A theory of change is the logic that connects what you do to the change you want to see. It explains, step by step, how your actions are meant to produce an impact — and under what conditions.
The term comes from the social impact world, but the idea is useful far beyond it: for a project, an initiative, even a personal decision. A theory of change simply makes explicit the reasoning “I do this so that this changes.”
A simple definition
It is a clear description of the path from a starting problem to an intended impact, passing through your activities and their intermediate results. It answers one question: why do we believe what we do will produce the change we hope for?
Why you need one
Without a theory of change, you act and hope it works, without knowing why. With one, you can test your logic, spot the fragile links, and measure in the right places.
It is useful even at a small scale. Someone who wants to change something at work benefits from spelling out: which problem, which action, which expected result, which final impact. That turns a good intention into reasoning you can check.
The elements of a theory of change
The starting problem
The point of departure: which situation you want to change, and for whom. A clear theory of change starts with a clear problem.
The activities
What you concretely do to act on that problem — train, build, support, produce.
The expected outcomes
What changes in the short and medium term because of the activities: new skills, new behaviours, improved access. These are the intermediate effects, distinct from mere activity.
The intended impact
The lasting change you aim to produce over the long term — the real transformation of the starting situation.
The assumptions
The conditions you take to be true for the chain to work. Making your assumptions explicit is often the most revealing step: that is where a project’s weaknesses hide.
The “if… then” logic
A theory of change reads as a chain of “if… then.” If we run these activities, then these outcomes should appear; if these outcomes appear, then this impact should follow. Each link is a testable claim, not a certainty.
That is exactly what makes it valuable: it turns a vague hope into a series of assumptions you can check one by one.
A simple example
Take a job-search workshop. Problem: qualified young people are not landing interviews. Activity: a workshop on CVs and interviews. Expected outcome: stronger applications and more interviews. Intended impact: more durable hires. Assumption: the main barrier was application quality, not a lack of openings.
Written this way, the logic becomes testable. If interviews do not increase, the assumption may have been wrong — useful information, not failure.
What practice taught me
At the Miller Center for Social Entrepreneurship, I developed impact measurement frameworks and monitoring and evaluation plans. The theory of change was usually the starting point: without it, you know neither what to measure nor why. And it need not be complex to be useful — a simple, honest chain with explicit assumptions beats an elaborate diagram nobody uses.
What’s next?
The theory of change is the backbone of impact measurement: it tells you where to look. It is a central tool of the fourth pillar.
To see how Impact connects with Intention, Continuity, and Mastery, start with the method.
Next step: Read the Method → — or see how to measure the impact of your work.
FAQ
What is the difference between a theory of change and a logic model?
Both map the path from activities to impact. A theory of change emphasizes the “why” and the assumptions linking the steps; a logic model is often the more schematic version. In practice, they complement each other.
Do you need to be an NGO to have a theory of change?
No. Any project or person aiming to produce change benefits from making their logic explicit. The scale changes; the principle stays: connect what you do to the change you seek, assumptions included.
How do I start writing mine?
Start from the end: define the intended impact, then work backwards — which outcomes lead there, which activities produce them, which assumptions must hold. A single page is enough to begin.
