What is this?

Who is behind this project?

Don't big foundations know what the good charities are? Why not just use their information?

Why does GiveWell give its own grants, instead of just publishing information?

Where does GiveWell get the money to make its grants?

How can GiveWell simultaneously evaluate charities and raise money for its own grantmaking? Does that create a conflict of interest?

GiveWell operates as a public foundation, pooling donations in order to give grants. This is a common model - it describes United Way among other organizations - the difference being that GiveWell publicly discloses its reasoning, so that even those who don't donate to GiveWell can still take advantage of our research.

We hope that those who respect the quality of our decision-making and research, and wish to help us do more of it, will donate directly to us; but we also want to help donors who would rather give directly to excellent charities without making any commitment to GiveWell. This decision is not relevant to the content of our writeups, which aim to compare our applicants to each other rather than to GiveWell. We therefore believe that our writeups do not suffer from any conflict of interest; if you believe that there are cases in which our writeups are unfair, we'd appreciate your bringing them to our attention.

GiveWell plans to raise funds both from donors who wish to support our mission in general, and donors who wish to "vote with their donations" on which causes we research next. We will cover any and all causes for which we have adequate demand, funding, and staff; there is no conceptual limit to how large we can grow or how many causes we can cover, although the challenge of finding and managing great staff will limit our rate of growth to some extent.

• There is a GiveWell registered in Massachusetts that appears to be defunct. It has not filed its required Form 990 in several years; its website (listed as givewell.org) returns a 403 error; and we have been unable to reach the organization by email or phone.
• There is a GiveWell in Australia that performs a similar function to the "watchdog" organizations listed here, providing information taken from public records. Like other watchdogs, it differs from us in its emphasis on breadth (providing basic, publicly available information on many charities) over depth (using a grant application process to examine what charities do and whether it works); it also differs from us in that it covers Australia-registered charities rather than US-registered charities.
• GiveWell.com is currently the home of the Healthcare Gift Card offered by Highmark, a product with no connection to charitable giving.

Getting involved

See our donation page.

See our donation page.

We are always looking for hard-working, passionate, intelligent people who can help us research the best methods of helping people, as well as contribute to our project in other ways. We hire based on ability and passion rather than experience and credentials.

At this point we generally require any job candidate to volunteer for us, for at least some period of time, before we offer compensation. The reasons for this are:

• Working directly with people is the single best way we can determine whether they have the passion, commitment and ability to contribute substantially.
• Because we are a charity, our candidates generally find this requirement reasonable and do not mind contributing to us without pay in the early going. (We do not require volunteers to be full-time; we simply ask that they prove their abilities and contribute to our project before being hired.)

Therefore, if you would like to pursue a job with GiveWell, we encourage you to submit our supporter form and mark that you are interested in volunteering. If you have particular interests and abilities, please indicate them in the comment field. We will be in touch with volunteer opportunities, which may lead to a hire.

You can help GiveWell by donating, volunteering, helping to spread the word, and sharing your feedback.

For the first three, submit our supporter form and be specific about how you’re interested in helping out; we will be in touch when appropriate. To share your feedback, simply use our forums or contact us directly.

GiveWell: strengths, weaknesses, and self-evaluation

We believe that our primary qualification is our willingness to share and discuss everything that goes into our decisions.

The question of where to give is a difficult one; we believe that it cannot be contained within a single academic field or intellectual approach, and we’ve also observed that there is no apparent "expert" consensus on it (even if it could be determined what qualifies one as an "expert" in charity evaluation). Charity evaluation inevitably involves analysis, intuition and judgment calls, and even philosophical decisions. Ultimately, there is no such thing as a giving decision that is "foolproof" and can’t be legitimately questioned and criticized. We believe this is why other grantmakers are hesitant to share their reasoning publicly - and exactly why it is so essential to do so.

We believe ourselves to have a very strong combination of analytical intelligence, common sense, and passion for the issues we study, which positions us well to make reasonable bets. Those who have worked with us in the context of highly selective and competitive environments will attest to this. We also consult with experts whenever relevant, and for each of the causes we have studied, we have spent significant time discussing our decisions with at least one person who is extremely experienced in the field. However, we take the position that transparency - the willingness to share all the details of our decisions in public - is more valuable by far than any of these qualities, or than any quality of a grantmaker. In other words, your single best reason for trusting us is that you don’t have to.

