Evaluation Framework 2025: How We Work

Purpose

The Life You Can Save envisions a world where there is no global poverty.1 Our mission is to improve the lives of people living in poverty by changing the way people think about and donate to charity. As such, we seek to understand the complex nature of global poverty and use robust evidence to recommend highly impactful nonprofits and engage with donors on best practices.

This evaluation framework describes the process of how we identify the nonprofit organizations we recommend. Through our approach we provide high impact giving recommendations. By ‘high impact’ we mean opportunities that bring substantial additionality to work towards alleviating extreme poverty. Our goal is to provide donors assurance that giving money to an organization in our list of recommended nonprofits is a great bet, highly likely to do good, and cheap enough to be scalable, durable, and have a deep impact on people’s lives. 

Our approach

In a nutshell, our approach consists of:

  1. Problem: Identifying the most pressing problems behind extreme poverty.
  2. Intervention: Identifying effective interventions to address these problems.
  3. Organization: Identifying the best nonprofit organization implementing these interventions.

We first define priority geographies where we focus on. These are geographies with highest multidimensional poverty. After taking steps 1-3 described above, we recommend nonprofit organizations with the most compelling evidence of impact and organizational quality. 

1 The Life You Can Save defines extreme poverty using the global multidimensional poverty index described in section 1. A person is identified as experiencing extreme poverty if they are deprived in a third or more of ten (weighted) indicators.

How we define poverty

At The Life You Can Save, we use the global Multidimensional Poverty Index (MPI) to define poverty. Apart from informing our understanding of poverty it guides our decisions on what geographies to focus on.

Leading experts at the Oxford Poverty and Human Development Initiative2 and the United Nations Development Programme have collaborated with stakeholders to develop a comprehensive understanding and measurement of the multidimensional and interlinked nature of poverty (UNDP and OPHI, 2022). This resulted in the MPI which has been included in the UNDP’s Human Development Report since 2010. The MPI provides a complex view and indicators of poverty and wellbeing, and moves beyond an income threshold to define poverty. Understanding the experience of poverty can be challenging for individuals in high-income countries. Many of us find it difficult to truly understand what it means to live on less than US$3 a day.3  Poverty encompasses various dimensions that profoundly impact individuals’ freedom to lead the lives they aspire to. 

The MPI defines three fundamental dimensions of poverty: health, education, and standard of living. These dimensions were selected through a comprehensive process that involved participatory surveys among individuals experiencing poverty, expert consultations, alignment with public consensus (including the Universal Declaration of Human Rights and Sustainable Development Goals (SDGs)), and philosophical examination (OPHI, 2024). The Life You Can Save uses the MPI given its robustness, broad consensus within expert communities, and availability of reliable data.

The MPI breaks down the three dimensions of poverty into ten indicators.4 A person is identified as being multidimensionally poor if they are deprived in more than a third of the (weighted) indicators. The intensity of poverty is then measured by the percentage of deprivations they are simultaneously experiencing. Table 1 shows the indicators and their weights.

 

2 Sabina Alkire and Maria Emma Santos.
3 In June 2025 the World Bank updated the extreme poverty line (for low income countries) to $3 per day. See ‘Measuring Poverty’ blog by the World Bank.
4 The MPI includes only two education indicators: years of schooling and school attendance. The Life You Can Save recognizes that access does not guarantee learning. Many countries now have high school participation, but also high levels of learning poverty. Learning poverty is defined as the share of children unable to read and understand a simple text by age 10. We therefore also support focused on improving foundational learning.

Table 1. The Global Multidimensional Poverty Index
Dimension Indicator Deprived if living in a household where… Weight SDG Area
Health (1/3) Nutrition Any person under 70 years of age for whom there is nutritional information is undernourished. 1/6 SDG 2: Zero Hunger
  Child mortality A child under 18 has died in the household in the five-year period preceding the survey. 1/6 SDG 3: Health and Well-being
Education (1/3) Years of schooling No eligible household member has completed six years of schooling. 1/6 SDG 4: Quality Education
  School attendance Any school-aged child is not attending school up to the age at which he/she would complete class 8. 1/6

SDG 4: Quality Education
Living Standards (1/3) Cooking fuel A household cooks using solid fuel, such as dung, agricultural crop, shrubs, wood, charcoal, or coal. 1/18 SDG 7: Affordable and Clean Energy
  Sanitation The household has unimproved or no sanitation facility or it is improved but shared with other households. 1/18

