The GiveWell Blog

Challenges of transparency

When we first started GiveWell, we wondered why major staffed foundations didn’t write more about the thinking behind their giving (and the results of it), in order to share their knowledge and influence others. We’ve tried to counterbalance normal practice by making transparency one of our core values.

Our commitment to transparency is as strong as it’s ever been; we derive major benefits from it, and we believe there’s far too little public information and discussion about giving. At the same time, we’ve learned a lot about just why transparency in philanthropy is so difficult, and we no longer find it mysterious that it is so rare. This post summarizes what we see as the biggest challenges of being public and open about giving decisions.


  • Everything we publish can help or hurt our brand. We put substantial effort into the accuracy, clarity and tone of our public content.
  • In most cases, writing about our thinking and our results also means writing about other organizations (the organizations we recommend and support, both via our traditional work and via the Open Philanthropy Project). We don’t want to hurt or upset other organizations, and we put substantial effort into making our public content both (a) amenable to the organizations we write about and (b) fair and complete in its characterization of our views.
  • The level of transparency we seek is unusual, meaning it often takes substantial effort to communicate our expectations and processes to the organizations we recommend and support.
  • The interaction of the above challenges can make it extremely difficult and time-consuming to write publicly about grants, recommendations, and grantee progress. In addition, it can be the cause of major delays between drafting and publication: much of our content takes weeks or months to go from draft to published piece, as we solicit feedback from parties who might be affected by the content.
  • The costs of transparency are significant, but we continue to feel they are outweighed by the benefits. Public writeups help clarify and improve our thinking; they play a major role in our credibility with our audience; and they represent a step toward a world in which there is far more, and better, information available to help donors give well.
  • We don’t think it necessarily makes sense for all philanthropic organizations to put as much effort into transparency as we do. Rather, we see transparency as one of the core areas in which we are trying to experiment, innovate, and challenge the status quo.

Challenge 1: protecting our brand
Because of the work we’ve put into explaining and defending our positions in the past, we benefit substantially from our reputation and from word-of-mouth. Nobody checks every statement and footnote on our site; even our closest readers rely often on the idea that content under the GiveWell name has a certain degree of thoroughness, reliability and clarity. (We believe a common way of approaching GiveWell content is to spot-check the occasional claim, rather than checking all claims or no claims; in order for our content to perform well under arbitrary spot-checks, our content needs to have fairly consistently high quality.)

Somewhat ironically, this dynamic means we’re hesitant to publish content that we haven’t thought through, checked out, and worded carefully (in order to say what we feel is important and defensible, and no more). We feel that poorly researched or poorly worded content could erode the brand we’ve built up, and could make people feel that they have to choose between checking everything we write themselves and simply placing less weight on our claims. (In general, most of our busy audience would likely choose the latter in this case.)

Giving decisions are generally impossible to justify purely with appeals to facts and logic; there are many judgment calls and a great deal of guesswork even in the most seemingly straightforward decisions. This makes it particularly challenging to write about them while preserving a basic level of credibility and a strong brand, and we don’t know of clear role models for this endeavor. (A funder once told me that s/he didn’t want to publish the reasoning behind giving decisions because this reasoning wasn’t up to academic standards, and so would not be perceived as reasonable or credible.)

Rather than aim to write only what we can back with hard evidence, and rather than write everything we believe regardless of the level of support, we put a great deal of effort into being clear about why we believe what we believe – whether it is because of solid evidence or simply a guess. (Phrases such as “we would guess that” are common in our content.) This allows us to share a good deal of our thinking (not just the parts of it that are strongly supported) while still maintaining credibility. But it requires a careful, thoughtful, and somewhat distinctive writing style that has been an ongoing challenge to develop and maintain.

As our brand becomes stronger, our audience becomes broader and our staff grows, the challenges of maintaining the appropriate style – and backing up our statements appropriately – intensifies. For example, we now put most public pages through a “vet” – in which a staff member who was not involved in writing the page goes carefully through its statements, making sure that each is appropriately supported – before publication. (We do not do this for all pages, and we generally do not do it for blog posts, which are more informal.)

