Theory of Change Template for Nonprofits, With an Example

A theory of change is a one or two page explanation of how your work leads to the change you want to see: it starts from the long-term goal, works backwards through the conditions that must be true for it to happen, names what you will do to bring each one about, and states the assumptions linking them. The template below has eight sections, with a worked example for a fictional literacy programme, and it is in the Word download. A logic model, which many funders ask for, is a related but different tool: a table that runs forwards from inputs and activities to outputs and outcomes. Both are covered here, with how they fit together.

Where the idea comes from

The term was popularised by the evaluator Carol Weiss, in her 1995 essay “Nothing as Practical as Good Theory”, written for the Aspen Institute Roundtable on Community Change. Her point was that community programmes were hard to evaluate because their designers rarely spelled out how and why they expected the programme to work. The Roundtable and the organization ActKnowledge went on to develop the practice, and run theoryofchange.org together. Logic models came into wide use through funders: the W.K. Kellogg Foundation’s Logic Model Development Guide, updated in January 2004, describes a logic model as a picture of how your organization does its work and the theory and assumptions underlying the program, and includes a chapter on developing a theory-of-change logic model.

A theory of change template

Fill it in this order: goal, preconditions working backwards, then interventions, then assumptions. Most people start with what they already do and work forwards, which produces a description of current activities rather than a theory. Working backwards is what exposes the step nobody is doing.

How to write one, step by step

1. Agree the long-term goal. Specific enough to measure, ambitious enough to matter, and about people or places, not about your organization. Your mission statement is the starting point, not the answer.
2. Map backwards. Ask “what has to be true before this can happen?” and write each answer as an outcome. Then ask the same of each outcome, until you reach things you can directly affect.
3. Mark what you will do. Attach your activities to the outcomes they are meant to cause. Outcomes with no activity attached are either someone else’s job or a gap.
4. State the assumptions. For each arrow, write why you believe one thing leads to the next, and what evidence you have. Weak assumptions are the most useful thing the exercise produces.
5. Set indicators and thresholds. How you will know each outcome is happening, and how much is enough.
6. Test it with the people involved. Staff, volunteers, participants and a funder or two. If participants do not recognise the pathway, it is probably wrong.
7. Draw it. A single diagram with the goal at the top and pathways underneath. Keep the full version for planning and a one-page version for funders.

The example, written out

A theory of change is often easier to check written as a paragraph. For the fictional programme above:

If second- and third-graders who are reading below level get one-to-one tutoring twice a week, have books at home they choose themselves, and have families who read with them, then they will read more often and more confidently, and most will reach grade level by the end of third grade. We believe this because regular one-to-one practice improves reading fluency, because children read more when they choose the books, and because families take part when events are held at school pickup time. We will know it is working if 75% of children move up one reading level each year. Changes in the school curriculum, attendance and tutor turnover could affect the result.

Every “because” in that paragraph is an assumption to test. If families do not come to the reading nights, the theory says the result will suffer, which tells you what to watch and what to fix first.

Running a theory of change workshop

• Who: six to twelve people, including programme staff, a board member, a volunteer and, if possible, a participant or two. One facilitator who is not responsible for the programme.
• Time: two to three hours for a first draft of one programme.
• Materials: a long wall, sticky notes in three colours (outcomes, activities, assumptions) and the template.
• Running order: agree the goal (20 minutes); map preconditions backwards (60 minutes); attach activities (30 minutes); name assumptions and evidence (30 minutes); agree indicators (20 minutes).
• Afterwards: one person writes it up within a week, circulates it for comment, and brings it to the board.

Choosing indicators that work

Good indicators are things you can actually collect, at a cost you can afford, that change when the outcome changes. Prefer a few measured well over many measured badly: one reading assessment each term beats a long survey nobody completes. Pair each outcome with one quantitative indicator and one source of stories or observations, and decide in advance what result would make you change the programme.

Theory of change vs logic model

The same programme, as a logic model:

Notice what the logic model leaves out: why tutoring twice a week should lead to grade-level reading, and what has to be true at home and at school for it to work. That is what the theory of change adds. Many organizations write the theory of change once, for strategy, and derive a logic model from it for each grant application.

A short version for small organizations

A volunteer-run organization with one programme does not need a workshop and a diagram to benefit from this. Answer four questions on one page: what change are we trying to make, and for whom; what has to happen first for that change to occur; which of those things do we do; and what are we assuming that we have never checked. Share the answers with your board and one funder, and ask them what is missing. That page will improve your next grant application more than any template, and you can expand it into the full version when you grow.

Where to use it

• Strategic planning. It is the backbone of the programme section of a strategic plan, and the clearest way to decide which activities to stop.
• Grant proposals. The methods, outcomes and evaluation sections of a grant proposal are a theory of change written out in prose. Some funders ask for a logic model as an attachment.
• Evaluation. The indicators are your measurement plan, and the assumptions are what an evaluation should test.
• Board orientation. A one-page diagram explains the work to new directors faster than any report; see the board guide.
• Reporting. The annual report can show progress against the same outcomes every year.

Common mistakes

• Listing activities as outcomes. “Run 80 tutoring sessions” is an output. “Children read more often” is an outcome.
• Skipping the assumptions. Without them, it is a logic model with arrows.
• Claiming the whole goal. Most goals depend on many actors. Say which outcomes you contribute to and which you are responsible for.
• Making it too complex. If it does not fit on two pages, it will not be used. Keep the detail in an appendix.
• Writing it once and filing it. Review it every year against what the indicators show, and change the assumptions that turned out wrong.

Writing outcomes well is also the hardest part of most grant applications; the grant writing courses page covers where to learn it, and the grants guide what funders look for.

How this page was put together

The origin of the term is from theoryofchange.org, the joint venture of the Aspen Institute Roundtable on Community Change and ActKnowledge, and the logic model definition from the W.K. Kellogg Foundation Logic Model Development Guide (updated January 2004), both read on 14 October 2026. The template, example and logic model are our own, for a fictional organization.

This is reference information, not legal or tax advice. Rules vary by state and change over time. For a decision that carries real consequences, check the current text at irs.gov or your state registry, and talk to a nonprofit attorney or CPA.