SDG 6: Clean Water and Sanitation
  Drinking water The household’s source of drinking water is not safe or safe drinking water is a 30-minute or longer walk from home, roundtrip. 1/18 SDG 6: Clean Water and Sanitation
  Electricity The household has no electricity. 1/18 SDG 7: Affordable and Clean Energy
  Housing The household has inadequate housing materials in any of the three components: floor, roof, or walls. 1/18 SDG 11: Sustainable Cities and Communities
  Assets The household does not own more than one of these assets: radio, TV, telephone, computer, animal cart, bicycle, motorbike, or refrigerator, and does not own a car or truck. 1/18 SDG 1: No Poverty

Source: Reprinted from What is the Global MPI by the Oxford Poverty and Human Development Initiative (OPHI), retrieved September 16, 2025, from https://ophi.org.uk/what-global-mpi. Note that content is identical to the original source.

How we identify those most in need

We use the MPI to define our priority geographies. The Life You Can Save focuses on countries and regions characterized by the highest levels of multidimensional poverty, as well as highest numbers of people experiencing it. This is determined by checking the most recent MPI values released by the Oxford Poverty and Human Development Institute (OPHI) and UNDP (OPHI and UNDP, 2024). This contains information on MPI globally and ranks countries across MPI dimensions and indicators. The MPI shows that the greatest concentration of multidimensional poverty is found in countries in Sub-Saharan Africa5  and South Asia. Based on this, The Life You Can Save has defined these two regions as its priority geographies. 

In these areas, a significant portion of the population grapples with deprivation across multiple dimensions, making them particularly vulnerable and underserved. We believe that focusing our efforts here is essential, as it indicates a concentration of structural barriers hindering human development. These obstacles span a wide spectrum, including issues such as limited access to healthcare, educational opportunities, and economic prospects, as well as deprivations stemming from entrenched social norms, conflicts, natural disasters, and the impacts of climate change.

 

5 This includes African countries located south of the Saharan Desert, namely those in Central Africa, Southern Africa, East Africa and West Africa.

1. The problem sphere

Within the context of multidimensional poverty, various underlying factors contribute to deprivation across its three dimensions. Our team conducts reviews of the scientific literature and data to identify the primary drivers of poverty within the most deprived regions. This enables us to gain a nuanced understanding of the critical challenges faced by vulnerable populations, whether they reside in specific locations, belong to particular ethnic groups, or fall within specific age groups. Moreover, we explore how poverty intersects with various household and community characteristics, ensuring a holistic analysis of the multifaceted nature of deprivation. Examples of key drivers are malaria and diarrheal diseases for child mortality, school costs and gender norms for education, and lack of access to markets or lack of capital for standard of living.

To identify problems to focus on, we rely on three main criteria: 6

  • Scale. Does the problem affect a vast majority of people in a given context? This ensures that we consider the most important drivers behind poverty, including those that affect a vast number of people as well as those that cause extreme suffering.
  • Solvability. Would additional resources go a long way towards solving the problem? This keeps our focus on problems that can in principle be solved using existing interventions that are backed by rigorous evidence.
  • Neglectedness or additionality. Will donor contributions add substantially to the current resources available to address the problem (i.e., by providing a given number of dollars or person-hours). This keeps our focus on poverty drivers that get insufficient attention, so that our contribution has a high impact and does not crowd out other efforts.

Further details on the process we use to determine which problems to focus on and the types of interventions we have identified as the most effective strategies for addressing them can be provided upon request. Our prioritization is based on a scoping review of the literature and on discussions with experts. General details on how we select high-impact interventions are described in the next section.

6 This is the Importance–Tractability–Neglectedness (ITN) framework used by most Effective Altruism organizations to identify priority cause areas (e.g., Founders Pledge, Give What We Can, Open Philanthropy). Its three components can be expressed as: [I]mportance (or scale) = utility gained / % of problem solved; [T]ractability (or solvability) = % of problem solved / % increase in resources; and [N]eglectedness = % increase in resources / dollar (Kwa, 2022). Then I x T x N = utility gained / dollar.

2. The intervention sphere

This section explains how The Life You Can Save selects interventions to focus on.

