Theory of Change

A planning and diligence pattern that states how a specific investment or program is expected to produce specific outcomes, then turns that pathway into assumptions, metrics, and learning triggers.

Also known as: ToC, impact pathway, results chain, program theory, logic model.

Context

Impact Measurement and Management starts before capital is committed. The office has a thesis, a target population, a structure, and a set of claims it expects to make later. A theory of change is the document that forces those claims into a causal sequence before the annual report, verification review, or family council meeting asks whether the work succeeded.

The pattern applies at several levels. A foundation may write a portfolio-level theory of change for a decade-long housing strategy. A family office may write an investment-level theory of change for a $12M catalytic credit facility. A GP may write a fund-level theory of change for a climate-resilience fund. A DAF sponsor may ask a donor family to write one before approving a recoverable-grant strategy. The form scales, but the discipline is the same: name the problem, name who experiences it, name what the intervention does, name what changes next, and name what evidence would make the office revise the claim.

In family-office practice, the theory of change also connects rooms that usually work separately. The program team knows the beneficiary problem. The investment team knows the instrument. The family council knows the purpose. The controller knows what the reporting stack can actually measure. If those four groups don’t share the same pathway, the office will measure what is easy, announce what is attractive, and discover too late that neither one proves the outcome the family cared about.

Problem

Impact claims often start with activity and end with aspiration. The office funds a workforce lender, a clean-energy fund, a childcare facility pool, or a recovery-oriented health provider. It then reports dollars committed, companies financed, people reached, megawatts installed, or clients enrolled. Those numbers may be accurate. They may also stop one step before the thing the family said it wanted.

Outputs are not outcomes. Financing 3,000 small-business loans is an output. The intended outcome may be increased household income, lower business failure, job stability, or stronger local ownership in a place where conventional lenders left. If the office doesn’t name that outcome at the start, it can’t choose the right metric, judge whether the intervention worked, or explain what to do when the data disappoints.

The deeper problem is attribution. A family office can sit near a good organization and still not know how its capital contributed. It can report the investee’s mission as if it were the investor’s effect. It can confuse a good story with a tested causal pathway. Theory of Change is the pattern for refusing that confusion before it hardens into impact washing.

Forces

• Clarity versus complexity. A clean diagram helps decision-makers, but real social and environmental change is not linear.
• Standardization versus fit. Portfolio comparison needs shared vocabulary, while each issue area needs outcome measures that fit the population, geography, and intervention.
• Ambition versus evidence. Families often want system-level change; the available evidence may support only a narrower near-term outcome.
• Accountability versus learning. The office needs targets, but a theory of change should also tell the committee when to revise the strategy.
• Investor contribution versus enterprise contribution. The investee may produce the outcome; the office still has to say what its capital, terms, networks, or governance rights changed.

Solution

Write the theory of change as a diligence artifact before approval, not as a communications artifact after the fact.

Start with the outcome, then work backward. Name the long-term change the family cares about in words a skeptical committee can test. “Economic mobility in the region” is too broad. “More low-income childcare workers in three counties retain full-time employment six months after placement because childcare supply became available within a twenty-minute commute” is narrow enough to underwrite.

Then map the chain from input to impact. Inputs are the resources the office supplies: a PRI, a recoverable grant, a staff secondment, convening power, a guarantee, technical-assistance money, a board seat. Activities are what the investee or intermediary does with those inputs. Outputs are the immediate products of that work. Outcomes are the changes experienced by people, enterprises, communities, or the environment. Impact is the longer-run condition the office hopes those outcomes contribute to.

Make the assumptions explicit. The theory should say which links are backed by evidence, which links are uncertain, and which contextual factors could break the pathway. The office might assume that lowering the cost of credit increases borrower survival, that additional childcare seats increase worker retention, or that a new patient-monitoring product reduces emergency admissions. Those assumptions are not decorative. They are the places the measurement plan has to look first.

Attach metrics only after the pathway is stated. IRIS+ metrics, survey questions, beneficiary interviews, operating KPIs, and third-party datasets should measure the named outcomes and assumptions. If the office chooses metrics first, the theory will bend toward what the metric catalog can count. If the office states the pathway first, it can use standardized metrics where they fit and add custom measures where the local outcome requires them.

Finally, assign learning triggers. A working theory of change says what would cause the office to hold, revise, scale, or exit. If customer outcomes are weaker than expected after four quarters, what changes? If the investee reaches the output target but affected people report no improvement, who revises the intervention? If the office’s capital did not change the financing terms, does the allocation move from impact-first to value-aligned? Without those triggers, the theory becomes a diagram no one is willing to falsify.

