Applications are currently open for the next cohort of AIM's Charity Entrepreneurship Incubation Program in August 2025. We've just published our in-depth research reports on the new ideas for charities we're recommending for people to launch through the program. This article provides an introduction to each idea, and a link to the full report. You can learn more about these ideas in our upcoming Q&A with Morgan Fairless, AIM's Director of Research, on February 26th.

Advocacy for used lead-acid battery recycling legislation

Full report: https://www.charityentrepreneurship.com/reports/lead-battery-recycling-advocacy 

Description

Lead-acid batteries are widely used across industries, particularly in the automotive sector. While recycling these batteries is essential because the lead inside them can be recovered and reused, it is also a major source of lead exposure—a significant environmental health hazard. Lead exposure can cause severe cardiovascular and cognitive development issues, among other health problems.  

The risk is especially high when used-lead acid batteries (ULABs) are processed at informal sites with inadequate health and environmental protections. At these sites, lead from the batteries is often released into the air, soil, and water, exposing nearby populations through inhalation and ingestion. Though data remain scarce, we estimate that ULAB recycling accounts for 5–30% of total global lead exposure.

This report explores the potential of launching a new charity focused on advocating for stronger ULAB recycling policies in low- and middle-income countries (LMICs). The primary goal of these policies would be to transition the sector from informal, high-pollution recycling to formal, regulated recycling. Policies may also improve environmental and safety standards within the formal sector to further reduce pollution and exposure risks.

Counterfactual impact

Cost-effectiveness analysis: We estimate that this charity could generate about 59 income doublings^[1] per $1,000 (USD). However, this model carries significantly more uncertainty than our typical estimates due to limited evidence. Our model assumes that the charity would operate in three countries in parallel, with a 20% chance of success per country. If successful, we estimate that it could reduce blood lead levels by 5% in the target country.

Scale this charity could reach: We estimate that, if this charity’s work is successful, it could generate around 14,000 income doublings per year at scale per country of the size of Thailand.

Potential for success

Robustness of evidence: This intervention does not have a very strong evidence base, as it largely relies on case studies. Most evidence comes from high-income countries, though around seven LMICs have also made good progress. The most relevant case is Brazil, where implementing the three policies listed above was associated with a shift away from informal recycling, reducing its share from 45% to 10%. Some of the progress made in these countries has been industry-led (like in Brazil) rather than driven by NGOs, indicating potential industry support and opportunities for collaboration.

Additionally, the experience of existing nonprofits working on lead exposure mitigation suggests strong government interest. However, we anticipate that progress on ULAB recycling will be less tractable than for other lead exposure sources due to the complexity of the lead-acid battery market.

Theory of change (ToC): The primary ToC focuses on country-level policy advocacy, supported by local data (quantitative and qualitative) collection on the ULAB-recycling market, along with technical assistance to governments on regulatory and enforcement issues. Initially, we propose exploring three main policies: a tax exemption for ULABs to reduce the price difference between formal and informal recycling; extended producer responsibility (EPR), which requires manufacturers or importers to ensure that ULABs are recycled in the formal sector; and the establishment of a producer responsibility organization (PRO) to oversee enforcement of these policies on behalf of the government. However, other policies may also be promising, and solutions will likely require significant country-specific tailoring.

A secondary ToC envisions accelerating global progress on this issue by generating new evidence and advising governments and NGOs on the best practices. Currently, almost no one is filling this role, and experts have emphasized that a team focused specifically on ULAB recycling would be highly valuable.

Neglectedness

Neglectedness: We are confident that there is room for a new organization in this space. While major actors working on lead exposure—such as Pure Earth, UNICEF, and UNEP—are addressing ULABs, their efforts appear to be ad hoc rather than a core focus. However, there is some chance this could change in the short to medium funding from the Lead Exposure Action Fund increases.

Geographic assessment: Our analysis prioritizes countries with high lead exposure, a large informal employment sector, a sizeable population, and relatively low fragility and corruption. Based on these criteria, we identify Indonesia, India, Pakistan, Bangladesh, Mexico, Peru, Thailand, Colombia, Argentina, and Egypt, among others, as promising locations for this intervention. In general, we expect the greatest potential in middle-income countries that already have at least one high-quality formal ULAB recycling facility.