Several interventions are available to tackle key drivers of poverty, and their design can vary according to contextual factors, the specific needs of beneficiaries, and the philosophical principles of implementing organizations. For example, some interventions involve active participation of beneficiaries in the design phase, tailoring solutions to their unique circumstances, while others may originate in different settings and be applied across various geographical areas. It’s important to recognize that there isn’t a one-size-fits-all approach to driving change and creating impact. However, extensive research has indicated that certain interventions can generate higher levels of impact than others.

At The Life You Can Save, we identify the most impactful solutions to key drivers of poverty. To do so, we assess interventions across the following criteria:

  • Evidence: Is the intervention backed by rigorous evidence? 
  • Cost: Is the intervention cost-effective?
  • Scale: Does the intervention have the potential to reach a high number of beneficiaries? 
  • Depth: How many MPI indicators does the intervention address? 
  • Durability: Is the change achieved by the intervention sustained over time? 
  •  

While all of these criteria are important, we place particular emphasis on evidence and cost-effectiveness. The interventions we support must be backed by rigorous evidence. They must also make the best possible use of limited resources. Scale, depth, and durability then help us distinguish between strong options. Scale is closely tied to cost, since it depends both on the complexity of the intervention and on the cost per beneficiary. For clarity, we treat it as a separate criterion.

We first look for evidence that an intervention works. Only if there is evidence of impact, we then assess the quality of that evidence. To evaluate the quality of evidence behind an intervention, we use an adapted version of the GRADE framework (Grading of Recommendations, Assessment, Development and Evaluation) (GRADE, n.d.; Schünemann, et al., 2013). GRADE is widely applied in health research, but we adjust it for the broader range of social interventions supported by The Life You Can Save. Our approach makes the assessment more systematic and transparent, while allowing space for judgment where evidence is mixed or incomplete.

We begin with the study design, which forms the foundation of our rating. Randomized Controlled Trials (RCTs), systematic reviews, and meta-analyses are considered the strongest forms of evidence, followed by quasi-experimental studies. Non-experimental studies such as observational research sit lower in the hierarchy. However, study design alone does not determine quality. We also consider four additional factors: risk of bias, imprecision, inconsistency, and indirectness. Together, these criteria determine whether an intervention’s evidence base is rated high, medium, or low.

While RCTs provide a strong way of isolating causal effects, they have important limitations. They are often impractical for large-scale or long-term interventions, and poorly suited to evaluating policy advocacy or systemic reforms. They can also miss the mechanisms through which change occurs. For these reasons, we consider quasi-experimental and non-experimental evidence as well, particularly when RCTs are unavailable or unsuitable. These complementary sources of evidence are important for understanding external validity and for assessing how interventions work in practice.

When relying on non-experimental evidence, we apply it carefully. Monitoring data, case studies, and beneficiary surveys are not included in the intervention evidence score, since they cannot easily establish causality. Instead, we use these forms of evidence to assess organizational quality and to better understand context and implementation (see section on ‘The organization sphere’). This distinction helps us avoid overstating causal claims while still learning from a wider body of data.

An important consideration is the external validity of studies. Results from one country or time period may not hold in another. For example, the impact of cash transfers in Kenya in 2015 may not be the same as in Liberia in 2022. Evidence is strongest when it directly reflects the intervention as delivered by the nonprofit organization we are evaluating. When direct evidence is unavailable, we rely on indirect evidence and judge how relevant it is to the nonprofit’s specific context.

We also look for convergence across methodologies. No single study or design is perfect. By comparing results across different study types and contexts, we can mitigate the biases of any one method, such as publication bias, limited replication, or measurement error. This approach allows us to balance rigor with pragmatism. For example, for direct service delivery of medicines, RCTs may be the most reliable. For interventions that involve government partnerships, such as universal salt iodization, correlational studies, legislative reviews, and policy analysis may provide better insights.

We assess cost-effectiveness using harmonized metrics that allow us to compare impact within sectors. These are widely used in the literature and provide a standardized basis of comparison for interventions that otherwise produce different outcomes. Our thresholds are informed by global health and education research (e.g., WHO-CHOICE,  Angrist et al. (2025)), as well as benchmarks used by other philanthropic evaluators (e.g., Livelihoods Impact Fund). Our aim is to assess the relative cost-effectiveness of similar types of interventions in health, education, and living standards, ensuring our recommendations are both rigorous and comparable.