Challenge 2: information about us is information about grantees
We seek to be highly open about the lessons we’ve learned and the results we’ve seen from our work – including developments that contradict our expectations and reflect poorly on our earlier decisions (which are often particularly valuable for learning). Because our function is to recommend other organizations for donations and grants, being open about our performance almost always means being open about another organization’s performance as well. (For simplicity, the rest of this section will refer to GiveWell-recommended organizations, as well as Open Philanthropy Project grantees, as “grantees.”)

While we want to be open, we don’t want to create a dynamic in which working with us creates significant risks for grantees. (This could lead good organizations to avoid working with us.) So we’ve had to find ways of balancing the goal of openness with the goal of making it “safe” for an organization to work with us. Doing so has been a major challenge and the subject of many long-running discussions, both internally and with grantees.

Things we’ve done to strike the right balance include:

  • Putting serious effort into communicating expectations up front. Simply saying “we value transparency” is not enough to communicate to a grantee what sorts of things we might want to write in the future. We generally try to send examples of past things we’ve written (such as our 2013 updates on Against Malaria Foundation and Schistosomiasis Control Initiative), and we often try to agree on an initial writeup before going forward with a grant or recommendation.
  • Giving grantees ample opportunity to comment on pending writeups that discuss them. There have been cases in which a writeup has been the subject of weeks, or even months, of discussion and negotiation.
  • Giving grantees a standing opportunity to retract non-public information, including even the fact that they’ve participated in our process. (Organizations considered as potential top charities have often been given the option to withdraw from our process and have us publish a page simply saying “Organization X declined to participate in our process”; this option has sometimes been invoked quite early in the process and has sometimes been invoked quite late, after a draft writeup has been produced and shared with the organization.)
  • Being generally hesitant to run a writeup that a grantee is highly uncomfortable with. We’re often willing to put substantial effort into working on a writeup’s language, until it both (a) communicates the important aspects of our views and (b) minimizes grantees’ concerns about giving misleading impressions.
  • We are in the process of creating a more formal process for negotiating about transparency with grantees up front. This process will draw on the agreement we negotiated with The Pew Charitable Trusts.

Challenge 3: transparency is unusual
In general, unusual goals are harder to achieve than common goals, because the rest of the world isn’t already set up to help with unusual goals. When we ask for budgets, project plans, confidentiality agreements, proof of 501(c)(3) status, etc., people immediately know what we’re seeking and are ready to provide it. When we bring up transparency, people are often surprised, confused, and cautious. In some cases people underestimate how much we plan to write, which could lead to problems later; in other cases people fear that we will disclose information carelessly and indiscriminately, leading them to be be highly wary. Discussions about transparency often involve extensive communication between senior staff at both organizations, in order to ensure that everyone is clear on what is being requested and expected.

We believe that we could achieve the same level of transparency with far less effort if our practices were even moderately common and familiar.

The difficulty of writing about grants
The interaction of the above challenges can make it extremely difficult and time-consuming to write publicly about grants, recommendations, and grantee progress. We can’t simply “open-source” our process: each piece of public content needs to simultaneously express our views, maintain our credibility, and be as amenable as possible to other organizations discussed therein. Much of our content takes weeks or months between drafting and publication.

With this in mind, we no longer find it puzzling that existing foundations tend to do little substantive public discussion of their work.

Benefits of transparency
The costs of transparency are significant, but we continue to feel they are outweighed by the benefits.

First, the process of drafting and refining public writeups is often valuable for our own thinking and reflection. In the process of discussing and negotiating content with grantees, we often become corrected on key points and gain better understanding of the situation. Writing about our work takes a lot of time, but much of that time is best classified as “refining and checking our thinking” rather than simply “making our thinking public.”

Second, transparency continues to be important for our credibility. This isn’t because all of our readers check all of our claims (in fact, we doubt that any of our readers check the majority of our claims). Rather, it’s because people are able to spot-check our reasoning. Our blog generally tries to summarize the big picture of why our priorities and recommendations are what they are; it links to pages that go into more detail, and these pages in turn use footnotes to provide yet more detail. A reader can pick any claim that seems unlikely, or is in tension with the reader’s background views, or is otherwise striking, and click through until they understand the reasoning behind the claim. This process often takes place in conversation rather than merely online – for example, see our research discussions. For these discussions, we rely on the fact that we’ve previously reached agreement with grantees on acceptable public formulations of our views and reasoning. Some readers do a lot of “spot-checking,” some do a little, and some merely rely on the endorsements of others. But without extensive public documentation of why we believe what we believe, we think we would have much more trouble being credible to all such people.

