Measuring Impact of Nonprofit Projects

Participatory impact assessment (PIA) has emerged as a transformative approach to evaluating development and humanitarian projects, placing communities at the center of assessing changes in their livelihoods. This Participatory Impact Assessment Design Guide provides a structured, adaptable framework for practitioners to measure project outcomes effectively. By emphasizing community-defined indicators and participatory methods, it moves beyond traditional evaluation metrics to capture nuanced, real-world impacts. Drawing from decades of field experience, the guide offers an eight-stage approach to PIA, blending participatory techniques with systematic analysis. From defining key questions and selecting indicators to triangulating results and validating findings with communities, this resource ensures comprehensive and credible impact evaluations. With practical examples ranging from livestock projects in Ethiopia to livelihood recovery initiatives in Zimbabwe, the guide demonstrates how PIA can reveal both expected and unexpected project impacts. Tailored for development practitioners, researchers, and policymakers, this guide is an indispensable tool for fostering accountability, enhancing learning, and driving policy reform. By adopting its principles, organizations can build stronger connections with communities, deliver more impactful interventions, and contribute to evidence-based decision-making across diverse contexts.

Are We Counting Activities… or Solving the Problem? 🤔 I was casually looking at the annual report of a housing nonprofit and something struck me that feels worth surfacing — not as criticism, but as an invitation to rethink how we measure impact. Most reports highlight numbers like: 🏨 Bed nights provided 🚿 Showers taken 🚽 Restroom visits These are important, human, real services. But when you add them up, you’re often looking at tens of thousands of “services”… supported by budgets in the tens of millions. In this case, the math works out to roughly $300 per service when you consider the sum of bed nights, showers, and restroom use. That’s a costly investment in the symptoms of homelessness rather than the resolution of it. And yet the one metric that matters most is often missing: 📉 Are fewer people experiencing housing insecurity next year than this year? This isn’t a failure of nonprofits — far from it. The people doing this work are heroic. The issue is structural: When we measure outputs (showers, meals, bed nights), we unintentionally reinforce a system where more services = more success. But in reality, more services often means the underlying problem has grown. It’s like applauding a hospital for a record number of emergency surgeries. Necessary? Yes. But it doesn’t mean the community is getting healthier. The real distinction in measurement is this: 📊 Outputs — how much activity the system is doing 🌱 Outcomes — whether the system is actually changing And in any system, what we measure becomes what we manage. Housing nonprofits don’t need blame — they need better dashboards. Imagine metrics like: 🏡 Housing retention at 6, 12, 24 months ⏳ Time-to-stability (from first contact to permanent housing) 📉 Fewer people experiencing homelessness overall 🔄 Reductions in repeat homelessness 🙋 Exits without reentry for 12–24 months 🌱 Improved health and wellbeing over time 📈 Income stability or gains after housing 🏘️ Availability of truly affordable, permanent units 💵 Lower total public cost per person (health, justice, crisis services) 🚑 Fewer emergency-system encounters year over year These metrics aren’t about activity — they’re about whether the system is actually healing. And the right list should be co-created by the community, for the community. And they might be part of a higher level set of metrics for a community which are supported by a thoughtful ecosystem of non profits each contributing to their part in the system of repair. I’m sharing this because I believe so many organizations and communities are ready for this shift. The sector is full of passionate people who want long-term change — and systems thinking gives us the tools to get there. If we update what we measure, we can update what we achieve. 💡✨

I asked a nonprofit CEO one question that made her go completely silent. Her organization: $12M budget. Award-winning programs. Thousands of families served over 20 years. The question: "Show me one family you've moved from needing your services to not needing them." She stared at me for 30 seconds. Then said the words that broke my heart: "Well... that's not really how we measure success." That's when I realized the uncomfortable truth about our entire sector: We're accidentally addicted to people staying broken. Think about it: → Success = more families in our programs → Growth = bigger budgets to serve more people → Impact = higher numbers on our annual reports But here's what we don't track: How many people graduated OUT of needing us? I watched this CEO's face change as it hit her. "So you're saying we should measure how many clients we lose?" Exactly. Here's the test that will make you uncomfortable: If your organization executed every program perfectly for 10 years, would the problem you're solving get smaller or bigger? If the answer is "bigger" - you might be treating symptoms while the disease spreads. The nonprofits creating real change? They're designing themselves out of business. → They measure food security achieved, not just meals served → They track permanent housing, not just shelter nights → They count families who no longer need services, not just families served I've seen organizations like Cradle Cincinnati reduce Black infant mortality by 34%. Like the Robinhood Foundation which increased graduation rates in low-income communities in NYC by 40%. And many more who’ve moved needle significantly towards their work not being necessary at all. They all asked the same question: "How do we make sure people don't need us anymore?" The uncomfortable truth? Community improvement without community ownership = Community removal with better PR. Program expansion without client graduation = Professional poverty management. I get it - people need help today. That work is essential and sacred. But if we're not measuring how many people move from dependence to independence, we might be part of the problem we're trying to solve. What would happen if your organization measured success by how many clients you "lost" to self-sufficiency this year?