  • Health: Disability-Adjusted Life Years (DALYs) averted.
    • For health interventions, we calculate the cost per DALY averted. Cost per DALY averted measures how much it costs to give someone one additional year of healthy life by preventing or reducing disease and disability. DALYs combine years of life lost to premature mortality with years lived with disability, providing a comprehensive measure of disease burden. Interventions that avert one DALY for less than $100 are considered highly cost-effective.
  • Education: Learning-Adjusted Years of Schooling (LAYS) gained.
    • For education interventions, we calculate the number of LAYS gained per $100 spent. Cost per LAYS tells us how much it costs to give a child one year of quality education. This is a year in which they actually learn, not just attend school. LAYS adjust years of schooling for learning quality, making results comparable across countries. Interventions that generate two or more LAYS per $100 are considered highly cost-effective.
  • Living Standards: Return on Investment (RoI). 
    • For living standards interventions, we use RoI to capture the economic returns generated per dollar spent. Interventions that achieve at least a 5x return considered highly cost-effective.

In practice, we use organizational data, published studies, and systematic reviews to estimate these metrics, while applying adjustments based on the literature (for example, to account for implementation at scale). This allows us to benchmark each program against cost-effectiveness thresholds, while recognizing the limitations of available data. 

We recognize that standardized metrics are better suited to some interventions than others, and they may not capture every dimension of impact. Nonetheless, using common benchmarks is essential to ensure transparency and comparability across interventions, and we view this as the best available approach for The Life You Can Save. 

We also recognize that separate sector-specific metrics limit direct comparability across health, education, and living standards. We accept this limitation, as we believe a single universal metric would risk oversimplifying important differences and introducing subjective judgments. Instead, we prioritize using measures that are meaningful within each sector, reflecting the distinct ways interventions improve people’s lives. At the same time, we note that these domains are deeply interconnected: progress in health, education, or living standards often generates spillovers into other areas (Suppa, Alkire, & Nogales, 2022; Alkire et al., 2021). This directly links to our understanding of poverty which is based on the Multidimensional Poverty Index (MPI), namely that poverty is multidimensional and driven by interlinked factors. Recognizing these interlinkages is important, even if our cost-effectiveness metrics focus on sector-specific outcomes. 

Even if a problem is large, not all interventions are equally scalable. We examine whether an intervention can be replicated across geographies, populations, or contexts, focusing in particular on its implementation complexity. For this, we draw on Andrews, Pritchett and Woolcock’s (2017) typology of tasks by capability required, which helps us assess how demanding an intervention is to deliver in practice.

Tasks that fall into categories like policy-making or logistics (e.g., salt iodization or vaccine distribution) are generally more straightforward to scale. Interventions that require implementation-intensive service delivery (such as curative care) pose more challenges, while those that involve behavior change enforcement or tackle “extremely hard” problems (such as preventative health campaigns or complex systems reforms) are the most difficult to expand. Based on this framework, we classify scalability as high, medium, or low. This qualitative assessment helps us judge whether an intervention has the potential to reach a large number of beneficiaries without a significant loss of effectiveness.

To assess depth, we look at how many Multidimensional Poverty Index (MPI) indicators an intervention addresses. The MPI captures overlapping deprivations that people experience across areas such as health, education, and living standards. Interventions that affect multiple indicators are considered deeper, as they reduce poverty across several dimensions at once, while those targeting a single indicator are narrower in scope.

We assess durability by looking at evidence of long-term impact. We review meta-analyses, systematic reviews, and follow-up studies of RCTs or other evaluations. The goal is to know whether the effect of an intervention lasts over time or whether it fades quickly and needs to be repeated every year. We classify interventions as short-term if the impact fades within a year. They are medium-term if there is evidence that effects remain for at least one year. They are long-term if the evidence shows the impact lasts three or more years, or if the intervention creates benefits in future outcomes. Whenever possible, we reference studies that confirm sustained impact.

3. The organization sphere

Our goal is to identify outstanding nonprofit organizations focused on high impact solutions. To streamline our selection process, we begin by compiling a longlist of nonprofits that operate in the areas with the highest need. We build a pool of nonprofits using different sources, and then vet them in-house before considering whether or not to recommend them. Our data come from the organizations’ program documents, review of studies by independent evaluators, conversations with sector experts and philanthropists, and field visits by us or our partners. Through this process we assess an organization’s capacity to be impactful, including its overall quality.

