Why Activities Keep Getting Mistaken for Outcomes
I have taught grant writing to upper-level undergraduates for some time, and there is one piece of feedback I give super frequently: “That is an activity, not an outcome.” Writing an effective grant application requires us to think about how we will create change in our communities. Students, as beginners in this work, will often describe a thoughtful program. They want to hold workshops, hire staff, create partnerships, develop resources, provide services, or reach people who are not currently being served. Then they move to the outcomes section and list the same things again, but this time with numbers:
We will provide six workshops.
We will distribute 500 resource guides.
We will serve 100 families.
We will hire two new staff members.
All of those things are important, but none of them tells us what really changed.
It usually takes a week or so to work through this distinction with students, and it becomes clearer with a simple example. Holding six caregiver workshops is an activity. Eighty caregivers attending is an output. Caregivers learning a new strategy is an outcome. Caregivers using that strategy at home is a more meaningful outcome. To truly measure impact, consider how these outcomes lead to longer-term changes like reduced caregiver stress or greater family stability, which are often the ultimate goals.
For those of us who care about the work of human service organizations, those later outcomes are often the brass ring. We want the positive change in the world. We do not simply want to hold the workshop. We want the workshop to matter and to inspire meaningful progress.
The problem is that each step in the chain moves farther away from what we can directly control. A program can schedule a workshop. It can advertise it, hire a facilitator, reserve a space, and count how many people attend. It has much less control over whether participants learn something, use it later, or experience meaningful change because of it. Planning around an outcome rather than an activity requires a substantial leap of faith in our methodology. We are making an argument that if we do this work, under these conditions, with these people, then something meaningful is likely to change. That is a much more difficult promise than saying we will provide six workshops, and it's understandable that this can be frustrating.
The more often I give this feedback, however, the less convinced I am that this is primarily a student problem. In education, healthcare, social services, and public programs, we count trainings, appointments, referrals, contacts, meetings, partnerships, materials, and participants. Those numbers fill quarterly reports, dashboards, board presentations, funding applications, and legislative updates. In healthcare systems, counts of appointments and services directly translate into the lifeblood of funding systems. Some of that information matters. A program that planned to serve 500 people and reached 40 needs to know that, as has recently become a source of significant stress in Portland’s Preschool for All program. A training initiative that never delivered the training has not succeeded. Outputs can tell us whether implementation occurred. They cannot tell us whether the work mattered. Counting is easy. Human change is hard.
A count can be entered into a spreadsheet. It can be compared with a target. It can be placed in a chart and reported at the end of a quarter. It gives organizations something concrete to hold onto and governing bodies something visible to review. Human behavior profoundly complicates the change process. In systems, research, and organizational change, we are the ultimate confounding variable. People learn unevenly. Behavior changes slowly. Trust is a currency that accretes through repeated experience. New organizational practices may take years to become routine. Families may improve in one area while struggling in another. A referral pathway may exist on paper long before people can use it consistently. A staff training may increase knowledge without changing practice. Even when change occurs, it may be difficult to determine how much of it came from one program. Multiple people, systems, and conditions influence most meaningful outcomes. Good programs contribute to change. They rarely create it alone.
This creates a predictable pressure. Systems measure what is easiest to count and then begin treating those numbers as evidence of what is hardest to know. Funding and oversight systems also review what they can see and count for themselves. Government agencies and foundations are accountable to governing bodies that may not be especially interested in complex, nuanced questions about slow change. Policymakers and organizational leaders want results, even when we know those results are difficult to achieve and slow to arrive.
Teaching this work, I often have to talk with students about walking the fine line between promising things that sound appealing and overplaying what is actually feasible within a tight one-year grant timeline. The difficulty is that many important outcomes will not occur within the timeline of the grant. A one-year initiative may be able to recruit participants, deliver services, create a referral process, increase knowledge, or improve initial confidence. It may not be able to demonstrate lasting behavior change, reduced hospitalization, lower workforce turnover, improved graduation rates, greater family stability, or population-level improvements in mental health. Some of those outcomes take years. Others depend on housing, transportation, staffing, insurance coverage, family circumstances, policy changes, economic conditions, or dozens of other factors outside the program’s control.
That does not make the outcomes unimportant. It also does not make increased services unimportant, particularly at a time when supports are generally lacking. It means we need to be honest about where activities, outputs, and outcomes sit in the change process. Some funders and oversight bodies understand this well. They are willing to hear that early outcomes may be modest. They allow programs to describe what can reasonably be accomplished within the funding period and what will require more time. They understand that an increase in knowledge may be an early outcome, while a change in behavior or community conditions may be much farther down the road. Others are much less able to hear that feedback. They want the major outcome during the contract period because that is when the report is due. They may ask a small program to demonstrate changes in community conditions that developed over decades. They may interpret an honest explanation of limits as a lack of ambition or evidence that the program is not working.
In turn, I see programs respond to that pressure to provide. In the resource-strapped fields where I consult, many systems choose things that are easy to count. Others stretch the meaning of the data they have. Participation becomes engagement. Satisfaction becomes effectiveness. A referral becomes access. A completed training becomes changed practice. A new partnership becomes improved coordination. Sometimes those things are early signs of progress. Sometimes they are simply evidence that an activity occurred.
This creates a strange system in which the most honest programs can look weaker than the programs making the strongest claims. A program that says, “We reached 200 people and increased short-term knowledge, but we do not yet know whether practice changed,” may appear less successful than one claiming that a single training transformed an organization. The second report sounds better, but it tells us less.
Over time, organizations can become very skilled at demonstrating effort without learning much about effectiveness. Developing the right reporting language becomes a vital skill for leaders trying to sustain their teams and programs. I have heard colleagues who evaluate programs talk about the work involved in piecing together this double language: translating what actually happened into the kind of results a funder expects to see without making claims the evidence cannot support. The process is not always intentionally deceptive. Often, it is how people survive inside systems that demand certainty where certainty does not exist. Still, it has consequences. Leaders begin to second-guess one another’s claims. Evaluators learn to read between the lines. Programs become cautious about admitting that an intervention did not work as expected or that meaningful change will take longer than the funding allows. Staff spend substantial time producing reports that satisfy oversight requirements but offer limited guidance for improving the work. Through the confusion about what we are actually trying to achieve, we inadvertently damage the openness and honesty needed to learn from the process.
This is the part I keep returning to in the classroom. Students need to understand the difference between an activity, an output, and an outcome. Their grants are stronger when they can explain that chain clearly. But they are also preparing to enter systems that routinely blur those distinctions themselves. I do not want them to learn that the solution is simply to make bigger promises. I want them to learn how to make an honest argument about change: what the program will do, what it can reasonably influence, what can be measured now, and what may not be known for years. Counting is necessary. Human service systems cannot operate without knowing who was served, what was delivered, and whether the work happened as planned. But counting is not the same as knowing whether people’s lives changed. That difference is harder to explain, harder to measure, and sometimes harder for funders to hear. It is also where the real work begins.