As our story explains, GiveWell would not exist if we’d been able to find a website that could help us substantially with our donation decisions. We have compiled a list of other donor resources we know of, available here.

We make no such assumption. GiveWell’s staffers are motivated 100% by emotion, and we believe that failing to do rigorous analysis is usually a result of not caring enough, rather than being over-emotional (as we’ve argued informally on our blog).

We do recognize that many donors may not share our values and worldview, and like any venture, we do not aim to serve everyone. Our research is specifically intended to help those who donate primarily in order to improve the world, and for whom accomplishing as much good as possible is more important than the other perks of donating.

Thoroughly evaluating even a single charity is an enormous amount of work, and we want to use our limited time and resources optimally. With that in mind, we believe that identifying the best charities is far more important than identifying the worst.

In the business world, regulating/exposing/preventing gross misdeeds is important; but the majority of investment decisions have far less to do with these issues than with difficult and intuition-laden predictions of which companies will perform best. Relentlessly funding the best performers leads to constant improvement (for example, cell phones continually get better because people reward quality, not because they are vigilant about cell phone scandals). We believe it is this dynamic - demanding and rewarding the best - that is currently absent from charity, and that is far more needed than more coverage of scandals.

Other organizations do report on charity scandals, likely better than we could. The American Institute of Philanthropy is one such organization.

The value of GiveWell is a product of the following:

1. How much money we influence, directly or indirectly, with our research.
2. How much value our research adds (i.e., how much more effective a donation is when informed by our research).
3. The extent to which we’re able to create higher-quality dialogue around giving, and spread the acceptance and use of our core values (particularly transparency in giving decisions).

(1) is relatively straightforward to measure. When people make donations to our recommended charities using the links on our site, we are able to track these donations; we also, of course, track how much is given directly to GiveWell (and thus granted to the charities we select).

Measuring (2) coincides with the goal of our research: determining what the charities we examine can be expected to accomplish, and how they compare to each other. There is often a great deal of uncertainty in comparing top charities to "average" charities, precisely because a lack of measurement is characteristic of "average" charities; but just looking at differences in our top candidates can give a sense of how widely strategies diverge and how much can be gained by more informed decisions.

(3) is the hardest to measure, because the spread of ideas is complex and wide-ranging; so although we can track changes in dialogue, attributing them to our specific activities is difficult. That said, the uniqueness of our message at this point in time makes it relatively feasible to trace relevant ideas back to us, something we’re currently working on documenting.

These metrics are complex and cannot be precisely quantified, just like the metrics we apply to our applicants. However, as with evaluating our applicants, we believe we can evaluate ourselves using a combination of empirical data, analysis, judgment calls, and guesswork (rather than simply the latter); we are currently writing up an end-of-year report on GiveWell that aims to do precisely this.

We strongly feel that it is. We believe, mostly intuitively but also based on the research we’ve done, that an excellent charity can be several times more effective than an average one - and that many charities may not be helping people in any significant way at all. In this context, spending time and money on evaluating charities - rather than just funding them - seems easily justified.

We believe that our research speaks for itself on this matter; we encourage you to read our reports and determine whether the differences between charities that we’ve identified are large enough to justify our costs.

We have not yet been required to conduct a financial audit or file a Form 990, as our organization was founded this year. We will make our Form 990 available publicly on our website.

We do not know of a good "one-size-fits-all" way of evaluating charities, so we don’t try to do so. Instead, we try to deeply understand the activities a charity carries out, the problems it addresses, and the evidence for its effectiveness. This is an enormous amount of work for even one charity, which is why we use a Round 1 application to go from a large field of applicants to a set that we can reasonably evaluate.

Rather than trying to evaluate every charity in the world, our approach is to use heuristics and imperfect information to find the ones we can have the most confidence in. We believe this is a superior approach for a donor whose primary aim is to help people as well as possible, not to evaluate a particular charity.

