Writing Your Theory of Change: What Funders Need to See

The W.K. Kellogg Foundation's logic model guide — one of the most widely used frameworks in nonprofit program planning — defines a theory of change as the explanation of "why you believe the actions you want to take will produce the changes you want to see." This definition emphasizes the word "believe" deliberately. A theory of change is not a proof. It is an argument: a structured claim that if certain inputs are applied through certain activities, certain outcomes will follow. The job of the theory-of-change document in a grant application is to make that argument legible to an evaluator who did not develop it.

Program officers at major foundations evaluate dozens to hundreds of applications per funding cycle. They are not rejecting proposals because the problems they address are unimportant or because the organizations proposing them are insufficiently committed. They are rejecting proposals where the causal logic — the chain of reasoning from inputs to impact — is either absent or cannot be followed. A theory of change that requires the reader to trust the organization rather than evaluate the logic is a theory of change that foundation program officers cannot recommend for funding, regardless of how much they believe in the mission.

What a Theory of Change Is Not

The most common error nonprofits make in theory-of-change writing is confusing a theory of change with a program description. A program description tells you what an organization does: "We provide after-school tutoring to 200 students in three Title I schools, delivering 40 hours of one-on-one instruction per student per year." This is accurate and useful. It is not a theory of change.

A theory of change tells you why those activities produce the outcomes they claim to produce: "Students who receive 40 hours of structured one-on-one tutoring in foundational reading demonstrate measurable gains in oral reading fluency (a predictor of reading comprehension established in the National Reading Panel's 2000 report); improved reading fluency at grades 3-5 is associated with higher likelihood of on-track promotion and reduced dropout risk in middle school (as documented in longitudinal research by Hernandez, 2011)." The second version makes an argument. The first describes an activity.

A theory of change without the causal logic — without the explanation of why the activities produce the outcomes — is a program description dressed up in theory-of-change language. Foundation program officers who review applications regularly recognize this pattern immediately, and it raises a fundamental question: if the organization cannot articulate why its program works, what is the basis for believing it does?

The Logic Chain: Inputs to Impact

The standard logic model framework, as articulated in the Kellogg Foundation guide and widely adopted by OMB and federal program evaluation guidelines, traces five linked elements:

1. Inputs: The resources the program uses — staff, funding, facilities, partnerships, community relationships.
2. Activities: What the program does with those inputs — the specific interventions, services, or programs delivered.
3. Outputs: The direct, measurable products of the activities — the number of students tutored, the hours of service delivered, the events held.
4. Outcomes: The changes in participants' knowledge, skills, attitudes, or behavior that result from the activities — the improvements in reading fluency, the changes in health behavior, the development of specific competencies.
5. Impact: The longer-term changes in community conditions that the outcomes contribute to — improved educational attainment, reduced disease burden, economic mobility.

The distinction between outputs and outcomes is where most nonprofit writing fails the funder's evaluation. Outputs are what you do. Outcomes are what changes as a result. Serving 200 students is an output. Those students improving reading fluency by one grade level is an outcome. Those students graduating high school at higher rates is an impact. Funders evaluate outcomes and impact. Many proposals only report outputs.

The Causal Chain: Why Your Logic Holds

The most rigorous element of a well-written theory of change is the explicit articulation of the causal mechanisms that connect activities to outcomes. Why does your program produce the outcomes it claims? This is where the evidence base for your intervention matters — and where the absence of an evidence base needs to be handled honestly rather than glossed over.

When a strong evidence base exists for the intervention, the theory of change should cite it: name the research, describe the populations and conditions in which the evidence was generated, and explain why those conditions are sufficiently similar to your program's context that the evidence applies. When the evidence base is weaker — when the program is newer, more innovative, or operating in conditions where existing research is limited — the theory of change should acknowledge this directly and explain what the organization is doing to build evidence over time.

Funders evaluate uncertainty differently than most nonprofits expect. A theory of change that acknowledges the limits of the evidence base and describes a rigorous evaluation plan is more fundable than one that overstates the evidence in ways a program officer can easily detect. The Center for Theory of Change's guidance on theory-of-change development consistently emphasizes that credible theories acknowledge what is assumed versus what is established — because the assumptions are where the organization's learning agenda lives.

What Major Foundations Actually Require

Different foundations have different specific requirements for theory-of-change documentation, and grant applications that do not match the required format are typically eliminated before substantive review begins. The most common requirements across major foundations include:

• A visual logic model: Many foundations require the theory of change to be presented visually — as a logic model diagram — in addition to or instead of narrative description. A diagram that cannot be followed visually without the narrative is a diagram that has not been designed for the reader.
• Measurable indicators: Outcomes must be tied to specific, measurable indicators that can be tracked over the grant period. "Improved reading skills" is not measurable. "A minimum 1.0 grade-level improvement in DIBELS Oral Reading Fluency scores over a 12-month intervention period" is measurable.
• Population specificity: The theory of change must specify the population being served — not "at-risk youth" but "students in grades 3-5 at three Title I schools in Baltimore City Public Schools with baseline reading levels two or more years below grade level."
• Timeline realism: Outcomes claimed within the grant period must be realistic given the intervention intensity and the research evidence on timeframes for change. Claiming long-term impact within a 12-month grant period is a credibility flag for evaluators who understand the evidence base.

Handling Uncertainty Without Undermining Confidence

Every theory of change rests on assumptions. Some assumptions are well-supported by evidence. Others are reasonable working hypotheses. Still others are genuine unknowns that the program is designed to test. The question is not how to avoid uncertainty — it is how to write about uncertainty in a way that builds rather than undermines funder confidence.

The pattern that undermines confidence is unacknowledged uncertainty: the theory of change that claims direct causal effects the evidence does not support, that projects outcomes on timelines inconsistent with the evidence base, or that assumes contextual conditions that may not hold. Evaluators who recognize these overstatements do not conclude that the program is strong despite the overclaiming — they conclude that the organization does not have an accurate understanding of what it is doing and why.

The pattern that builds confidence is acknowledged uncertainty with a learning plan: the theory of change that says "we are testing the assumption that X leads to Y in this specific population, because the evidence base from comparable programs suggests this mechanism; we will measure X and Y separately and evaluate the assumption at the 12-month mark." This signals evaluation sophistication, intellectual honesty, and organizational learning capacity — all of which are characteristics that sophisticated funders value in multi-year grant relationships.

Outputs and Outcomes: The Distinction That Gets Applications Funded

Foundation program officers who evaluate grants consistently report that the outputs-outcomes confusion is the most common writing failure in nonprofit applications. The confusion is understandable: outputs are what organizations actually control and can reliably report, while outcomes depend on participant response and are harder to measure. The natural instinct is to report what is certain — outputs — and describe outcomes aspirationally.

But funders are not funding outputs. They are funding outcomes. An organization that can clearly articulate what changes in participants and how it will measure that change is an organization demonstrating that it understands the difference between delivering a service and producing an effect. That distinction is the foundation of accountability in grant-funded work, and funders who support organizations over time consistently distinguish between the two.

A theory of change written for funders is a theory of change that makes the causal logic evaluable — where the steps from input to impact are explicit, the evidence base is cited, the assumptions are named, and the measurement approach is specific. This is harder to write than a program description. It is the document that gets funded.