Most non-profits struggle with impact measurement. The reason is simple: their work is multi-dimensional and inter-related, but measurement frameworks tend to reduce everything to a single axis: reach. Reach, as measured in the number of lives touched, has been the cornerstone of impact measurement because it is simple, measurable, objective and easy to comprehend. Sometimes, reach has been supplemented by the scale of impact on the life of each individual reached. But even then, it flattens the story. In practice, this creates three problems: 1. It reduces diverse types of work to the same metric. A think tank that shifts policy, a digital platform that reaches millions, and a grassroots NGO working deeply with 200 families are not comparable. 2. It creates pressure to show scale in numbers, rather than outcomes or systemic influence. 3. It makes it harder for funders and nonprofits to have a shared language of what impact looks like. During my six years at Omidyar Network, I worked on its impact measurement framework, one that I still find incredibly valuable. It captures both the direct impact that impact organisations have (as measured by reach, depth and inclusion), as well as the indirect impact that perpetuates their influence in other ways (for e.g., capital mobilised, replication of practices, institutional and policy shifts) I have found this framework incredibly useful because: 1. It moves beyond “just reach,” allowing nonprofits to tell their story in multiple ways. A for-profit impact start-up may focus on reach, but may also want to document its policy engagements with governments. 2. It works across organisational models, including grassroots NGOs, digital-first orgs, or policy think tanks. Each model may emphasize a different part of the framework but can still be placed on it. 3. It creates multiple valid pathways to being a high-impact organisation (e.g., low reach but high depth, or a pioneering idea that gets widely replicated). 4. It allows nonprofits to adapt the “indirect impact” dimension to their own context. For e.g., a think-tank may customise the policy impact pathway, based on its theory of change. Impact is rarely linear. A holistic framework like this creates space for nonprofits to be seen in their full richness, while still giving the ecosystem a common language to work with. #SocialImpact #ImpactMeasurement #Nonprofits #Philanthropy

As a donor, 90% of the grant proposals I read fail to include strong, measurable goals. If a proposal lacks strong goals, why should a donor approve it? Many organizations focus on their activities such as how many papers they’ll write, how many events they’ll host, or how many social media posts they'll create. But while important, these numbers alone don't create impact. Activities only create impact when they contribute to a clear and measurable goal. Foundations may call them outcomes, deliverables, or something else, but the real question is: Are your goals focused on the impact of your work, and are they both measurable and meaningful to your mission? Your goals should reflect what you hope to accomplish because of your work, not just the work itself, and they may vary depending on what you're trying to accomplish. For example, if your project involves writing research reports, the goal isn’t just to produce a certain number of reports. The real question is what impact will those reports have? Are you hoping to educate the public? Then tracking reads or media mentions might be the right measure. A goal here might be 10 media mentions in the next 6 months. Are you aiming for policy change? Then citations in legislative or academic discussions might be more relevant than raw readership numbers. In this case, a better goal might be 6 citations in the 3 months following the report's release. In your personal life, you might set a goal to go to the gym 3 times a week (an activity), but that doesn't tell you how long to go, what exercises to do, or why 3 times a week is effective. But if your goal is to gain 5 lbs of muscle in 6 months (the impact), you can start answering those questions with clarity. Start with your big-picture goal, then ask yourself: What would need to happen for this to become a reality? 🤔 How can we track progress toward that outcome? 📈 Don’t just set goals to satisfy a donor’s requirements. Make them meaningful to your mission. When your goals align with the change you want to see, measuring progress becomes not just a reporting requirement, but a powerful tool for driving impact.

