THE IDEA
Making your assumptions visible
A theory of change is a description of how and why you believe a set of activities will lead to the outcomes you want. It’s not a plan. It’s the logic underneath the plan - the chain of reasoning that connects what you do to what you hope will happen.
Every organisation, project, and intervention has a theory of change, whether they’ve made it explicit or not. A charity that provides mentoring to young people has an implicit theory: mentoring builds confidence, confidence improves engagement with education, education improves life outcomes. A company that invests in employee training has an implicit theory: training builds skills, skills improve performance, performance improves results. Making the theory explicit forces you to state your assumptions - and stating your assumptions is the first step to testing them.
The danger is that most theories of change are far too linear. They present a clean chain of causation: if we do X, then Y happens, which leads to Z. Real change in complex systems doesn’t work like a chain. It works like a web - with feedback loops, delays, multiple pathways, contextual factors, and the ever-present possibility that the system will respond in ways you didn’t anticipate. A useful theory of change acknowledges this complexity rather than pretending it away.
IN PRACTICE
The logic you didn’t know you were using
A government funds a job training programme. The implicit theory: training gives people skills, skills make them employable, employment reduces poverty. Each link in the chain seems reasonable. But the theory doesn’t account for the labour market (are there jobs for these skills?), the barriers to employment beyond skills (childcare, transport, discrimination), or the possibility that the training might build confidence more than competence. The programme delivers training. Employment outcomes are disappointing. The theory of change was never tested - it was assumed.
A product team launches a new feature to improve user retention. The theory: users churn because they can’t find what they need, the new feature makes discovery easier, easier discovery increases engagement, engagement reduces churn. But the team never tested the first assumption. Users might be churning because of price, performance, or a competitor - not discoverability. The feature works perfectly and retention doesn’t budge. The theory of change was wrong at the first link.
A parent decides to help their child with homework every evening. The theory: help builds understanding, understanding builds confidence, confidence builds independence. But the child experiences the help as pressure, the pressure creates anxiety about homework, and the anxiety reduces independence. The theory of change ran in the opposite direction because the mechanism (help) was experienced differently by the recipient than by the giver. The theory was plausible. The reality was different.
WORKING WITH THIS
Building a theory that works
Start by making your current theory explicit. Write it out: we believe that doing X will lead to Y because of Z. Then stress-test each link. Is there evidence for this connection? What are the assumptions? Where could the chain break? What else would need to be true for this to work?
Build feedback loops into the theory. Don’t just describe the linear pathway from activity to outcome. Ask: what would tell us this is working? What would tell us it isn’t? How quickly will we know? If the feedback loop is too slow (you won’t know for three years) or too weak (you’ll measure outputs but not outcomes), build in earlier, faster signals.
Treat the theory as a hypothesis, not a fact. The best theories of change are the ones that are designed to be updated. If the evidence shows that link two in the chain isn’t working, revise the theory and change the approach. A theory of change that’s never revised is a theory that’s never been tested.
THE INSIGHT
The line to remember
If you can’t explain how your actions will lead to your outcomes, you don’t have a strategy. You have a hope. Making the theory of change explicit is how hope becomes something you can test.
RECOGNITION
When this is in play
You need a theory of change when a project’s activities are clear but its pathway to impact isn’t. When different stakeholders have different assumptions about how the change will happen. When a programme is delivering its outputs (workshops delivered, reports written, meetings held) but the outcomes aren’t following. When the question “how does this actually lead to that?” produces hand-waving rather than a clear answer. When the evaluation asks “did we do what we said we’d do?” instead of “did doing it produce the change we expected?”