We don’t believe there’s any such thing as perfect information, and it isn’t realistic to thoroughly evaluate every charity in the world. That’s not our goal. Our goal is to find charities that we have high confidence in, and provide much better information than an individual can get on his or her own (as well as to create a starting point for a global dialogue around how to improve the world as well as possible).

Our recommended charities are the organizations we would bet most confidently on, based on the information we have - which is very far from perfect information, but also very far from what individual donors would have access to otherwise.

We generally don’t think this is a major concern, for two reasons:

• We believe that those most likely to use our site are donors who do not have a strong prior commitment to a particular charity. At this early stage, we don’t expect our analysis to sway those with deep-rooted personal connections and commitments to the organizations they already support.
• If this turns out not to be the case, it will be because donors are choosing to trust our analysis over their existing convictions. We see this as a purely good thing - a case of funds' being redirected to superior organizations. In all other areas of our economy, people demand the best deal they can get, and fund only the companies that are the best at providing it; we believe that consumers are generally better off for this dynamic, and would be happy to see the same dynamic benefit the people charities seek to help.

We generally don’t think this is a major concern. Some people recognize the complexity and difficulty of what charities do, and would feel more confident in their giving if they could see the truth; others naively expect perfection, and would feel less confident in their giving if they saw facts that contradict marketing materials. We would guess that our website is likely to be used by the former and ignored by the latter, resulting in more giving rather than less. More on this idea in this blog post.

Our official bylaws state that the Board of Directors must have at least three full meetings per year, consistent with the guidelines of the Better Business Bureau. We meet more often when there are major issues to discuss; we have had four meetings since our project began last June.

We use Network for Good to process donations to recommended charities, rather than linking directly to the charities' own donation forms, because:

• Network for Good allows anonymous donations (something that not all charities offer on their own websites).
• Network for Good provides a single set of URLs; thus, we don't have to change our links every time an individual charity updates its website.
• Network for Good allows us to track the donations that are made through GiveWell, which in turn helps us to monitor our own effectiveness.

Evaluation process and possible pitfalls

The short answer is that we have no universal or one-size-fits-all way of measuring an organization’s impact. Unlike many evaluators, who insist on "objectivity" and assess only things that can’t be argued about (and thus, to our minds, things that matter little if at all), we take the approach of making subjective judgment calls - informed as much as possible by empirical evidence and thorough research - and making our reasoning as transparent as we possibly can; this way, others can come to their own conclusions as well as critique us, which helps us to continue learning and improving.

Generally, we evaluate an organization within the context of a predefined cause and goal, which we lay out explicitly at the top of our cause overview (i.e., saving lives in Africa or employment assistance in New York City). We do what we can to get a sense of how well the organization accomplishes this goal on a "per-dollar/per-year" basis. However, when we do so, we generally consider any such estimates to be extremely rough, and also place weight on general organizational quality (as we perceive it), and benefits unrelated to the cause (for example, even in the "saving lives in Africa" cause, we give strong weight to organizations that have non-mortality-related benefits).

We use empirical data and hard analysis to inform our decisions as much as is feasible (and always have much higher confidence in an organization that can demonstrate its impact empirically), but there are always limits to what can be proven about an organization’s impact on human lives. We quantify impact as much as we can to get a sense of cost-effectiveness, but there are always limits to how well two organization’s effects can be compared. We consult with people who are highly experienced with the issues in question, but we also believe that fresh and outside perspectives are important, particularly for emotion-laden issues such as the ones we deal with.

We believe that trying to eliminate the human element from giving is futile; we aim to fully incorporate this human element, as all grantmakers do, while (a) informing our decision as much as possible with research and analysis; (b) being transparent enough that one who wishes to examine our reasoning can do so.

Ranking charities has the following benefits:

• Some donors do not have the time or inclination to perform their own comparisons; see us as hard-working, passionate, intelligent people who make reasonable decisions; and therefore prefer to put their money where we’re putting ours. We find this wholly reasonable, and seek to help these donors.
• Ranking enforces discipline, which in turn enforces thorough analysis. It’s the intellectual exercise of comparing two charities that leads us to ask more and more questions, using facts as much as possible before filling in what we don’t know with intuition and judgment calls. Publishing a standard set of information on each charity would lead us to request, publish, and use far less information.
• Ranking charities provides context for the information. We believe it is significantly easier to read and engage critically with our reviews - especially for someone who doesn’t have time to read every word of them - when it is clear where we are coming from and what conclusion our analysis is ultimately supporting.

