Stop Calling It Impact: What Your Program Data Is Really Telling You

You served 200 families this year. You delivered 500 hours of tutoring. You ran a 16-day course three times over. These are real accomplishments, and you should feel proud of every single one.

But here is the quiet worry that keeps a lot of us up at night: what if all that activity is not actually changing anyone's life? What if we are so busy counting what we do that we have never stopped to ask whether it works?

That question has a name. For 30 years, funders mostly asked nonprofits to report on what staff do: people served, hours delivered. Then a better question crept in, and it is deceptively simple. It is the "So what?" test. So what if you provide 10 hours of parent training? Are the parents actually better able to raise their children? Do they? Learning to answer that honestly is one of the most important skills you can build as your organization grows.

Outputs are not impact (and why the difference matters)

The single most common mistake in our sector is reporting outputs and calling them impact. Let me untangle the words, because getting them right changes everything about how you tell your story.

• Inputs are what you put in: funding, trained staff, a curriculum, your time and talent.
• Activities are what you do: the course, the peer support group, the one-to-one advisory session.
• Outputs are the direct, usually numerical results: number of students attending, hours of service delivered.
• Outcomes are the change over time in people's lives: increased confidence, new skills, a secured job, better health.
• Impact is outcomes minus what would have happened anyway.

So "we trained 200 parents" is an output. "Parents are better able to raise their children" is an outcome. And true impact reserves that word for the change you can genuinely link to your work, once you subtract what would have happened regardless (the counterfactual, sometimes called deadweight).

This matters because outputs live above the waterline, easy to see and easy to count. The outcomes that justify your existence live below it. If you only report outputs, you are describing motion, not change.

A gentle word on what you can honestly claim

Here is where I want to protect you from a trap. There is a meaningful line between two kinds of evidence.

Outcome measurement explores what your program provides, the change it intends to create, and whether that change is happening. It does not prove your program caused it.

Evaluation research, using control groups and comparison designs, tries to establish causation. It takes considerably more time, money, and expertise.

Most of us live in the first world, and that is completely fine. The honest move is to claim a plausible contribution rather than to overclaim proof. Saying "participants improved, and our program plausibly contributed" is more credible than declaring you single-handedly reduced a problem. Funders trust organizations that do not overclaim.

Name the change you intend to create

Before you measure anything, you need a clear picture of the change you are aiming for. A simple logic model does this beautifully. It is a visual map linking your inputs to activities to outputs to outcomes and impact. Drawing it out surfaces your assumptions, clarifies who is responsible for what, and, crucially, tells you which indicators to track.

When you write outcomes, the BACKS lens keeps you honest. Good outcomes describe intended change in:

• Behaviour (for example, reduced reoffending)
• Attitude (increased confidence)
• Condition (more people in permanent housing)
• Knowledge (improved parenting skills)
• Status (increased college enrolment)

Some outcomes are soft (a subjective shift in knowledge, attitude, or behaviour). Others are hard (a tangible change in condition or status, like a qualification or a job). You want both, because the soft outcomes often explain why the hard ones happened.

Set a baseline, then measure the change

You cannot show change if you never captured the starting point. A baseline is the condition you measure before you begin working with someone. Without it, a hopeful "they seem better" is just a feeling. With it, you can point to real movement.

For each outcome, choose an indicator: a specific, observable, measurable marker that turns a fuzzy concept into something you can track. Then gather two kinds of evidence. Quantitative data tells you how many, how much, how often. Qualitative data, the stories and interviews, unpacks the why. You need both. Numbers prove scale; stories give it meaning.

Make it a habit, not a scramble

The worst version of evaluation is bolted on at the end, in a panic, the week before a grant report is due. The best version is woven into daily work. Frontline staff capture data routinely, program managers coordinate it, and leadership carves out regular time to actually reflect on what it says.

This is what building a social-impact-focused culture looks like: easy feedback capture, routine reflection, and time set aside to make sense of what you learn. Good evaluation is useful, feasible, ethical, and accurate, and always proportional to your capacity. A small organization does not need a research department. It needs a baseline, a few honest indicators, and the discipline to look at them.

What to do next

Start small and start now. Pick one program. Draw its logic model on a single sheet of paper. Name one outcome using BACKS, choose one indicator, and decide how you will capture a baseline for the next person who walks through your door. That is a complete, honest measurement loop, and it is more than most organizations ever build.

If you want to see where measurement fits alongside your other growth priorities, our milestones can help you place it, and the assessment will show you where to focus first.

Your challenge this week

Take your most important program and rewrite one of its headline numbers. Move it from an output to an outcome. Change "we served 200 families" into a sentence about what actually changed for them, and then ask yourself one honest question: do you have the evidence to say that yet? Whatever your answer, you now know exactly what to measure next.