It is paramount to assess an organization’s operational capacities to ensure impact. It is not enough to implement an evidence-based intervention that is cost-effective. Organizations also need the resilience to endure political or economic shocks, and the ability to adapt to changing circumstances while continuing to deliver results. Strong organizational quality helps ensure that interventions remain effective and sustainable over time.

We only consider nonprofit organizations implementing interventions that are backed by rigorous evidence of impact and that are cost-effective. Furthermore, we define high-impact nonprofits as those that consistently achieve measurable, evidence-supported, and sustainable improvements in the lives of individuals, which would not have happened without their contribution. We support these organizations to ensure that every dollar invested in them can go the furthest in generating positive outcomes and reduce suffering. Two key questions we examine are:

  • Is there evidence of impact of the intervention when implemented by the organization? 
  • Is the organization able to implement the intervention at a cost-effective rate?

Since interventions are implemented by the nonprofit organizations we recommended, there is overlap between the intervention and organization spheres. Therefore, as part of our consideration of nonprofits to recommend, we pay close attention to an intervention’s: scale, depth and durability. 

From an organizational perspective the questions to assess these elements are formulated as:

  • Does the organization have the ability to operate at scale?
  • Does the organization’s work reflect depth by addressing multiple MPI indicators?
  • Is there evidence that the intervention delivered by the organization has durable impact?

In our evaluation and monitoring of nonprofit organizations, we take into account these three dimensions and consider how organizations can achieve similar levels of impact through different strategies. We illustrate this in Box 1. 

 

Box 1. Organizational impact across scale, depth, and durability
Examples:

 

1. Nonprofit A provides a narrow intervention that it distributes to millions of people. The intervention prevents people from contracting disease X but it cannot help beneficiaries in other ways. The intervention must be repeated every year. Their impact can be visualized as follows:

 

 2. Nonprofit B provides an educational programme that supports beneficiaries in various aspects of their life. The intervention has proven to increase wellbeing, income and health of beneficiaries. The cost of the intervention is high and the organization cannot directly reach vast amounts of people directly. A longitudinal study showed intergenerational effects of the intervention. Their impact can be visualized as follows:

 

Nonprofits A and B are impactful in different ways, and both would be considered for recommendation. 
 
Our recommendations are limited to organizations implementing interventions that are rigorously evidence-based and highly cost-effective. For this reason, evidence and cost-effectiveness are not shown in the figures, which focuses on the additional criteria of scale, depth, and durability.”


Even highly effective interventions require organizations with the capacity to implement them broadly and reliably. We therefore assess an organizations’ capacity to operate at scale by looking at the number of people reached each year and whether the organization meets its yearly targets. A strong track record of delivery shows that the organization can reach large populations consistently and reliably. We classify organizations into three levels. Low scale means reaching 100,000 to 1 million people per year. Medium scale means more than 1 million to 2 million. High scale means more than 2 million. These thresholds reflect our experience with the types of organizations we recommend and are consistent with broader frameworks on organizational maturity and absorptive capacity.

In addition to assessing scale, depth, and durability, we also consider a broader set of elements when evaluating organizations. We prioritize those that operate in The Life You Can Save’s priority geographies. We also assess whether an intervention that is cost-effective in general remains cost-effective when implemented by the organization in its specific context. Beyond this, we examine the following elements:

  • Is there evidence of high organizational quality?
  • Does the organization uphold beneficiaries’ dignity in their work?
  • How is the organization embedded in local systems?
  • What is the organization’s funding need?
  • What are the risks the organization faces?

Next, we provide details on what each of the above entail.

Assessing organizational quality is essential because even strong, evidence-based interventions require organizations with the leadership, systems, and adaptability to deliver them effectively in real-world contexts. Randomized Controlled Trials and other forms of impact evaluation tell us whether an intervention works, but they do not show whether a particular organization can implement it reliably across different settings. To address this gap, we draw on insights from realist evaluation (Pawson & Tilley, 1997; Deaton & Cartwright, 2017; Vivalt, 2020), which emphasize understanding what works, for whom, and under what circumstances. This lens highlights the importance of context, organizational capabilities, and the mechanisms through which change occurs.