As our recommendations and reasoning should show, we consider many aspects of our applicants besides data. That said, data absolutely plays a large role in our process, and charities with strong self-documentation have a decided advantage. We believe this is a strength of our process, not a weakness, for the following reasons.

• Self-evaluation and -documentation allows an organization to learn from its mistakes, in a way that nothing else can. Based on our own life experiences and our knowledge of the history of science, we believe that the world is far too complex to rely on what makes logical or intellectual sense. Furthermore, we generally focus on charities aiming to create significant and lasting life change; in this context, observations of how clients react to a program while enrolled in it simply aren’t enough to predict how the program will affect their lives. Ultimately, a charity that does not systematically and empirically evaluate itself is a charity that has little ability to test its reasoning and learn from its mistakes. We have much higher confidence in a charity that takes the effort to test its work against empirical reality.
• We seek charities that we can confidently recommend to individual donors. We believe there are times when intuitive and informal knowledge - particularly about the quality of an organization’s people - can substitute for empirical evidence. But when recommending that an individual send a check to an organization s/he has never personally interacted with, often serving people hundreds or thousands of miles away, we prefer to have as much factual evidence as possible. While some charities may have poor documentation while still doing valuable work, assessing this is not realistic for us or the donors we serve. Our goal is not to perfectly name the best charities, but to recommend excellent charities with high confidence.

We hesitate to answer this question in the abstract - it all depends on what is being measured, what can be learned from it, and how much the measurement costs. However, we firmly believe that measurement is worth significant expense, mostly for the reasons detailed immediately above.

Without rigorous self-evaluation, a charity can be missing out on opportunities to make its activities several times more effective (or to identify activities that aren’t effective at all); we believe that conducting this measurement will generally be more beneficial than spending the extra 10-20% of funds on activities that haven’t been critically evaluated.

As our rankings and reasoning should show, our "cost per life saved" type metrics do not play the dominant role in our decision-making. They are one factor we consider among many.

We believe that making a decision without any attempt to estimate such metrics would be an enormous mistake. It would be impossible, without such metrics, to reasonably assess high-intensity programs (programs that serve a few people in thorough, expensive ways) vs. lower-intensity programs (programs that serve many people in simpler less involved ways). If we believe both a high-intensity and low-intensity program to be helping people, it becomes essential to form a sense of whether the lower-intensity programs serves enough more people to make up for its presumably weaker effects.

But the above purpose is the only purpose for which we use "cost per life saved" type metrics. We also consider the overall viability, simplicity, and straightforwardness of a program; the empirical evidence that it is effective at all (we tend to favor programs we are highly confident in vs. programs that serve far more people with far more questionable effects); what we know of the organization as an organization; and anything that we feel our metrics (which are always simplifications) are leaving out.

In short, we ask anyone who feels put off by "cost per life saved" type metrics to read through a significant amount of our reasoning before accusing us of hyperfocusing on such metrics.

The short answer is that this is the same challenge faced by nearly all donors. If we insisted that all giving decisions be informed by thorough on-the-ground experience, we would be insisting that most people give only locally, or refuse to give at all.

We believe the benefits of being able to help the people in most need - no matter where they are - are too great to leave on the table. Therefore, we look for charities that have the capacity to document their work, and the ability to explain and prove it (as much as possible) to those who haven’t experienced it directly.

It’s also worth mentioning that the goal of a charity often has far more to do with what clients perform after a program than with their behavior during a program. For example, when evaluating job training programs, we fear that "direct observation" may reflect that a program is enjoyable, as well as that the people invested in the program are seeing what they want to see - rather than that the program successfully helps clients to get and keep jobs, something that can only be assessed through rigorous collection of data.

We do not believe that minimizing overhead cost is a good goal for a charity, or a good way to judge a charity.