Finally, we believe that there is currently very little substantive public discussion of philanthropy, and that a new donor’s quest to learn about good giving is unnecessarily difficult. Work on the history of philanthropy is sparse, and doing new work in this area is challenging. Intellectuals tend to focus their thoughts and discussions on questions about public policy rather than philanthropy, making it hard to find good sources of ideas and arguments; we believe this is at least partly because of the dearth of public information about philanthropy.

We don’t think philanthropic transparency is easy, and we certainly don’t believe it’s something that foundations can jump into overnight. We don’t think it necessarily makes sense for all philanthropic organizations to put as much effort into transparency as we do. Rather, we see transparency as one of the core areas in which we are trying to experiment, innovate, and challenge the status quo.

In doing so, we hope to continue refining the processes necessary to achieve transparency, encouraging future (as well as present) foundations to adopt them, and making it easier for future organizations to be transparent than it currently is for us, so that one day there will be rich and abundant information available about how to give well.

Our ultimate goal is to do as much good as possible, and if we ever believe we might accomplish this better by dropping the emphasis on transparency, we will give serious consideration to the possibility. But at this time, the chance to promote philanthropic transparency is a major part of the case for GiveWell’s future impact, and we plan to retain transparency as a costly but essential goal.


  • GiveWell’s transparency has been a huge deal for me and presumably other donors in just figuring out what effective giving looks like. When I first followed y’all’s work I shared some general beliefs (on the importance of aid to the global poor, and of cost-efficiency and evidence) but there was a lot more to learn (on the relevance of room for funding, how much uncertainty remains even after well-done studies, the trickiness of things like negotiation with the countries where aid is needed). I don’t think those are things one can fully appreciate without seeing how they shape real decisions. It’s huge to me that GiveWell helps nonprofessional donors really understand effective giving.

    And without openness I don’t think the recs could have become, or can remain, so important. Other practically-minded groups converged on similar recommendations to GiveWell’s, probably influenced by the open, rigorous assessments even if they don’t say so. And recs that I trust alright in the abstract, but can’t see the details of, aren’t necessarily ones I’m psychologically able trust a chunk of my annual income with; not sure that’s rational, but it’s true for me and probably many others.

    I get grantees’ worries–it’s human to want to see only glowing reviews. But they can hurt themselves being too protective. I’m in tech, and I recently saw a company put language in its terms of service that outright bans testing their product’s reliability or performance. That’s a remarkably effective way to make me doubt not only the reliability and performance of the product, but the way they do business–can’t they defend the totality of their work, warts and all? If they’re trying to hide their flaws, do they never intend to fix them?

    A mixed report from GiveWell at least puts a charity on my radar–it means what they do is interesting enough to put them towards the top of a very long list of causes, and marks them as folks to watch. The negatives in any report are, inevitably, rebuttable propositions, and GiveWell seems happy to give anyone they deal with a chance to rebut. But no report leaves them entirely out of consideration for me, and vague statements are a sign that the organization is wary of scrutiny which, like the “don’t benchmark us” clause, puts a worse taste in my mouth than the facts are likely to do.

  • So, talking too much because this is a really hard question and seems fundamental to GiveWell’s work. As ever, I know I don’t know anything about this, and write out of sympathy with folks who have a very complicated job as much as anything.

    1. Many organizations are probably getting a writeup that’s great for the organization and don’t recognize it. Every top-charity recommendation and intervention report GiveWell publishes comes with huge caveats. GiveWell is relevant *because* it’s candid and rigorous. And the active audience that reads these things will hugely appreciate when a group is open about things including flaws, risks, and difficulties, and can be skeptical about anything that might smell cagey. Charities that don’t completely withdraw are going to wind up with some opinion published, and it’s going to be mixed to some degree, and better that they come out looking like forthright transparency champs.

    2. Figuring out the right relationship and framing for these conversations seems incredibly fraught. Groups doing GiveWell’s sort of work need to find you good to work with, yet you can’t promise they leave ecstatic, precisely; you’re working for people needing help, not for aid groups. Indeed, negotiating against yourself by compromising before it’s ever requested or being too reassuring could plausibly set expectations too high and backfire. Letting groups opt out as if the review process had never started, and making it clear you’ll publish an accurate representation of your views otherwise, seem like a smart couple of ideas in a very tough situation. I don’t have much useful to say, but it just seems hard, in an unavoidable way.