Additional Considerations

Expert views: We spoke with 13 experts, who we believe represent the majority of those working in this field. Most were enthusiastic about the prospect of a new organization, emphasizing that increasing engagement in this space could generate valuable new insights. The most common concern was that Brazil’s approach may not be directly applicable to other countries, as policy solutions would need to be tailored to local contexts.

Implementation factors: Our main concerns relate to access to information and execution challenges. There is a significant lack of data in this space, which will likely slow progress. Even basic details—such as the scale of lead poisoning from ULAB recycling and the locations of recyclers—are not readily available. Combined with the economic interests of the informal ULAB recycling industry, these factors make us believe that failure is more likely than success, with the probability of successfully reducing lead exposure estimated at around 10–25% per country. However, even if the charity does not achieve policy change, it would still generate highly valuable information for other actors in this space.

Scaling up Kangaroo Care

Full report: https://www.charityentrepreneurship.com/reports/kangaroo-care-scaleup 

AIM has incubated one organization providing Kangaroo Care (KC) working in India, Ansh (launched in early 2023; see original report here). KC is a package of interventions which involves skin-to-skin contact between a low birth weight baby and a caregiver, exclusive breastfeeding, and close monitoring of the baby and their relevant vital and danger signs. KC is a cheap and easy-to-administer form of neonatal care that is evidenced to be effective at reducing mortality and morbidity.

Cost-effectiveness analysis: We updated our 2022 cost-effectiveness analysis using our new template, incorporating data from Ansh’s pilot^[2] and revised effect sizes for KC’s impact on mortality and morbidity based on our updated evidence review. Embedding KC capacity in hospitals in Pakistan is expected to be highly cost-effective, surpassing the cost-effectiveness anticipated by our previous model. The updated model estimates that the intervention could avert one DALY for every $50 spent.

Potential for success

Robustness of evidence: Overall, we are encouraged by the early-stage success of Ansh in building strong relationships with key stakeholders, which we feel is necessary for success in this space. Despite this update to feasibility, Ansh’s pilot shows that not all parents reach 8 hours of daily skin-to-skin contact. This may decrease the effectiveness of KC, as our understanding of its effects is largely based on evidence of 8 or more hours per day. Considering these factors and new evidence since publication, we have slightly reduced our estimate of KC’s mortality effects, but slightly increased our estimate of its morbidity effects.  

Theory of Change: We still think embedding capacity into the health system is a strong ToC that addresses many barriers to scaling up KC. We also weakly endorse exploring context-specific alternative ToCs, in particular, KC for emergency transport and initiating KC outside hospital settings.

Neglectedness

Geographic assessment: KC still remains highly neglected. We updated our assessment using a new template and more recent data. The recommended target countries remain largely the same, with most shifting positions within the top 10.

Additional Considerations

Expert views: We spoke with Ansh, GiveWell, and another KC implementation org whose insights mostly gave us positive updates, particularly regarding the value of starting an organization sooner rather than later.

Implementation factors: It would benefit the founding team to have a healthcare/medical background and familiarity with the target country, either through being a local or having previous work experience. We think a local founder would be preferable to make stakeholder relationships and access easier.

Differentiated Learning

Full report: https://www.charityentrepreneurship.com/reports/differentiated-learning 

Differentiated Learning (DL) is an educational approach that groups children by learning level, rather than their age, providing targeted instruction to improve foundational literacy and numeracy skills. This approach ensures that students who have fallen behind receive tailored instruction, allowing them to catch up on foundational skills. DL has been implemented at scale in multiple countries and evaluated by several randomized controlled trials (RCTs), making it one of the most promising interventions for improving the quality of education in low- and middle-income countries (LMICs).

Counterfactual impact

Cost-effectiveness analysis: We estimate the cost-effectiveness of this intervention to be between $12 and $45 (USD) per person doubling their future income. This estimate is based on projected impacts in the Philippines, the most promising country in our geographic assessment. The extent of this effect depends on the assumed relationship between education improvement and income gains.^[3] This meets our $30 benchmark in most cases. Additionally, our model indicates that dozens of other countries also meet this cost-effectiveness bar.