It's probably true that any charity spending 75%+ of its budget on overhead is a fraud. But this is never the case with the charities that apply for our grants. Frauds take advantage of uncritical people, not rigorous and competitive evaluation processes. We believe that every single one of our applicants is well-established and well-intentioned - and that that isn't enough.

We believe that helping people is difficult. Charities work on challenging problems from disease epidemics thousands of miles away to the achievement gap. As with any other difficult problem, we believe that no amount of money can replace great ideas and intelligent plans. Spending 99.9% of funds on programs helps no one if these programs are ineffective; and making them effective means working hard on designing them, evaluating their results, and learning from them - all activities that often fall under the "overhead" classification.

When overhead is good

We would guess that a great charity is typically built on:

• Great people. Great people command high salaries, and keeping payroll low means losing them - not a good way to save money.
• Strong self-evaluation. Charities try to change people's lives; we believe a good one should know whether it has been succeeding, and should be constantly re-examining itself to see how it can do better. The only way to do this is through rigorous study of what happens to its clients (and, ideally, similar non-clients): is the charity improving their health, or grades, or income as hoped? Such studies are difficult and expensive, but the only alternative is to keep running a program every year, without ever finding out whether the theory behind it is working in practice. To us, that seems far more wasteful than spending money on evaluation.

These are both examples of "diverting money from programs", and both things we believe nonprofits should be doing more of, not less.

How much money goes to overhead? is a question about accounting classifications, not about lives changed. We believe that people currently focus on charities' accounting because so little information is available about anything else. The IRS Form 990 is the only form that is filled out and made publicly available by all charities, and it requires a detailed financial audit but practically no information on activities and results; so most attempts to rank charities try to find meaning in this information, even though you'll have trouble finding any experienced person in the field who believes there is any meaning there to be had.

We take a different approach: we focus on a charity's results (lives changed). We seek to fund charities that are great at helping people, whatever the mechanics of their operation and the details of their accounting. Evaluating results is time-consuming, difficult, and costly, so we can't rate all charities at once. But we think this evaluation is worth it. It's the difference between measuring success in dollars spent and measuring success in lives changed; it's the difference between throwing money at the world's problems and taking our best shot at solving them.

We do not believe that minimizing fundraising expenses is a good goal for a charity, or a good way to judge a charity.

As long as fundraising brings in more than it costs, it is a smart and reasonable investment for a charity. Even in cases where charities overspend on fundraising (something that standard financial documents, by themselves, can’t reveal), we believe that fundraising efficiency tells you more about the fundraising staff than about the programs and their effectiveness, which we believe are far more important in the scheme of things.

To be sure, wasting money is a concern - but the way we calculate our "cost-effectiveness" metrics (see above) incorporates all fundraising and administrative expenses, and therefore should include any problems caused by inefficiency - without hyperfocusing on this inefficiency. (For example, if Charity A is 10% more wasteful than Charity B, but runs programs that are twice as effective, it is clearly the better choice.) Ultimately, we believe that what matters isn’t how much is "wasted," in and of itself, but how much is accomplished and for how much money. That is what we seek to assess.

Thoroughly evaluating even a single charity is an enormous amount of work, and we want to use our limited time and resources optimally. With that in mind, we believe that identifying the best charities is far more important than identifying the worst.

In the business world, regulating/exposing/preventing gross misdeeds is important; but the majority of investment decisions have far less to do with these issues than with difficult and intuition-laden predictions of which companies will perform best. Relentlessly funding the best performers leads to constant improvement (for example, cell phones continually get better because people reward quality, not because they are vigilant about cell phone scandals). We believe it is this dynamic - demanding and rewarding the best - that is currently absent from charity, and that is far more needed than more coverage of scandals.

Other organizations do report on corrupt charities, likely better than we could. The American Institute of Philanthropy is one such organization.

To a large degree, yes. In our limited experience, small/start-up charities rarely have the capacity to strongly document their effect on people’s lives. We find this wholly appropriate given the nature of our project: finding charities that we can confidently recommend to individual donors, donors who have no direct experience with the charities and issues in question.

