A theory of change template that explains the reasoning
A theory of change template helps you describe the change you want, how you expect it to happen and why that explanation is plausible. It should make assumptions, context and evidence visible alongside activities and outcomes. Filling the boxes does not prove the theory; it gives people something specific to examine together.
Use the editable worksheet below for a program, membership initiative, service or funded project. Start with the decision the model needs to support. A proposal, a redesign and an evaluation may need different levels of detail.
If you are new to the concept, begin with what a theory of change is. This page focuses on building a usable first version, checking it with others and connecting it to an evidence plan.
Download the editable theory of change template
Download the theory of change worksheet — CSV for Excel or Google Sheets →
Open or import the CSV in a spreadsheet. Replace the starter text, add pathways where needed and keep questions you cannot yet answer visible. The worksheet includes the situation, intended changes, proposed actions, reasoning, assumptions, evidence and review responsibilities.
There is no universal requirement to use exactly six boxes. A simple sequence may be enough for one initiative. A program with several partners or interacting pathways may need a diagram and a short narrative. Use a required funder format when one applies, while retaining the reasoning in supporting notes.
Scroll horizontally to see all columns →
| Part of the worksheet | What to write | What to check |
|---|---|---|
| Situation and people affected | The problem or opportunity, relevant context and whose perspectives matter | Does the description reflect evidence and affected perspectives? |
| Intended changes | Early, intermediate and longer-term changes | Are these changes distinct from the activities delivered? |
| Actions and resources | What the team and partners propose to do and need | Are the important capabilities and dependencies available? |
| Reasoning and assumptions | Why one change may support the next, and under what conditions | What could prevent the link from holding? |
| Evidence and uncertainty | What supports the explanation and what remains to be learned | Which alternative explanations deserve attention? |
| Review and revision | Who will review what evidence, when and for which decision | How will changes to the theory be recorded? |
Filled example: helping members apply a new practice
Fictional teaching example. A professional network wants members to use a new practice in their workplaces. Members say the existing guidance is difficult to apply. The network proposes workshops, guided practice and follow-up support. The intended result is useful application, not simply attendance.
Scroll horizontally to see all columns →
| Proposed pathway | Reasoning to examine | Evidence that could help |
|---|---|---|
| Accessible workshops and guided practice → members understand the method | Examples are relevant and participants have time and support to practise | Feedback on accessibility, participation records and an appropriate learning exercise |
| Understanding and support → members try the method at work | Members have suitable tasks, permission and an opportunity to apply it | Follow-up accounts of use and barriers, with the timing and respondent group recorded |
| Appropriate use → improved professional practice | The method suits the task and works alongside other necessary conditions | Suitable evidence of practice and results, considering other influences |
The arrows are proposed relationships. A positive workshop rating does not demonstrate the final outcome. A member might understand the method but have no opportunity to use it. That finding could change the support offered or the intended audience.
Include possible unintended effects too. A new practice might add work, exclude people with limited access or be unsuitable in some settings. A useful theory makes room for those findings instead of only describing the desired result.
How to fill in the template
1. Agree the purpose and describe the situation
Write the decision the theory will inform. Then describe the problem, relevant context and people or organizations affected. Gather perspectives that may challenge the delivery team's initial explanation. A concise statement is useful; an oversimplified one can hide the problem you need to address.
2. Name the changes you hope to support
Distinguish early changes from later ones. “Run four workshops” is an activity target. “Members can demonstrate the method” is a proposed learning outcome. “Members use the method appropriately” is a different outcome requiring different evidence.
3. Work backward to the actions and conditions
Ask what would need to happen for each important change to occur. Identify the actions your team can take, the role of partners and conditions outside your control. Do not imply that your organization alone produces every longer-term result.
4. Explain the connections
For each important link, state why it is plausible and what might prevent it. Use existing research, professional knowledge and stakeholder experience where relevant. Note when an explanation is still a hypothesis. Disagreement is useful information for the next review.
5. Choose priority learning questions
You do not need a new survey for every box or an indicator for every minor detail. Focus on the most consequential uncertainties and the decisions they affect. Identify existing evidence before adding collection work.
6. Review and record the next version
Check the model with the people who deliver, experience and use the results of the work. Record what changed and why. A first draft can start a conversation; complex initiatives may need substantial consultation. The quality of that discussion matters more than completing a diagram within a fixed number of hours.
BetterEvaluation's guidance on describing a theory of change provides further context for articulating the explanation and using it in evaluation.
