Clean water. Most of us take it for granted. We can turn on the tap and have safe water to drink whenever we want, without having to give it a second thought. That’s not the case for more than a billion people around the world who lack access to uncontaminated drinking water. This is a significant health concern: According to the World Health Organization, more than 500,000 people die each year from diseases caused by contaminated water.

We’ve been thinking hard about how best to build on our past grantmaking to address the huge need for clean water. In January, we launched a public request for information (RFI) to identify organizations who would be interested in implementing chlorination programs in countries with a high burden of waterborne disease.

RFIs are a relatively new strategy for GiveWell aimed at identifying highly cost-effective grantmaking opportunities. The call for water chlorination programs was just the second RFI we issued. The first, completed in 2024, was for research grants, and we’ve since completed another, for programs aimed at increasing vaccination rates. We’ve been excited about the overwhelming interest our RFIs have received. By the March 7 application deadline for water chlorination programs, we had received more than 200 applications.

From that group, we recently funded a portfolio of grants totaling around $15.5 million to support pilots of in-line chlorination programs in six African countries. Adding chlorine to untreated water is an inexpensive method of making it significantly safer to drink,¹Chlorination reduces the concentration of waterborne bacteria and viruses, thus protecting against diarrheal illness, though it does not deactivate all diarrhea-causing pathogens. For example, it has limited effectiveness against the protozoan parasite Cryptosporidium, one common cause of diarrhea in children in low-income settings. For more on the effect of chlorination, see this section of our report on water quality interventions. and we’re excited about the good these grants might accomplish—and what we can learn from them.

• First, the pilots will provide thousands of vulnerable people with access to safe drinking water. We estimate that the in-line chlorination devices installed as part of the pilots will provide the equivalent of one year of safe water for more than 2.8 million people and avert the deaths of more than 2,000 people, most of them children under five years old.
• Second, because we funded a wide range of implementing organizations working in different contexts and with different program models, we will have the opportunity to learn a lot very quickly that we can then apply to our future grantmaking decisions.

After the pilots are complete, we’ll consider funding to scale up the programs for the successful pilots. We estimate that there might be more than $100 million in highly cost-effective funding opportunities in the countries where the pilots are being conducted, potentially offering an opportunity to reach millions of additional people with safe and clean drinking water.

We’ll also be able to use what we learned about the choice of chlorination devices, the feasibility of implementation in various settings, the organizations that are most likely to be successful, and much more as we evaluate future grants. We’re excited about the potential of these pilot grants—and about this approach to grantmaking in general—to lead to even more opportunities for our donors to save and improve lives.

GiveWell’s past grantmaking

Programs to improve water quality via chlorination have been on GiveWell’s radar for many years—we made our first grant to support water treatment in 2017, and since then we’ve provided more than $125 million to improve water quality in low- and middle-income countries, most of it following a major update in our assessment of water quality interventions in 2022.

The vast majority of that funding has gone to two programs implemented by Evidence Action: chlorine dispensers and in-line chlorination. We funded Evidence Action to install chlorine dispensers in Kenya, Malawi, and Uganda and in-line chlorinators in Malawi, and to work with state governments to provide in-line chlorination in India.

The Dispensers for Safe Water program sets up small tanks containing liquid chlorine near communal water collection points in rural areas. People using the water points then dispense chlorine into their water containers so that it disinfects the water they collect. In-line chlorination, on the other hand, automatically disinfects water at collection points by adding chlorine as it flows through a pipe. The device requires no action on the part of the user, making it more convenient than programs that require users to add chlorine themselves.

A new approach

We estimate that chlorination programs can be highly cost-effective in many locations because chlorine is inexpensive and our research suggests it can substantially decrease child mortality. As a result, we have been looking for other opportunities to fund these programs. In June 2024, we wrote about our current plans for our water grantmaking portfolio. As we mentioned then, two of the highest-priority goals for our research subteam focused on water quality are to expand the number of implementers of large-scale chlorination programs and to learn more about the relative merits and drawbacks of various approaches to improving water quality. Our RFI was designed to help us accomplish these goals.

We reviewed all the proposals—more than 200—ourselves, then had chlorination experts review a subset of the proposals again. We scheduled multiple calls with each organization we were considering, and we talked with experts and other organizations to understand the local context and how each pilot would fit into other efforts taking place in the same area. At the end of that process, we approved 12 grants to seven implementing partners to pilot in-line chlorination programs in six countries in Africa—creating a portfolio that will allow us to learn quickly about a wide array of contexts and program models.

*The coverage estimates above are based on GiveWell’s calculations and include adjustments based on the possibility that some people may use alternative water sources or may not be successfully reached. The map was created using MapChart.