Scale potential: A successful charity could operate at a large scale. If expanded to cover 10% of the Philippines, we estimate it could generate 1.6–6.2 million consumption doublings per year.

Potential for success

Robustness of evidence: We are highly confident in the effectiveness of this intervention. DL is one of the most studied education interventions in developing countries, with numerous RCTs demonstrating effectiveness.

Theory of Change: The primary Theory of Change (ToC) is to deliver DL—likely in collaboration with the NGO Pratham—in a country where no organization is currently delivering it or where only small local actors are active in limited regions.

There are various delivery models, including:

• Training teachers to implement DL in their classrooms.
• Running short-term learning “camps” led by staff or volunteers.

It is unclear to us which model would be most effective, and the best approach may vary by country or region.

An alternative ToC would involve partnering with existing NGOs that already deliver DL, providing them with training and monitoring and evaluation (M&E) to help them scale up and improve delivery quality. This approach could allow the charity to operate even in countries where DL is already present while still having a cost-effective impact.

Additional Considerations

Expert views: Experts from Pratham, TaRL Africa, J-PAL, and the World Bank were generally supportive of us working on DL. Some raised concerns about whether we could deliver DL to a high standard, though we suspect these concerns stem from past experiences with low-quality or uncooperative implementers—a challenge unlikely to apply to an AIM charity. Overall, key stakeholders in this space appear highly collaborative and were generally enthusiastic about the prospect of new implementing partners.

Implementation factors: DL is a complex intervention and evidence suggests that both uptake and fidelity of delivery strongly influence its effectiveness. While we expect the charity would be able to implement DL successfully at a small scale, we have some concerns about its ability to maintain high-quality delivery at a larger, sustainable scale.

Would this charity be additional?

While DL is not being implemented at scale in many countries, major organizations are already planning expansions, raising concerns that a new charity might provide limited additional value. However, existing actors are capacity-constrained, with more funding available for DL than they can absorb. Additionally, leading organizations such as Pratham and TaRL Africa typically operate via local implementing partners, and an AIM charity could serve in this role. Finally, most experts we consulted supported the entry of new high-quality implementers.

Which exact ToC should this charity pursue?

While both a volunteer-led model (more effective but less scalable) and a teacher-led model (more scalable but less effective) are viable approaches, it is also unclear whether the charity should deliver the intervention directly or support existing NGOs in scaling up and improving delivery.

The most suitable approach may depend on the country and context, and we do not yet have strong evidence to recommend one over the other. Potential founders of this charity should conduct follow-up research and consult experts—particularly those with prior experience implementing DL and local expertise in the target country—to determine the best model.

How does DL compare with other education interventions, such as structured pedagogy?

AIM previously recommended incubating a charity to deliver Structured Pedagogy (SP), another highly studied and cost-effective education intervention, which led to the creation of the Learning Alliance. This raises the question of how DL and SP compare and whether they could be synergistic. Our models suggest that both interventions have overlapping ranges of cost-effectiveness, though DL may be less cost-effective due to being more labor-intensive.

While SP focuses on improving teaching quality in standard classrooms, DL provides remedial education for students who have fallen behind. Given these distinct aims, the two approaches may be complementary, with a recent study in Morocco showing promising results from a combined approach.^[4] In the future, this charity could explore co-delivering SP as a potential expansion.

Mass Communication for Education

Full report: https://www.charityentrepreneurship.com/reports/edu-mass-comm 

A nonprofit organization that leverages mass-communication strategies (mobile messaging) to provide caregivers and students with information on various aspects of education—including the returns to education, student effort, and institution quality—as a low-cost way to improve both attendance and learning outcomes.

Counterfactual impact

Cost-effectiveness analysis: We modeled an SMS-based intervention targeting grades five and six in South Africa and estimated the number of students it would need to reach to meet AIM’s cost-effectiveness bar of $30 (USD) per income doubling^[1]. In our conservative model (assuming a 10% income increase from a one standard deviation (SD) improvement in test scores), the intervention would need to reach approximately 58,000 students per year at scale. In an optimistic model (assuming a 40% income increase from a one standard deviation improvement in test scores), it would need to reach approximately 8,700 students per year at scale. For context, this represents roughly 2.0% (conservative) or 0.3% (optimistic) of South Africa’s 10–12-year-old students.