Our assessment of organizational quality focuses on an organization’s impact creation logic—the knowledge, processes, and systems that drive its ability to create and sustain meaningful change (Seelos & Mair, 2017). We evaluate whether organizations demonstrate a deep understanding of the problems they address, align their mission closely with their solutions, and adapt interventions effectively to context. Leadership and team capacity are also central: we look at the stability, experience, and structure of teams, and at whether organizations set and consistently achieve clear milestones.

We assess organizational quality across five main criteria:

  • Problem space: Understanding of the problem and its root causes, including socio-political and cultural dynamics.
  • Identity space: Alignment between mission, problem, and solutions, and resilience in pursuing long-term goals, and articulation of a clear, well-developed theory of change that links activities to outcomes.
  • Solution space: Evidence that interventions are contextually appropriate, adaptable, and supported by strong monitoring and feedback systems (from beneficiaries and partners).
  • Team and leadership stability: Depth, experience, and continuity of leadership and staff. We assess whether leadership combines technical expertise with proximity to the communities served, whether decision-making is inclusive, and whether the organization has succession planning to ensure continuity.
  • Commitment to learning and improvement: A culture of reflection, adaptation, and transparency, supported by robust monitoring, evaluation, and learning (MEL) systems and strong feedback mechanisms that generate evidence and drive program improvements.

We combine desk research, document review, and interviews with organizational staff, as well as external perspectives from donors and partners, to score organizational quality on a three-point scale (high, medium, low). A high-quality organization shows a strong understanding of the problem, effective and adaptable solutions, stable leadership, and a demonstrated commitment to continuous learning.

Research shows that international development beneficiaries deeply value respect for dignity and that, unfortunately, they are also frequently denied. Even when material needs are met, beneficiaries can come away from interactions with service providers feeling mistreated, ignored, or disempowered. For The Life You Can Save, any charity considered impactful must also be one that respects the dignity, priorities, and agency of those it serves.

The Life You Can Save partners with IDInsight’s Dignity Initiative to integrate dignity in our evaluation framework. IDinsight describes three main pathways to consider when reflecting and operationalising the concept of dignity. These are:

  • Equity: Treat beneficiaries as if they were equals and reduce power asymmetries.
  • Agency: Give beneficiaries choices and meaningful chances to consent.
  • Representation: Offer beneficiaries the chance to be and feel seen by the institution delivering aid.

We assess nonprofits on how well they respect dignity alongside other impact factors in deciding whether to recommend them to donors. In our evaluations, we consider the following criteria:

  • Priority: When looking at a nonprofit’s website, reports or listening to a presentation, evaluate whether the service they are providing is something people in the area value and are calling for. Most importantly, how does the charity know this is something people need and want? A respectful nonprofit makes reference to participatory processes of preference elicitation, willingness to pay and transparently discusses trade offs with other priorities. 
  • Interactions with beneficiaries: A respectful nonprofit references concrete steps taken to ensure that beneficiaries are treated fairly and in the way they wish to be treated. Donors should look out for references to the three pathways in nonprofit’s websites, social media, presentations and reports. 
  • Culture and People: Internal culture is key to developing respectful interactions both within and outside the organization. Donors should look for examples of concrete steps organizations take to ensure staff are treated with respect.
  • Listening and Learning: A highly impactful and respectful organization needs tools and mechanisms to know that they are treating people with respect. A nonprofit that respects people’s dignity mentions ongoing and proactive measurement of beneficiaries’ experiences, feedback mechanisms, and shows a track record of adapting their programs based on these mechanisms. 
  • Safeguarding: A respectful nonprofit has clear and accessible safeguarding policies in place to protect against harm or misconduct. It provides trusted reporting mechanisms that beneficiaries can use to raise concerns and demonstrates how it acts on them.
  • Representation: A respectful nonprofit portrays beneficiaries as active agents in their own lives. It avoids paternalistic or disempowering depictions in its communications and highlights dignity, agency, and equality in how it represents its work.

For interventions to be sustainable, nonprofits must be well connected to the communities and systems in which they operate. We therefore assess how organizations fit within the local ecosystem. This includes looking at whether they are locally led, how they engage with community stakeholders, and the extent to which they collaborate with governments. Organizations that are strongly embedded in their local context are more likely to understand community needs, adapt to challenges, and secure the buy-in needed for lasting impact.

