From Activity to Impact: Building Performance Systems That Actually Work for NGOs

Authored by: Jenine Saleh Esq.

In the social sector, performance management often defaults to the easiest KPIs to track: dollars raised, services delivered, supplies distributed, trainings held, or clients reached. These metrics matter, but they are usually outputs, not outcomes. They show activity and scale, but they do not necessarily prove impact.

My experience building Global Health Conscious (GHC), a nonprofit I founded to expand access to medical and humanitarian resources for displaced and underserved populations, made this distinction clear. At scale, GHC coordinated millions of dollars in donated medical supplies annually.

1. Define the outcome before selecting KPIs

A strong performance system starts with a clear outcome statement. Before choosing KPIs, an organization needs to define what success means in practical, measurable terms.

At GHC, the obvious KPI was the value or volume of medical supplies distributed. But that was an output metric. The actual outcome was whether those supplies increased clinical capacity, supported procedures, or improved access to care for communities with limited resources.

That distinction changed the measurement framework. We still tracked inputs and outputs, but we began prioritizing outcome-oriented indicators: partner utilization, appropriateness of supplies, deployment rates, and whether clinics could integrate materials into care delivery.

Takeaway: Do not confuse output KPIs with outcome KPIs. Output tells you what happened; outcome tells you whether it mattered.

2. Build a logic model from inputs to impact

A useful performance framework should map the full value chain:

Inputs → Activities → Outputs → Outcomes → Impact

For GHC, the model looked something like this:

Inputs: donated medical supplies, volunteer time, partner relationships, logistics supportActivities: sourcing, sorting, coordinating shipments, partner vetting, distribution planningOutputs: supplies shipped, clinics supported, partner organizations reachedOutcomes: increased availability of usable medical supplies, improved clinic capacity, more procedures supportedImpact: expanded access to healthcare for displaced and underserved populations

This kind of logic model helps identify where assumptions may be weak. At GHC, we learned that sending supplies did not automatically produce impact. Supplies had to match local needs, clear logistics barriers, reach trusted partners, and be usable in the clinical setting.

Takeaway: A logic model should not be decorative. It should be used to test whether each operational step is actually producing the next intended result.

3. Measure beyond internal activity

Many organizations are strongest at tracking internal performance data: budgets, staff time, inventory, case counts, completion rates, and deliverables. These are important administrative metrics, but they do not fully capture results.

The harder and more valuable data often sits outside the organization. For GHC, internal metrics could tell us how much we shipped. External outcome data told us whether partners used the supplies, whether the supplies were appropriate, and whether they strengthened service delivery.

That required feedback loops with clinics and humanitarian partners. It also required accepting that some impact measures would be imperfect. Long-term health outcomes were difficult to track across international contexts, so we used credible proxy indicators, such as clinic utilization, procedure volume, and supply deployment.

Takeaway: Push the measurement system beyond internal outputs and toward external outcomes.

4. Use KPIs for learning, not just accountability

Performance management should not operate only as a compliance or accountability mechanism. In complex social-sector work, underperformance is often not caused by lack of effort. It is usually caused by unclear assumptions, capacity constraints, weak infrastructure, or changing field conditions.

At GHC, when a distribution model underperformed, the question was not simply whether someone met a target. The question was why the KPI moved the way it did. Was the issue partner capacity? Customs delays? Poor fit between supplies and local needs? Weak follow-up? Insufficient training?

By using KPIs diagnostically, we could adjust the operating model rather than simply report the shortfall.

Takeaway: KPIs should support adaptive management. The goal is not just to monitor performance, but to improve it.

5. Align strategy, capacity, and stakeholder support

A strong initiative needs more than good intentions. It needs strategic alignment between value, organizational capacity, and stakeholder support.

At GHC, some expansion opportunities appeared mission-aligned but were not operationally sound. We had to ask:

• Do we have the capacity to execute well?
• Do partners have the infrastructure to use the resources?
• Is there enough local support to sustain the intervention?

Performance data helped us avoid overextending.

Takeaway: Use performance data to guide strategic prioritization, not just retrospective reporting.

Conclusion

Social-sector performance management is not about collecting more data. It is about building a disciplined system that connects inputs, activities, outputs, outcomes, and impact.

GHC taught me that scale is not the same as effectiveness. A high output number may look impressive, but it only matters if it produces meaningful outcomes for the communities served.

The core leadership shift is moving from “What did we do?” to “What changed because we did it?”

Author Bio: Jenine Saleh Esq. Executive Director, Global Health Conscious NFP