Scale this charity could reach: Because this intervention primarily relies on SMS or other mass messaging, which are inexpensive per individual reached, we believe its potential scale is comparable to other nonprofits we have incubated, such as Family Empowerment Media (FEM).

Potential for success

Robustness of evidence: This intervention design has been evaluated many times with generally positive findings—however, it is less studied as a mobile intervention and has not been evaluated as a radio intervention. It is endorsed by the Global Education Evidence Advisory Panel’s 2023 Cost-effective Approaches to Improve Global Learning Report (GEEAP, 2023). More recently, researchers from the Centre for Global Development published a meta-analysis on the effects of providing information on the returns to education to parents and students, finding positive and significant average impacts on school participation and student learning (Evans & Acosta, 2024).

Theory of Change (ToC): The ToC behind this intervention is that providing persuasive and informative messaging on education can lead to increased caregiver and student engagement, resulting in higher attendance and improved learning outcomes. The specific design of a new nonprofit’s intervention will depend on contextual factors such as the suitability of different delivery mechanisms, and the availability of key information, such as test scores and data on school quality.

Neglectedness

Neglectedness: We identified one organization that works in this area as a for-profit focused on tertiary education in Brazil. Our understanding of the field leads us to believe that this intervention is popular but lacks implementers delivering at scale, mostly remaining as an evaluative pilot or delivered sporadically through government programs.

Geographic assessment: Our geographic assessment suggests that several countries would be strong candidates for an intervention. We are confident that a new nonprofit could identify countries where its work would be additional.

Additional Considerations

Fit for the CEIP: Our best guess is that this idea fits the typical CEIP participant profile since it does not require specialized expertise, has historically been attractive to participants, and is well-supported by the international education and development community. However, the founding team must be comfortable with an intervention that depends on large-scale implementation for cost-effectiveness and has weak feedback loops.

Expert views: We spoke with Guilherme Lichand, who founded a nonprofit that initially used SMS delivery before pivoting to work in tertiary education. He was generally positive about SMS interventions as part of a wider package of support to education. Other experts we briefly consulted were also generally supportive but expressed reservations about radio delivery.

Implementation factors: Overall, we see this intervention as similar to others being tested or delivered by past cohorts, including FEM, No Violence at Home, Learning Alliance, Notify Health, and Suvita. We did not identify any major concerns, such as risks of harm or lack of funding. However, our main concern is the challenge of measuring impact. Given that the expected effect on student test scores is likely to be small, any impact evaluation would require a large sample size to detect significant effects. Monitoring intermediate outcomes, such as attendance and drop-out rates, may be more feasible.

Establishing a Livelihoods Evaluator

Full report: https://www.charityentrepreneurship.com/reports/livelihood-charity-evaluator 

Incubating a new evaluator which would approach evaluation from a livelihoods perspective (‘improving lives, not averting deaths’, i.e., focusing on incomes, education, etc.). The evaluator is expected to be rigorous and produce public recommendations, like GiveWell.

Counterfactual impact

Cost-effectiveness analysis: We conducted a fairly conservative back-of-the-envelope calculation (56% chance of success; 70% discount on the average money moved per year by other evaluators in their first five years; average donation around 5x GiveDirectly).

Scale this charity could reach: We are highly uncertain where a reasonable estimate for a peak in funds moved for a successful evaluator could be. Nevertheless, we expect any ceiling to be well in excess of an organization's operating expenses, and as such, do not expect this to reduce an evaluator’s cost-effectiveness below our target bar.

Potential for success

Robustness of evidence: Non-profit evaluators can be categorized as “giving multipliers,” organizations that aim to increase the amount of money going to non-profits. To that end, we note several anecdotal cases of evaluators who direct between 6 and 31 times their operating budget to non-profit organizations. We think their success is a marginal positive update that other actors can also achieve. Still, we suspect that early movers in the space benefitted from the relative dearth of competition.

We also note with caution that the literature examining the psychology of effective giving is mixed and tends to find that information on effectiveness does not motivate donors to give more. We think this should continue to be evaluated and used to tailor strategies for converting donors.