We also assess whether organizations have strategies to strengthen local systems over time. This includes efforts to build government capacity, foster community ownership, or create pathways for markets and institutions to take responsibility for effective solutions. While philanthropy can sometimes create risks of dependency (Easterly, 2006; Moyo, 2009), it can also act as a catalyst for innovation and sustainability when paired with strong system-strengthening strategies (Barder, 2014). We prioritize nonprofits that demonstrate clear, operational strategies for embedding their work in local systems and reducing long-term reliance on external support.

When evaluating a nonprofit, we assess its funding needs to determine whether additional resources are likely to generate significant impact. Our assessment considers several factors:

  • Unrestricted funding: We look at what proportion of the organization’s budget is unrestricted. Unrestricted funding is particularly valuable because it gives organizations flexibility to allocate resources where they are most needed, invest in capacity, and adapt to challenges. A low proportion of unrestricted funds signals a high need.
  • Reliance on TLYCS and philanthropic funding: We assess whether the organization’s funding base is diversified or overly dependent on The Life You Can Save or philanthropy more broadly. Heavy reliance on a single source can compromise financial stability and autonomy, while a medium level of reliance suggests healthier diversification.
  • Funding urgency and runway: We consider whether the organization faces time-sensitive needs where delays in funding could disrupt services or lead to missed opportunities. As part of this, we assess the organization’s financial runway, or how long it can continue operating at its current scale without new resources. A short runway increases urgency and signals greater funding need.
  • Capacity to absorb funding: We examine whether the organization has the operational systems, leadership, and scaling plans in place to effectively manage additional resources and deliver impact.

We score organizations on each of these dimensions and then determine whether their overall funding need is high, medium, or low. Organizations with a high funding need, typically combine limited unrestricted funding, medium reliance on philanthropy, pressing financial urgency, and strong capacity to absorb new resources.

As part of our evaluation process, we also consider potential risks that could limit an organization’s ability to deliver services to a high standard and achieve the intended impact. This is a qualitative assessment based on all the evidence and documentation we review, combined with our professional judgment. The goal is not to exclude organizations because they face risks, but to be transparent about possible challenges and to identify areas that may require closer monitoring over time.

In assessing risks, we look at both internal and external factors. Internal risks may include organizational capacity, leadership, or systems that affect implementation. External risks may relate to stakeholder engagement, local or national politics, or other contextual challenges. By weighing these factors carefully, we aim to provide a balanced view of each organization’s strengths as well as the uncertainties that could affect its work. This helps us, and our donors, remain aware of potential vulnerabilities while recognizing the impact these organizations are achieving.

Recommending high impact nonprofits

After assessing the three spheres of interest (problem, intervention and organization), we award a score to the organizations under review. We use a weighted scoring model that covers the sixteen criteria that we examine. This helps us compare organizations and identify those scoring the highest. Based on this scoring, we develop a list of nonprofits that we can confidently recommend as high-impact funding opportunities. These cover all three dimensions of poverty, a wide range of theories of change, and interventions. The list is dynamic and subject to new evidence, contextual changes within the countries of operation, and organizational changes. More information on the weighted scoring model and thresholds we use to select top charities can be provided upon request.

Limitations and uncertainty

Our evaluation framework is designed to be rigorous and transparent, but it is not without limitations. All assessments involve a degree of uncertainty, which requires professional judgment from our team. The evidence and data available vary across interventions and organizations, and cost-effectiveness models are highly sensitive to assumptions. To reflect this, we communicate our recommendations with different confidence levels, indicating where we have stronger or weaker certainty. We aim to provide donors with the clearest possible information while acknowledging the limits of our analysis.

A core limitation is our reliance on global evidence and aggregate data, which may not fully capture local context or implementation realities. We often depend on information provided by the nonprofits themselves or by partners in our network, which can introduce bias and reduce visibility into local dynamics. This creates a risk of false precision, where quantitative models obscure important qualitative factors such as cultural nuances, community preferences, or political economy considerations. As a result, we may underestimate unintended consequences such as dependency, exclusion of marginalized groups, or fragile implementation that cannot withstand shocks.

We also face practical constraints. Our work depends heavily on existing networks of experts and evaluators, which tends to direct our attention toward organizations already well known in philanthropic and evaluation circles. Language barriers and limited team capacity further narrow the scope of organizations we can identify and review. These constraints mean that we are unlikely to capture the full diversity of innovative or smaller-scale actors working in low-visibility contexts.

