Building a Theory of Change that funders trust

Why funders ask for one

Funders ask for a theory of change because they want to understand why you believe your program will produce the results you promise. A list of activities and a list of outcomes does not answer that question. A theory of change does: it lays out the chain of conditions that must happen, in order, for your activities to produce lasting change, and it names the assumptions holding the chain together.

Program officers read many of these. The ones that earn trust are specific, honest about uncertainty and grounded in evidence. The ones that do not are diagrams with arrows that connect 'job training' to 'economic prosperity' without explaining anything in between.

Theory of change and logic model are not the same thing

The two terms are often used interchangeably, but they do different jobs. A logic model is a structured summary of a program: inputs, activities, outputs, short-term outcomes, intermediate outcomes and long-term outcomes, usually in a table. It describes what the program does and what it expects to measure. Federal NOFOs often require one as an attachment, and its columns are fairly standard.

A theory of change explains why the logic model should work. It works backward from the long-term goal and identifies the preconditions that must be in place for it to happen, then the preconditions for those, until it reaches the program's activities. It makes the causal reasoning explicit and lists the assumptions at each link. A strong proposal often has both: the theory of change in the narrative and the logic model as the summary table.

Start from the end and work backward

Begin with a long-term outcome that is specific and within reasonable reach of your program. 'End poverty in the county' is a mission, not an outcome. 'Increase the share of program completers employed in jobs paying at least the county's living wage 12 months after exit' is an outcome a funder can believe and you can measure.

Then ask what must be true just before that outcome happens. For the employment example: participants hold an industry-recognized credential, they have been placed with an employer, and they have the transportation and child care to keep the job. For each of those, ask again: what must be true first? Keep going until you reach the activities your program actually delivers. The chain you have built is the backbone of your theory of change.

Do this as a working session with program staff and, where possible, a few participants or alumni. They know which steps actually break. A case manager may point out that participants who earn the credential still lose job offers because they cannot pass a background check, which adds a precondition the development team would never have spotted from the office.

• Name one long-term outcome you can measure
• List the preconditions immediately before it
• Repeat for each precondition until you reach your activities
• Mark which links your program controls and which depend on others

Make the assumptions visible

Every causal link rests on assumptions. A training program assumes local employers are hiring for the credential it teaches. A home visiting program assumes families will stay enrolled long enough to benefit. A food access program assumes people will use a new market if it is close enough. Writing these assumptions down is not a sign of weakness. It shows a funder you have thought about what could go wrong.

For each major assumption, say how you know it holds, or how you will check. Employer hiring can be checked with letters of commitment and labor market data. Retention can be checked against your own enrollment history. If an assumption is shaky, say what you will do if it fails. Reviewers and program officers consistently respond better to an applicant who names a risk and has a plan than to one who pretends no risk exists.

Ground the links in evidence

The links in a theory of change should rest on something more than conviction. There are three common sources. Published research, including evidence clearinghouses that review program models, can show that a type of intervention has produced similar outcomes elsewhere. Your own program data can show that your earlier cohorts followed the expected path. Practitioner and participant input can explain how and why a step works in your community.

Cite each source precisely. If you rely on an evidence clearinghouse rating, name the clearinghouse, the program model and the rating, and check that the model you run actually matches the one that was reviewed. Adapting an evidence-based model is common and reasonable, but describe the adaptation honestly. Claiming the evidence base of a model you have substantially changed is a credibility risk.

Keep it to one page, then use it

A theory of change that runs to six pages will not be read. Aim for a single diagram or a one-page narrative, with the assumptions and evidence in a short table underneath. Use plain words. If a board member cannot follow the chain from activity to outcome, a trustee will not follow it either.

Then use it beyond the proposal. Your evaluation plan should measure the key preconditions, not just the final outcome, so you can see where the chain breaks. Your funder reports should refer back to it. ELIXIR CONSUMERS stores theories of change and logic models in the vault alongside the outcome data that tests them, so each new proposal starts from the version your last evaluation refined. A theory of change that changes as you learn is the kind funders trust most.