The theme for IDinsight's 2024 year-in-review is "Innovating for Inclusive Impact". The review showcases inclusivity in the process of creating impact: new tools and frameworks that IDinsight teams have developed to make data more accessible to decision-makers (Ask-A-Metric), to help organizations self-diagnose M&E needs and focus on the highest-priority M&E activities (M&E Health Check, Impact Measurement Guide), to include the voices & experiences of program recipients in design & implementation (Dignity Initiative), and other very cool innovations. Anyone interested in the frontier of data methods & AI tools for social impact is definitely encouraged to take a look. https://lnkd.in/dnTRvNTM   But another interpretation of "inclusive impact" could be "programs or interventions that benefit all participants." I'll admit that when doing impact evaluations it's easy to focus on the big headline average treatment effects. We'll usually do some subgroup analysis that shows we can't reject the null hypothesis of equal treatment effects for men vs women or older vs younger participants. But this is hardly suggestive of a program being impactful across the board. I've seen several evaluations where, when you dig into the data, the statistically significant treatment effects in the headline disappear when you omit the top 5 or 10% of performers.   For this reason, one of my favorite graphs from this year (below) comes from our RCT of the Luminos Fund program, where we measured the impact of their accelerated learning program for out-of-school children in Liberia. This graph was conceptualized by my colleague Mico Rudasingwa as a way of exploring impacts across the study sample. The graph shows the average change in reading fluency (words per minute) in each of the 98 communities in our study from baseline to endline, sorted by communities with the largest change (top) to smallest change (bottom), and color-coded by whether the community received the Luminos program (red) or not (blue).   Technically we're not pinpointing the program impact for each community; we don't know the counterfactual for each community, and at least for some communities the counterfactual would involve some improvement in reading ability (after all, learning gains do vary in control communities). But to me at least it's pretty convincing that children are benefitting from the program across the board. Not only are average learning gains positive in every community that got the Luminos program, but nearly every program community has larger average gains than nearly every control community. I've rarely seen such clear evidence of a program having inclusive impact.   If you're interested to dive into the data yourself, check out our interactive visualizations of the RCT results posted earlier this year! https://lnkd.in/dhf4wGbd

🧠 Logic models might seem academic—but they’re really about one thing: serving people well. If you work in a nonprofit or local agency, you probably don’t have much time to slow down. The need is urgent. The workload is real. And taking time to build a logic model can feel like one more thing on an already packed list ✅ A logic model isn’t red tape—it’s your blueprint for doing the work well, without wasting time, trust, or resources. 🚫 BEFORE (no logic model): “We’re going to run financial literacy workshops for youth in transitional housing.” Sounds good. But what does that actually mean? Who’s running them? How often? Will youth show up? What happens after the workshop? How do we know it worked? Too often organizations receive funding then scramble to figure the program out after the fact. ✅ AFTER (logic model clarified): “We offer trauma-informed financial coaching for youth in transitional housing—through weekly group workshops and 1:1 follow-up. Youth get transportation support, meals, and real-world tools to build savings, avoid predatory lending, and access banking. We track confidence levels, savings rates, and stability six months after program completion.” Same mission. But now it’s a real program—grounded in logistics, accountability, and human needs. Here’s the logic model framework that helps you get from “vague idea” to “impactful plan”: 🔧 INPUTS – What are you working with? Do you have trained, trauma-informed facilitators? A safe space to meet? Bus passes, food, or translation? Are youth involved in design? 📚 ACTIVITIES – What exactly will happen? Are the workshops weekly? Are they interactive or lecture-based? Is there any follow-up? Will youth leave with tools they can use the next day? 📈 OUTPUTS – What will you count? Number of sessions. Number of youth served. Number who return for more than one session. These are the basic metrics of reach. 🔄 OUTCOMES – What will change in people’s lives? More confidence with money. Fewer overdrafts. More consistent savings. Knowledge that’s used, not just heard. 🌍 IMPACT – What long-term difference are you aiming for? Are youth more stable financially in 6–12 months? Less likely to cycle through housing insecurity? Do they feel in control of their futures? 👇 A lot of orgs don’t plan—because they’re under-resourced, moving fast, and trying to meet real needs. But skipping the planning stage can cost more in the long run. #SocialImpact #NonprofitLeadership #HumanCenteredDesign #ProgramPlanning #GovernmentInnovation #Evaluation #WatkinsPublicAffairs

