What counts as an output, and what counts as an outcome?
An output counts anything the program produces or delivers, and an outcome counts a change in the people or conditions the program is meant to affect. A useful test is to ask whether the figure would still be true if nobody's knowledge, skills, behavior, or circumstances had changed at all. If it would, it's an output.
An example: a made-up tutoring program
| Outputs | Short-term outcomes | Intermediate outcomes | Long-term outcomes |
|---|
| 40 tutoring sessions held; 60 students enrolled; average attendance per student | Greater confidence and skill in the subject | Better course grades and higher course completion | Students continuing in their education |
The figures and outcomes in the table are invented for illustration. Your own program's outputs and outcomes will depend on what it's designed to accomplish. If you're building a logic model, where outputs and outcomes sit side by side with goals, inputs, and activities, What's a logic model, and does my proposal really need one? covers it.
Why do funders ask about outcomes instead of how many people we served?
Funders ask about outcomes, and not only how many people you served, because a funder supporting a program usually wants to know whether it's making a difference, and the number of people served doesn't show whether that change is occurring. Saying that a program served 200 people describes activity. Explaining who participated, what they experienced, what changed, why that change matters, and what the organization learned describes the program's progress toward what it set out to do.
That doesn't mean output figures are unimportant to funders. Many require them in reports, and they're often the first thing a funder checks to see whether a program reached the people it was designed to serve. What a particular funder requires is in your grant agreement or reporting template, so read those before deciding what to report.
What can outcome figures hide?
Outcome figures can hide important differences between groups, the experiences of people who left the program early, and the role of influences outside the program. An overall average may show improvement while one group of participants made little progress, so looking at results by group, site, or level of participation often reveals more than the total. A survey given at the end of a program usually reaches only those who finished it.
Outcome figures also can't establish on their own that the program caused the change. Participants' lives are shaped by many influences, and positive outcomes don't automatically tell you what produced them. Being clear about what your evidence can and can't support makes an outcome report more credible to a funder.
Do outputs still matter if funders care about outcomes?
Outputs still matter when funders care about outcomes, because outcomes are difficult to interpret without them. If an outcome falls short, output information helps you determine why: whether the program was delivered as planned, who participated, and how much of the program they actually received. Participants who attended two sessions and participants who attended twenty didn't receive the same program, and output records are often the only place that difference is documented. Are we collecting the right data to show that our program is working? explains why information about how a program is carried out belongs alongside its results.
The Grantibly perspective
Lauren's perspective
Organizations doing meaningful work often know their programs are making a difference but struggle to demonstrate that difference. Anecdotes matter. Numbers matter. Participant voices matter. Outcomes matter. The strongest stories often combine them.
Evaluation can help an organization move beyond saying, "We served 200 people," to explaining who participated, what they experienced, what changed, why that change matters, what the organization learned, and what happens next. That's the difference between reporting activity and communicating impact. Numbers can show patterns and change, and participants' own accounts can help explain why a pattern may be occurring. Used together, they provide a much more complete picture.