How It Plays Out

Consider a $850M single-family office with a $95M foundation and a family council mandate around maternal health in rural counties. The foundation is considering a $10M PRI into a community health fund and a $1.5M grant for technical assistance. The draft memo says the investment will “improve maternal health access.” That statement is not underwritable.

The impact committee rewrites the work as a theory of change before approval. The long-term impact is a reduction in preventable maternal-health complications in five counties. The near-term outcome is narrower: more high-risk patients complete prenatal visits and postpartum follow-up within the clinically recommended windows. The fund’s activity is to finance three mobile-clinic operators and one data-coordination nonprofit. The foundation’s input is a seven-year 1% PRI plus a grant-funded reporting and patient-navigation layer.

The memo then states the assumptions:

The office does not approve the investment because the diagram looks good. It approves because the theory gives each body a job. The investment committee underwrites the PRI terms. The program team owns patient-navigation quality. The foundation board accepts the grant budget. The controller confirms that the reporting stack can collect visit completion, follow-up, and patient-survey data without creating a parallel spreadsheet system. The family council understands the claim it will be able to make later: not “we improved maternal health,” but “our concessionary capital and grant-funded navigation layer expanded visit access for high-risk patients, and we are testing whether that changed completion and complication rates.”

Twelve months later, output numbers look strong. Three clinics are financed, vehicles are operating, and 2,400 patients have been reached. Outcome data is weaker. Postpartum follow-up improves only modestly because patients are missing appointments after delivery, not before. The theory of change makes the revision obvious: the office keeps the PRI in place, but moves part of the technical-assistance budget from prenatal intake to postpartum transportation and peer outreach. The report to the family council says the initial assumption was partly wrong. That is not failure. It is the reason the theory was written.

A failure case is easier to recognize. A $300M family foundation makes a $6M grant to a national education nonprofit and receives a dashboard showing 18,000 students served, 700 teachers trained, and 42 school districts reached. The dashboard is polished, but no one can say which student outcome changed, why the family grant mattered, which assumptions were tested, or how the program would change if the first year disappointed. The grant may still fund useful work. It doesn’t yet have a theory of change strong enough to support an impact-first claim.

Consequences

The benefit is decision quality. The office can compare investments against the outcome pathway rather than against the charisma of the founder, the attractiveness of the issue area, or the size of the activity count. The investment memo becomes sharper because it has to answer how the structure produces the outcome, who experiences the change, how much change would count, what the risks are, and where the investor’s contribution sits.

The pattern also makes later measurement cheaper and more honest. IRIS+ selection becomes a filtered exercise rather than a hunt through a catalog. The Five Dimensions become usable because the “what” and “who” are already named. OPIM disclosure becomes less performative because the office can show the path from objective to assessment to monitoring to exit review. Independent verification has something to test beyond the existence of a policy.

The liabilities are real. A theory of change takes staff time. It can make principals impatient because it slows a gift or investment they already wanted to approve. It can expose weak evidence behind a favored issue area. It can also create false confidence if the diagram is treated as proof rather than a hypothesis. The office has to keep the theory alive after closing, or the exercise becomes another polished appendix.

The most important second-order effect is cultural. Once the office uses Theory of Change discipline, impact reporting becomes less theatrical and more adult. The office starts saying, “Here is what we thought would happen, here is the evidence so far, here is what surprised us, and here is what we changed.” That is a less glamorous sentence than “we touched thousands of lives.” It is also the sentence sophisticated principals, rising-generation members, co-investors, and verifiers can trust.

Related Patterns

Sources

• Rockefeller Philanthropy Advisors, Impact Investing Handbook: An Implementation Guide for Practitioners, 2020 — the asset-owner implementation guide that places theory of change in the “Why” stage, merging impact goals and investment goals before portfolio construction and measurement.
• Operating Principles for Impact Management, Principle 1: Impact Objectives and Principle 4: Ex-Ante Impact Assessment, 2025-2026 — current signatory-practice guidance treating theory of change as the throughline from strategic intent through investment decision, monitoring, exit review, and learning.
• Impact Frontiers, Five Dimensions of Impact, 2024-2026 — the field vocabulary for what, who, how much, contribution, and risk, including the distinction between enterprise contribution and investor contribution.
• W.K. Kellogg Foundation, Logic Model Development Guide, 2004, and Center for Theory of Change, How Does Theory of Change Work? — the program-evaluation lineage for backward mapping, assumptions, indicators, and the input / activity / output / outcome / impact chain.

This entry describes a structural pattern and is not legal, tax, or investment advice. Consult qualified counsel and tax advisors licensed in your jurisdiction before adopting any structure described here.