Theory of Change: The theory of change for this intervention is relatively straightforward and reflects the approach of established players like GiveWell, Giving Green, and Animal Charity Evaluators. The organization will produce and market rigorous evaluations designed to appeal to a wide audience. These credible and persuasive analyses are expected to enhance donors’ understanding of what constitutes a high-impact non-profit and increase their motivation to direct funds towards effective organizations. Ultimately, this should result in more funding for impactful charities and improved well-being for their beneficiaries.

The greatest uncertainties in this ToC relate to the evaluator's ability to gain recognition and transform its research into donations. While existing organizations have demonstrated success in these areas, smaller actors often struggle to establish visibility and communicate the value of their work. Additionally, producing rigorous evaluations requires a high level of research expertise, as evidenced by the challenges and errors documented by evaluators in the past.

Additional Considerations

Is there a gap in the market?

We think that many gaps in the market could draw attention from prospective funders: there are only a handful of evaluators, and arguably only one or two very large and popular ones. Increasing the number of evaluators in the field will likely improve viewpoint diversity and plurality of approaches.

Livelihoods emerged as a strong option among the handful we considered. We think existing evaluators in this space give much greater moral weight to health-focused interventions—especially life-saving ones—over income-focused interventions. Many experts were excited by the idea of increased evaluation in this space. However, some felt the space was already well covered by existing organizations.

Will there be sufficient donor interest?

We are unsure but marginally confident that there will be sufficient early interest in the outputs of an evaluator's work. We have already received some informal interest from two or three foundations and/or individuals who have suggested they would be interested in either funding or consuming the research produced to inform their donations. It should be noted that given more time, we would have scoped the donor market more for this type of moral view and interest. However, we note the existence of several potentially aligned funders, such as the Livelihood Impact Fund, the Agency Fund (and potentially those listed on their site as partners), and other large philanthropic donors.

Will the new organization be successful in gaining an audience?

By audience, we mean attention from the broader public. We think gaining traction within aligned audiences, such as effective altruism-aligned individuals and organizations, would account for earlier and more tractable success for the organization. However, we have a strong expectation that the organization seeks to broaden the types of funders and individuals it appeals to. GiveWell and Giving Green have succeeded in becoming “household brands” in the charity evaluation sector which gives us some confidence. However, other evaluators have remained relatively unknown and have adopted approaches to reaching out to high-net-worth individuals or donors.

Expert views: We carried out several informal consultations with individuals who have expertise in the field of non-profit evaluation and affiliated efforts, including a survey consultation (n = 11). Livelihoods was marginally the most favored option among the three we considered. Yet, there was considerable disagreement about potential funding sources, talent availability, and other implementation factors.

Implementation factors: We are most concerned about talent and tractability—we think that to achieve early success, an evaluator will need an early-stage team with strong research skills and resilience. We view non-profit evaluation as a highly technical skill, and expect that marketing those evaluations will also take considerable effort to get right.

Excited about these ideas?

We’d love to see you apply to found a new organisation based on one of these ideas through the Charity Entrepreneurship Incubation Program. 

There are no specific qualifications or previous expertise that we require - what we care about is finding talented, ambitious, mission-driven individuals. Our training and mentorship will give you the specific skills you need.

Applications are open until March 24th. To answer questions you may have, we’ve organised a series of online Q&A sessions with:

• Our Director of Research, Morgan Fairless, on February 26th to discuss these ideas
• Our Director of Recruitment, Ben Williamson, on March 6th to discuss our application process and questions about the program

1. ^{^}

    An income doubling refers to an increase in a person’s lifetime earnings equivalent to twice what they would have earned in a single year without the intervention, based on projected career earnings and adjusted for present value.

2. ^{^}

    Our CEA relies on pilot data from work in two hospitals in Rajasthan from January to March 2024, but Ansh is now in their scale-up phase so specific numbers used in this model could already be out of date.

3. ^{^}

    At AIM, we model the relationship between one standard-deviation improvement in test scores (a meaningful academic gain relative to peers) and income to be between 10% and 40%, based on our review of the literature.

^{^}

 Ibrahim et al. (2024)

Show all footnotes