Building a portfolio to maximize learning

As we evaluated the projects, we considered a range of factors, and we intentionally selected a portfolio of programs with differing strengths and opportunities. Together, the pool of grants will teach us more, and more quickly, than we would learn by making the grants one at a time over a period of years. Among the factors we considered are the following:

Cost-effectiveness. As with the vast majority of our grants, we modeled the cost-effectiveness of the programs. We created separate estimates for the pilot period and for longer-term scale up of the project. Some programs showed the potential for high cost-effectiveness during the pilot period; others have lower initial cost-effectiveness, but we estimate that they’ll reach our funding threshold as they scale up. All of the partners expressed interest in continuing the projects, or potentially expanding them, following the pilots.

Direct impact. Our estimates of direct impact incorporated mortality rates in the regions where the pilots would be implemented, along with the proportion of those deaths attributable to intestinal infections, which are often caused by contaminated water. We estimated more specific mortality rates for subnational regions, when available, and tried to adjust for areas where mortality rates might differ from the country as a whole, such as in places with a high population of internally displaced people.

Organizational experience. Some of the implementing organizations are local NGOs with extensive experience in the area but limited experience with in-line chlorinators—we think that partnering directly with NGOs headquartered in Africa could potentially be more effective and cost-effective than working with organizations headquartered in the United States or in Europe, and we expect to learn more about this during the grant period. Others are international organizations with extensive experience constructing communal water points, including those being used in the pilots. Another is a company that manufactures chlorination generation devices in partnership with a company that manufactures chlorine dosing devices.

Chlorination devices. The pilots, which will be 12 to 18 months long, range from the installation of 10 devices to more than 500. The chlorination devices themselves will differ based on what is most appropriate for a given context. For example, some of the pilots plan to generate chlorine solution on site, while others will procure it from suppliers. Some devices will use liquid chlorine and others will use chlorine tablets. This diversity of devices will help us learn more about which devices work best in which settings.

Local context. Some of the pilots are being conducted in rural areas, others in or near urban settings. Many of the pilots will provide chlorinated water to refugee camps or to camps for internally displaced persons. Some pilots will include installation of devices on private boreholes.²A borehole is similar to a well, but is typically deeper and narrower than a well, reducing the chance for surface contamination. By diversifying the contexts in which the programs are implemented, we expect to learn more about the feasibility of in-line chlorination in a wide variety of settings.

Implementation model. Some program models are similar to ones we’ve funded in the past, while others include innovative components that we’d like to learn more about. For example, one of the pilots will use both imported chlorination devices as well as devices that are fabricated at a local vocational school, allowing us to compare the quality and cost of each. Another pilot will use onsite chlorine generators, which may reduce the logistical challenges of supplying chlorine to remote locations.

Addressing our uncertainties

The grants will provide us an opportunity to learn about water infrastructure and the chlorine supply chain in new areas where we haven’t previously funded water projects. In addition, we are interested in addressing several of our uncertainties:

• Where are in-line chlorination programs feasible, and what steps can be taken to increase feasibility? We’ll examine how effectively the organizations can manage the chlorination systems and keep the devices stocked with chlorine.
• What chlorination rates are achievable in each context? We’ll measure chlorination rates in households and at water collection points before the pilots begin, during the pilots, and afterward.
• How much do in-line chlorination programs reduce water contamination? We’ll measure contamination rates in households and at water collection points before the pilots begin, during the pilots, and afterward.
• How many people use the water points where the devices are installed? We know it’s difficult to collect data for water collection points with a large number of users, and we’re considering an array of strategies for estimating usage, such as using customer lists, GIS data, or surveys of households near randomly selected GPS coordinates.

Scaffolding for success

While we have conducted a thorough analysis of all these grants, it is still likely that some of the pilots will not be successful. We have heard—and have experienced via our other grants—that in-line chlorination is challenging to implement, primarily because water infrastructure differs significantly from one geographic context to another, requiring adaptations to the chlorination devices. We are therefore taking proactive steps to improve the likelihood of success—and the opportunities for learning.

All of the organizations will receive support from our technical assistance partner, Mangrove Water, which provides advice and support for the implementation of in-line chlorination programs.³We’ve previously funded Mangrove Water to develop an implementation guide for organizations considering in-line chlorination programs and to create an adapter so in-line chlorination can be used with hand pumps. In addition, each partner will work with an external, independent survey firm to collect baseline data before implementation begins and then again after the pilot has been completed. In this way, we think even the expected failure of some pilots will provide critical data that will improve our future grantmaking in an area with such great need and opportunity to make a difference.