Finally, our framework does not yet include a systematic way of assessing all risks that may affect the success of interventions and organizations. Factors such as political instability, regulatory changes, or sudden shifts in local systems can have major consequences, yet are difficult for us to predict or incorporate consistently. While we aim to capture the best available evidence, we acknowledge that our evaluations cannot fully account for every contextual nuance or long-term uncertainty.

References

Alkire, S., Kanagaratnam, U., & Suppa, N. (2020). The global Multidimensional Poverty Index (MPI): 2020 revision. OPHI MPI Methodological Note 49. Oxford Poverty and Human Development Initiative, University of Oxford.

Angrist, N., Evans, D. K., Filmer, D., Glennerster, R., Rogers, F. H., & Sabarwal, S. (2025). How to improve education outcomes most efficiently? A review of the evidence using a unified metric. Journal of Development Economics, 172(7), Article 103382. https://doi.org/10.1016/j.jdeveco.2024.103382

Barder, O. (2014). Is aid a waste of money? Significance, 11(1), 31–34. https://doi.org/10.1111/j.1740-9713.2014.00723.x

Deaton, A. S., & Cartwright, N. (2017). Understanding and misunderstanding randomized controlled trials (NBER Working Paper No. 22595). National Bureau of Economic Research. http://www.nber.org/papers/w22595

Easterly, W. (2006). Planners versus Searchers in Foreign Aid. Asian Development Review, 23(2), 1-35. https://doi.org/10.1142/S0116110506500053

GRADE Working Group. (n.d.). GRADE: From evidence to recommendations – transparent and sensible. https://www.gradeworkinggroup.org/

Kaffenberger, M., & Pritchett, L. (2021). Effective investment in women’s futures: Schooling with learning. International Journal of Educational Development, 86, 102464. https://doi.org/10.1016/j.ijedudev.2021.102464

Kwa, T. (2022, May 4). Most problems don’t differ dramatically in tractability (under certain assumptions). LessWrong. https://www.lesswrong.com/posts/WyJX7Rz4rq8nu3kCb/most-problems-don-t-differ-dramatically-in-tractability

Moyo, D. (2009). Dead Aid: Why Aid Is Not Working and How There Is a Better Way for Africa. Farrar, Straus and Giroux.

OPHI (Oxford Poverty and Human Development Initiative). (2024). Multidimensional poverty and the Alkire-Foster method. University of Oxford. https://ophi.org.uk/md-poverty-and-AF-method

OPHI (Oxford Poverty and Human Development Initiative) and UNDP (United Nations Development Programme) (2024). Global Multidimensional Poverty Index 2024: Poverty amid conflict. United Nations Development Programme and Oxford Poverty and Human Development Initiative. https://ophi.org.uk/Publications/GMPI15-2024 

Pawson, R., & Tilley, N. (1997). An introduction to scientific realist evaluation. In E. Chelimsky & W. R. Shadish (Eds.), Evaluation for the 21st century: A handbook (pp. 405–418). Sage Publications, Inc. https://doi.org/10.4135/9781483348896.n29

Schünemann, H., Brożek, J., Guyatt, G., & Oxman, A. (Eds.). (2013). GRADE handbook for grading quality of evidence and strength of recommendations. GRADE Working Group. https://gdt.gradepro.org/app/handbook/handbook.html

Seelos, C., & Mair, J. (2017). Innovation and scaling for impact: How effective social enterprises do it. Stanford Social Innovation Review. https://doi.org/10.48558/TCD0-D753

Suppa, N., Alkire, S., & Nogales, R. (2022). The many forms of poverty: Analyses of deprivation interlinkages in the developing world (OPHI Research in Progress No. 63a). Oxford Poverty and Human Development Initiative. https://ophi.org.uk/sites/default/files/2024-05/OPHIRP63a.pdf

UNDP, & OPHI. (2022). 2022 Global Multidimensional Poverty Index (MPI): Unpacking deprivation bundles to reduce multidimensional poverty. United Nations Development Programme. https://hdr.undp.org/content/2022-global-multidimensional-poverty-index-mpi

Vivalt, E. (2020). How much can we generalize from impact evaluations? Journal of the European Economic Association, 18(6), 3045–3089.