    3. It’s good that y’all seem to be finding a sensible place in your work for folks that don’t meet all of the criteria to be a top charity this year. The world needs more things pursued than just the two or three interventions where a dollar has the highest proven marginal impact, and many organizations are not in a spot where they’re ready to accept millions to scale but might get there. The Labs and the other work GW does seem like important things that occupy places in between the fuzzy notion of “doing important work” and the very specific criteria GW uses for yearly recs.

    Thanks for all the work you do.

  • Alexander on September 15, 2014 at 7:22 pm said:

    R – thanks for the thoughtful comments!

  • Accuracy and clarity are two important components which you must take care of. We love to see transparency in all the matters.

  • Colin Rust on September 25, 2014 at 12:53 pm said:

    Organizations considered as potential top charities have often been given the option to withdraw from our process and have us publish a page simply saying “Organization X declined to participate in our process”; this option has sometimes been invoked quite early in the process and has sometimes been invoked quite late, after a draft writeup has been produced and shared with the organization.

    Thanks for explaining this. I imagine that must be pretty frustrating for GiveWell staff when a draft is pulled late in the process at the request of the charity being reviewed. I can certainly appreciate why you might give a charity this option: knowing they can silently back out will encourage them to engage and be open with GiveWell in a risk-free way.

    Still, as you formalize your process — and as GiveWell has become more established — I’d hope you’d consider revisiting this by negotiating up front permission to publish a report (whatever your conclusions). Perhaps the charity should also have the right to have a reply attached to a report if they feel it is wrong-headed.

  • Alexander on September 25, 2014 at 8:38 pm said:

    Colin – thanks for the suggestion. We’ve continued to give charities early in our process the opportunity to opt out at any time, but we’ve moved in the direction that you suggest. With funding from Good Ventures, we’re planning to make “top charity participation grants” to organizations that agree to the publication of a preliminary review and commit to allowing us to publish a final, more comprehensive, review.

    We always allow organizations to publish a reply along with our review if they wish to, though relatively few tend to take us up on the offer.

  • Chris Corliss on September 30, 2014 at 11:57 pm said:

    Just two things

    1. Transparency is asymmetrical. Your readers and critics don’t open themselves in any way remotely similar to the way you do.

    2. There is no comment page on this. I want to critique you.

  • Alexander on October 1, 2014 at 2:26 am said:

    Chris – Thanks for the comment. We only have comments enabled on the blog, not on the whole website, but you’re welcome to email us at with any feedback you have on that page or any others.

  • Colin Rust on October 4, 2014 at 9:14 am said:

    Alexander, thanks for the reply. I still wonder if even without the guaranteed financial incentive, charities could reasonably be required to allow you to publish once the process has gotten to a certain point.

    I think there are broadly two kinds of issues:

    1. the charity doesn’t like what GiveWell has to say about their flaws (even after GiveWell has dong everything they can to address their point of view as you discuss above) and
    2. working with GiveWell is just a lot of work.

    I have more sympathy in the latter case. But I’d hope, while no doubt sometimes unpleasant, going through the analysis with GiveWell can actually be helpful for the charity and potentially point to them in directions they can improve, even if they don’t ultimately get prominent billing or funds from GiveWell.

    Would you consider publishing statistics, say annual counts of charities you worked on and got to various points with?

    Transparency is hard, no question about it, doubly so with the prevailing culture among existing foundations.

    Alexander, re Chris’ comment: I can appreciate why you don’t have comments enabled for the whole website. For one thing, with comments sparsely scattered in many different places, they would be easy to miss. But why not have a monthly open thread? For a few things, like say a list of typos on a report, email is great, but with substantive critiques there’s at least a potential benefit to broader discussion.

  • Alexander on October 7, 2014 at 7:06 pm said:

    Colin, thanks for the suggestions. I think both publishing annual counts of charities that we substantively interacted with (and where they fell out in the process) and periodic open threads for comments are good ideas. We’ll plan to try them out.

Comments are closed.