Theory of Change & Outcome Pathway

This workbench helps a group explain what change it seeks, for whom, through which activities, outputs and mechanisms, under which assumptions, and how it will learn whether the pathway remains credible.

It is a structured contribution story—not proof of attribution. A complete diagram can still be wrong. Every material arrow is a causal claim that needs an explanation, evidence, uncertainty, alternatives, responsible review and a plan to learn.

What the workbench produces

• a bounded change objective, focal population or system, perspective, horizon, owner and intended use;
• activities, outputs and near-, medium- and long-term outcomes connected as an explicit pathway;
• mechanisms, assumptions, external factors, supporting and challenging evidence, confidence and alternative explanations;
• indicators, data sources, baselines or targets when genuinely known, disaggregation choices and learning questions;
• visible adverse or unequal effects, evidence gaps, dissent, limitations and reservations;
• human review tied to substantive content, plus a structured downstream handoff;
• a comprehensive web report and branded PDF containing the normalized, machine-readable record.

Human governance

AI can draft questions and candidates from supplied context. It cannot invent evidence, targets, achieved outcomes, consultation, agreement or causal certainty. It cannot endorse the pathway for learning or planning, or claim impact. Humans remain responsible for consequential causal judgments, affected perspectives, measures, safeguards and decisions.

Readiness means the record is sufficiently complete for its declared use. It does not mean the intervention will work, that outcomes occurred, that every stakeholder agrees, or that attribution was established.

Continue with Frame and outcomes.