We believe that small/start-up charities generally should be (and generally are) funded by those who personally know and believe in the people running them; this direct and personal knowledge can easily compensate for a lack of hard empirical evidence. Indeed, GiveWell’s startup capital came entirely from our former coworkers, at a time when we had no product or results to show. We have no intention (or possibility) of replacing this sort of charity.

Instead, we seek to help individual donors, who are looking for a charity they can give to with confidence despite no personal connection. We believe that these donors are best served by mature, scalable, high-capacity charities that can convincingly demonstrate their impact and can reliably translate money into lives changed.

Extremely large charities are generally particularly well-suited to our evaluation process. In our experience, they generally have large enough development staffs to complete our application without requiring executive time; so even though our grants represent a small part of their budget, turnout among large charities has been strong (see this page for details). And we believe that an extremely large charity must have thorough self-evaluation and self-documentation if its executives are to manage it in any informed way; examining this documentation (or the lack thereof) can tell us a lot about the organization.

We generally conduct site visits once we have identified the strongest applicants within each cause. We have conducted site visits to The HOPE Program and Year Up (within the Employment Assistance cause); KIPP, Teach for America, and the Children's Scholarship fund (within the Education cause); and Nurse-Family Partnership (within the Child Care cause). We have not conducted site visits to our international applicants, but may do so in the future.

We continue our dialogue with all evaluated charities, to the extent that they’re interested in doing so. In the future, this dialogue may or may not go along with formal re-evaluations for further funding (this depends on which causes we cover in the future). Either way, if we lose contact with a charity to the point where we can no longer have confidence in it, we will eventually withdraw our recommendation. All recommended charities are organizations that we would bet on, with confidence, today.

The causes we analyze

Understanding even a single charity is difficult; understanding a cause is extremely involved. Rather than trying to reduce all of charity down to simple formulas, we do few enough causes at a time so that we can gain understanding of the issues, context, and charities.

Our choice of causes is influenced both by donor demand and by our own interests and passions.

We want your help deciding what to research next. If you'd like to influence our future research, please take our survey and let us know what causes are important to you.

Take our survey

We are currently developing an expansion model under which people will "vote with donations" on the direction of our future research. Those who donate directly to GiveWell will have a stronger say in what causes we pursue (though our own interests and preferences will also play in). If you are interested in participating in this, please be sure to answer Question 7 on the above survey: provide your email address and check the box saying "Contact me about future opportunities to influence the direction of GiveWell’s research."

The charities we evaluate

A full account of our process for finding applicants is available here; scroll to the bottom (or just click this link) for the list of contacted charities.

It takes a great deal of time and work to develop a good understanding of even one charity - that's work both for us and for the charity. The process we use, described here, involves soliciting interest from many charities, quickly identifying a few that are most likely to be able to demonstrate their effectiveness, and then studying these finalists in depth. The goal is not to rate each of the world's charities, but to find the ones we can confidently recommend.

While we hope to cover more charities as time goes on, there are many organizations that do not receive reviews for one of these reasons:

1. The organization is not a US-registered 501(c)(3) tax-exempt organization.
2. The organization is outside our current research scope, either because it is not within one of our causes, not large enough, or not within our geographical area of focus. See this page for more detail on how we determine eligibility.
3. The organization was invited to apply, but did not. This may be due to any combination of time constraints, staff capacity, etc.

Gaining true understanding of even a single charity is an enormous amount of work; so we cannot review a charity upon request. However, you can use this form to let us know about a particular charity, and we will be sure to include it if and when we research relevant causes in the future.

You can use this form to let us know about a particular charity, and we will be sure to include it if and when we research relevant causes in the future.

Once we choose a cause to investigate, we generally try to contact as many charities as possible within that cause (subject to the provision that they must be US-registered 510(c)(3) tax-exempt organizations), and use our Round 1 application to narrow the field down to finalists.

Last year, we found relevant charities mostly by purchasing and examining a set of over four thousand tax records (more on this process here); we also contacted any and all charities that were recommended to us by friends or others in the field. In future years, we will likely take a similar approach.

If you have a particular charity in mind that you would like to make sure we contact when appropriate (i.e., when we investigate a relevant cause), please use our charity submission form to register it with us.

