A theory of change is the broader narrative that a logframe summarises. Where the logframe is a matrix, the theory of change is a story — a structured account of how and why a particular set of actions is expected to lead, through a chain of intermediate steps and under specified conditions, to a particular set of long-term outcomes.
It adds three things the logframe typically omits: an explicit account of the contextual factors that shape the pathway, a fuller treatment of the assumptions that link each step to the next, and an honest acknowledgement of the rival hypotheses that might also account for any change you observe. A good one contains a clearly stated long-term outcome, a backwards-mapped chain of intermediate outcomes, the interventions meant to make them more likely, the assumptions that must hold, a contextual analysis, the rival explanations, and the signals that will tell you whether the theory is unfolding as expected.
All of that is in the textbooks. What the textbooks under-emphasise is how a theory of change should be produced.
A theory of change developed by a single consultant in a hotel room over two days is almost always inferior to one argued into existence collaboratively by the people who will implement the programme.
The collaborative process is what surfaces the implicit assumptions, the disagreements, the contextual nuances, and the uncomfortable questions that a solo author would miss. It is also what builds the shared ownership without which the theory of change will be ignored the moment the consultant leaves.
The protocol I keep returning to
Convene ten to twenty people who together represent the implementing team, the partners, the intended beneficiaries, and one or two outsiders who can ask awkward questions.
Spend the first session defining the long-term outcome in concrete terms — working until the group can describe what success would look like to a stranger.
Spend the second backwards-mapping: “for that to happen, what would have to happen first, and what before that?” Capture each step on a sticky note and arrange them on a wall.
Spend the third session interrogating the assumptions: “why do we believe this step will lead to that one? What would have to be true? What could break it?”
Spend the fourth identifying rival explanations: “if we observed the outcome, what else could have caused it?” and the evidence the MERL system will need to distinguish the programme’s contribution.
Then transcribe the wall into a written narrative and a diagram, share it for revision, and treat it as a living document revisited at least annually.
FROM THE FIELD
During the development of a theory of change for a coalition-led anti-corruption project across six Nigerian states, one lesson emerged clearly: coalition alignment is often harder than technical design.
Civil society organisations, media actors, legal advocacy groups and community accountability structures all approached corruption from different lenses.
Through participatory workshops, stakeholder mapping and problem-tree analysis, the coalition collectively identified weak citizen accountability, limited institutional transparency, and fear of retaliation as core drivers.
The process revealed that a theory of change is not merely a planning tool but a negotiation framework — one that builds shared ownership, clarifies assumptions, and aligns diverse actors toward measurable systemic change.
This is why the logframe-versus-theory-of-change debate is a false choice. The theory of change is the deeper, richer narrative; the logframe is the disciplined summary of that narrative in a form that supports planning and reporting.
The theory comes first; the logframe is derived from it.
If a donor template forces you to begin with the logframe, reverse-engineer the theory of change as soon as you can, and use it as the working document for internal management while treating the logframe as a reporting artefact.
Every theory of change rests, in the end, on a judgement about plausibility. We cannot prove in advance that a chain of activities will produce a long-term outcome; we can only assess whether the chain is reasonable.
The role of MERL is not to manufacture false certainty but to test and refine plausibility over time, replacing assumption with evidence wherever possible.
A theory of change without a MERL system is just a story. A MERL system without a theory of change is just data. Together they form a learning loop.
TAKE IT TO YOUR PRACTICE
Before your next theory-of-change workshop, write down who will be in the room. If the list does not include someone who can argue with the design and someone who will have to live with it, you are about to produce a diagram, not a negotiation.