Things I’m Not Falling For (Nonprofit Edition) 🙅🏾♀️ ❌ “Participant numbers = impact.” Serving 500 people isn’t impact — it’s activity. Funders don’t care how many walked through your doors. They want to know how lives changed. “We had 200 participants” tells me nothing about transformation. “85% secured stable employment within 6 months” tells me everything. Stop counting bodies. Start measuring change. ❌ “You can figure out evaluation later.” Wrong. Impact measurement isn’t an afterthought — it’s baked into your program from day one. Nonprofits scrambling to pull together inconsistent reports usually waited too long. The ones landing multi-year, six-figure grants? They planned for impact before serving their first participant. ❌ “Good work speaks for itself.” Not today. Your passion got you the first $50K grant. Scaling to $250K+ requires proof. Funders invest strategically, not emotionally. Your heart might be in the right place, but your data better be too. ❌ “Relationships are all you need for funding.” Relationships open doors. Impact keeps them open. I’ve seen nonprofits raise millions on founder connections — then lose funding when leadership changes or tough questions come. The survivors? They built programs that prove their worth beyond personal ties. Ready to move beyond myths and build real impact? Let’s talk.

Impact vs. Output: A CSR Story Told Through Bollywood & Real-World Examples. A simple thought!!! Colleagues from development sector keep discussing about the word impact, output, outcome etc. Half the time we tend to interpret and interchangeably use these words. In a way I often get tired of these words. In CSR, “impact” is often misunderstood. Many reports showcase big numbers but does that truly define impact? Let’s break it down using a Bollywood analogy and real-world examples. 🎬 Scene 1: Output vs. Outcome vs. Impact (A “Taare Zameen Par” Moment) TZP illustrates the difference between output and outcome. Ishaan, a child with dyslexia, is overwhelmed by relentless academic demands, resulting in academic failure. However, Nikumbh Sir's intervention shifts the focus to personalized outcomes 📝Output: Conducting a special training session for Ishaan 📝Outcome: Ishaan's reading skills improve dramatically 📝Impact: Ishaan develops a lasting self-belief and cultivates a genuine love for learning 👉 To demonstrate meaningful impact in CSR, we must move beyond simply reporting outputs, such as '10,000 children given books,' and instead measure sustained improvements in learning 🎬 Scene 2: Real CSR Example – The ‘Toilet Dilemma 🚾Output: A total of 100,000 toilets were constructed. 🚾Outcome: If daily usage reaches 70% or more, the project will have demonstrably succeeded. 🚾Impact: If, within five years, open defecation is no longer practiced, diseases are significantly reduced, and communities have integrated hygiene practices into their daily lives, then the project will have achieved its intended impact. 🎬 Scene 3: Any robust micro finance program 📄Output: ₹10,000 for a woman as a microloan. 📄Outcome: She starts a small micro business. 📄Impact: In 5 years, she has a stable income, her children finish school, and the family moves out of poverty. This may be a tall or hypothetical scenario. However , putting it here for a framework of reference 👉 Sustainable livelihoods create multi-generational impact. So, How Do We Measure Impact in CSR *? A robust monitoring and evaluation (M&E) system must include: 📖Baseline & Endline Surveys 📖Control Groups – Did the change happen because of CSR, or were other factors involved? 📖Qualitative Stories + Quantitative Data – A mix of human narratives + hard data is key. 📖Long-Term Tracking 📖Independent Assessments The CSR Mindset Shift: From Reports to Real Change At its best, CSR is not about “100,000 beneficiaries” but about lasting transformation and building positive community 👉🏼For NGOs: Shift from just reporting numbers to showcasing life-changing stories. 👉🏼For Corporates: Move beyond annual compliance and fund long-term interventions. When we align vision with measurement, we don’t just create CSR reports—we create real, measurable, and lasting impact. 💬What’s your take? Pic: Pradan and other partners Saroj Mahapatra

What if we measured impact not just to prove what works—but to learn what matters? At the Skoll Centre for Social Entrepreneurship, we’ve been asking this question with our partners at the The Old Fire Station, Oxford. The result is the Meaningful Measurement Playbook, a resource designed to help social impact organisations and funders navigate complexity with curiosity and purpose. 📽️ In this short video, I share the thinking behind the playbook and why we believe learning, not just accountability, should be at the heart of impact measurement. Meaningful Measurement is grounded in four key principles: 1️ Treat learning as the central purpose 2️ Centre the voices of those closest to the work 3️ Engage openly with all stakeholders including funders 4️ Embed learning in organisational culture Whether you're a social entrepreneur or a funder, I hope this approach helps you reflect, adapt, and deepen your impact in today’s fast-changing world. 🎥 Watch the video and 📘 Explore the playbook: https://lnkd.in/eHratwKg #MeaningfulMeasurement #ImpactMeasurement #LearningCulture #SocialImpact #SkollCentre #SystemsChange #EquityInImpact #LeadershipInPractice