compressed: a 60-person company deciding whether to adopt a four-day week pilot.
- 1Frame: root claim — "We run a 6-month four-day-week pilot for the product org, starting Q1." Scope explicit (product org, pilot, 6 months), exclusions explicit (not company-wide, not permanent). One challenge lands immediately: should support be included? Argued, resolved, frame amended — in writing.
- 2Surface: everyone contributes written pros, cons and variants over three days — client-coverage risks from support, focus-time evidence from engineering, two staged variants nobody had voiced in meetings. Twenty-three arguments, each with evidence, on the tree.
- 3Challenge: the coverage-risk argument gets a four-turn challenge (is the risk measured or assumed?); the productivity claim gets its evidence questioned and upgraded from anecdote to the pilot metrics that would test it. Two weak arguments retract; the rest are now examined.
- 4Weigh: the org rates the arguments before results are visible. The verdict: strong support, with the coverage risk rated the one serious con — visible to everyone, including leadership.
- 5Conclude: leadership decides for the pilot, with a coverage rota answering the top-rated con — the reasoning recorded on the root. Six months later, the review starts from the tree: what did we believe, and what happened?
The method, the standard, and the traps
Three vocabularies, one system. Decision quality is the standard — six elements a good decision must satisfy, judged at decision time. The Argumentree Method is the practice that produces decisions meeting the standard: framing serves the frame element, surfacing serves alternatives and information, challenging and weighing serve sound reasoning, concluding serves commitment. And the 7 Argument Traps are the named failure patterns each element defends against — every element of the method is an anti-trap by construction. The decision frameworks plug in as input generators, and the tutorials apply the method to specific rituals — sprints, pre-mortems, Delphi studies.
Frame it, surface it, challenge it, weigh it, conclude it — on a tree, on the record. Everything else on this site is a deep dive into one of those five verbs.