We currently consider only US-registered 501(c)(3) tax-exempt organizations. Beyond that provision, any charity is eligible, as long as there is a reasonable case that it provides the most proven, effective, scalable solution for the cause in question.

We currently consider only US-registered 501(c)(3) tax-exempt organizations. It is common for such organizations to work outside the US (all of the charities we consider for international causes fit this description); it is also common for organizations based in another country to have a US-registered affiliate, in which case we can consider them.

Narrowing our geographic scope made it easier for us to avoid being stretched too thin in our startup year.

By choosing a single city within the U.S., we were able to learn about the relevant labor market and regulatory environment, essential for evaluating the organizations we looked at. In the future, we hope to examine more cities, though the extent to which we do so will depend on donor demand.

It was not possible to focus as narrowly for international causes, as most large eligible charities in these causes serve Africa (or all developing nations) as a whole. We chose to focus for this year on Africa, as it made it slightly easier for us to put together independent analysis; but on reflection, we would have been better off without such a restriction for developing-world causes, and in the future we are more likely to look at the developing world as a whole rather than focusing on Africa.

We are open to evaluating any US-registered 501(c)(3) tax-exempt organization that believes it can make a strong case for having the most proven, effective, scalable methods for accomplishing our goals (such as saving lives in Africa or providing employment assistance in New York City). In finding applicants for our first year, we focused on public charities (i.e., charities that get a significant amount of their support from individual donors), guessing that they would be better equipped for, more interested in, and more receptive to our evaluation process.

Not in the near future. Evaluating a single charity thoroughly takes an enormous amount of work, and we are much more interested in finding charities we can recommend with confidence (even if doing so requires heuristics and imperfect shortcuts) than in covering every charity. However, as we grow and cover more charities, we will organize what we have more robustly.

Advice

Because evaluating a single charity is an enormous amount of work, and because we don’t believe in "one-size-fits-all" evaluations, we hesitate to respond to inquiries as general as this one. If you have a specific question, we’re generally happy to share our opinions.

Unfortunately, we’ve found that evaluating a charity well is extremely difficult to do on a part-time basis (see our story), so we can’t be very helpful with this question. Your options include:

• Hiring a philanthropic advisor such as Geneva Global, Rockefeller Philanthropy Advisors to do custom research. Disclosure: GiveWell Board member Tim Ogden is a former employee of Geneva Global.
• Taking your best guess with an organization you can volunteer with. Working directly with the organization may give you a better sense of its organizational quality over time.

We do not know of any easy way to get a good charity recommendation (it was this problem that led to the creation of GiveWell).

We recommend giving to the charity you have the most confidence in, regardless of its size. For a donor who has very little knowledge, however, we recommend large charities. It is true that small charities may be wrongly overlooked, but they can just as easily be rightly overlooked; just as with businesses, there is no particular reason to believe they are doing great work unless you have seen evidence of it. Though we would like to see more scrutiny on charities of all kinds, we would guess that - knowing nothing else about two charities - the larger one has probably survived more scrutiny over its history.

This is not our area of expertise; we focus on finding the best charities, not dealing with overly solicitous ones.

Although we disapprove of Charity Navigator’s one- to four-star ratings, the website has some resources that are excellent when taken in context, including a writeup on how to avoid direct mail solicitations that we recommend.

Questions about our survey

We encourage you to skip this question if it bothers you. We realize that not all users will have a preference, but we feel that separating out "global" issues will give them undue appeal to the casual survey respondent, resulting in less useful information for us.

Our survey asks how important each cause is to you, and separates them by developing- and developed-world mostly for clarity (for example, something like "education" means very different things in the different contexts). Whether you value a cause is more important to us than whether you consider it to be "global" in nature.

No.

Site navigation

Use the links at the top of this page ("Saving lives," "Global poverty," etc.) to see our research and recommendations. We currently have research and recommendations available for saving lives in Africa and employment assistance in New York City; on other causes’ pages, you can sign up to be alerted as we publish more.

Yes, use our Get Involved page.

Yes, see our donation page.

A full account of our process for finding applicants is available here; scroll to the bottom (or just click this link) for the list of contacted charities.

See our story.

See our Official Records page.