Turn the theory into a manageable evidence plan
Keep detailed definitions outside the main diagram, in a companion data dictionary or measurement plan. Link each priority question to a source, relevant unit, timing and responsible person. Identify which evidence is already available and which collection would be proportionate.
Scroll horizontally to see all columns →
| Priority question | Collection decision | Interpretation condition |
|---|---|---|
| Who could participate? | Use appropriate registration and participation information | Protect privacy and distinguish people from attendance entries |
| Did members try the practice? | Use a defined follow-up window and a clear question about application | Report eligible and responding groups, including missing follow-up |
| What prevented application? | Collect useful open-ended accounts or other suitable evidence | Do not treat a few comments as the experience of everyone |
| Did practice improve? | Choose an evaluation method suitable for the claim | Consider alternative explanations and the limits of the evidence |
For a federated network, agree a small shared core needed for comparison while allowing local questions. Define the common outcomes, periods, units and response categories. Identical-looking survey questions alone do not guarantee comparable data; local context and collection methods can still differ.
Collect stable registration information once where appropriate and update changing details when needed. Use person-level matching only when the learning question requires it and the collection arrangements support it. Some questions are better answered at organization, site or group level.
How evidence can change the theory
Fictional example. A cohort has 100 eligible members. Sixty answer the follow-up, and 36 report using the practice. That is 60% of respondents and 36% of the eligible cohort known to report use. The remaining 40 members have no follow-up response; they should not silently become successes or failures.
Suppose some respondents describe a lack of suitable work tasks. The team can examine whether opportunity to apply is a missing condition in its theory. It might adjust recruitment, involve workplace managers or offer a different type of practice. The accounts inform that review; they do not establish the cause of every member's outcome.
Linking comments and measures makes questions easier to investigate. It does not automatically test a causal claim or replace evaluation judgment. Read how qualitative and quantitative analysis work together for reviewing themes alongside the relevant numbers.
Theory of change, logic model or logframe?
Scroll horizontally to see all columns →
| Format | Useful when you need | Companion resource |
|---|---|---|
| Theory of change | An explanation of how and why change is expected in context | This reasoning worksheet and supporting narrative |
| Logic model | A concise representation of program components and intended relationships | Editable logic model template |
| Logframe | A matrix connecting results, indicators, evidence and assumptions | Editable logframe template |
The tools overlap. A logic model can include assumptions and context; a theory of change can include measurable results. Choose the representation that helps the audience understand and use the reasoning rather than treating the labels as rigid competing methods.
Keep the reasoning connected to the work
A spreadsheet and diagram can be enough to develop the theory. The operational challenge comes later, when repeated surveys, partner updates and documents need to support the same questions. Definitions change, follow-up is incomplete and the report must explain which evidence was included.
Sopact's connected approach brings collection, context, analysis and review into the workflow. The team owns the definitions and decisions. Where this is useful, evaluate it with one real evidence question: can reviewers follow the reported result to the included records, understand the denominator and see the relevant supporting accounts?
Use the Impact Measurement & Reporting course to develop the evidence plan. For a report that communicates findings and their limits, continue to the impact-report writing guide and Sopact reporting examples.
Watch: use the theory beyond the diagram
These companion videos discuss the connection between a theory of change and ongoing evidence review. Use them alongside the worksheet; the team's reasoning and evaluation choices remain essential.
Turn Theory of Change Into Daily Decisions
Theory of Change Done Wrong: The Most Expensive Mistake Organizations Make
Frequently asked questions
What should a theory of change template include?
Include the situation, intended changes, proposed actions, reasoning, assumptions and relevant context. Add evidence questions and review responsibilities so the theory can inform learning rather than only presentation.
Does every theory of change need six boxes?
No. The appropriate structure depends on the initiative and audience. A simple pathway may work for one program; multiple pathways and a narrative may be needed for another.
Can I use the template before collecting results?
Yes. It is useful for planning. Distinguish existing evidence from hypotheses and unknowns, then decide which learning questions deserve attention first.
Is a theory of change proof of impact?
No. It explains how change is expected to happen. Assessing whether that explanation holds requires appropriate evidence and evaluation methods, including consideration of other influences.
Can different local teams use different surveys?
Yes, where local needs differ. Agree the limited shared fields and definitions needed for comparison, retain local context and document differences that affect interpretation.
Choose your next step
King Center
Related practice: bringing program feedback into review. The worksheet example is fictional and is not a causal claim about this customer.
Read the customer story →