May marked my three-year anniversary as a Philanthropy Advisor at GiveWell. It’s a job I adore (as I’ve written about here and here), and I’ve recently been tasked with the exciting process of interviewing candidates for our growing team.

One of the best questions I’ve been asked in this process is: What’s the hardest part about fundraising for GiveWell? The short answer: GiveWell is funding constrained, but we can’t point at a specific opportunity and say, “If you donate now, here’s the impact your donation will actually cause.”

Instead, our answer is fairly abstract, and pretty far from traditional fundraising language. We tell donors that we would spend additional money on opportunities at or above our cost-effectiveness bar (which translates to saving a life for about $5,000), but we’re unable to explain in advance precisely what we will allocate additional funds to. That answer isn’t as compelling as telling someone a vivid story about how their money alone would allow us to fund a great program we’ll otherwise have to decline, but it has the advantage of being completely accurate and true.

It’s natural that people aligned with GiveWell’s approach would ask about the true impact their funds unlock, and also about what would happen if they don’t give. After all, these are key questions¹To learn more about how we fund, check out our “How We Work” blog series. we think about as a funder. Donors make careful decisions about how much to give, when to give, and where to allocate gifts according to their priorities; to make those decisions, they need to know what we’d do with their money and what we would be prevented from doing if they don’t donate.

But the GiveWell research process doesn’t lend itself to easy answers to these questions. So indulge me, if you will, in an extended metaphor:

I’m at the grocery store shopping for a huge dinner party. I choose everything on my list, get to the checkout, and realize I don’t have enough money to cover my selections. I choose a few things to put back, and then the kind bystander behind me magnanimously steps up to pay for the things I had cut.

This illustrates what many donors want: they don’t want to give if we can do just as much good without them, but they certainly don’t want us to be without the funds we need for the things we deem necessary. So they say: “Hey, if you’re ever at the checkout and come up short, give me a call!”

But that’s not how GiveWell shops. Our research team is constantly looking at how many items are left on the list and making small adjustments along the way, based on how much money we have available: changing from a wild mushroom risotto entrée to simple rice and beans, buying fewer cheeses for the appetizer tray, deciding to invite one fewer guest.

If someone were to hand our research team an extra metaphorical $100 bill as they entered the metaphorical store, the products in our cart at the end of the shop would be different than if they shopped with our original budget in mind. But if someone gave us an additional $100 at checkout, the adjustment we might be able to easily make (say, grabbing a side we had cut while shopping) might not be the best use of money. The best use of that unexpected money might ultimately be saving it for the next time we host a dinner party, because (to further complicate an already unwieldy metaphor) we’re shopping in a store where items might not be on the shelves when we circle back to them. Our research team makes time-sensitive grants throughout the year, and we’re not able to retroactively increase the size or specifics of most grants after the fact: the organizations we fund start cooking once their groceries are delivered.

Unlike our metaphorical grocery shop, the budget compromises we make throughout the year matter a lot. Less money available means:

• We might fund a new program in one country rather than two (reducing the number of people who benefit from the program and reducing our ability to learn how the program works in different contexts).
• We might reduce an initial grant horizon from five years to three years (creating slightly more work for the research team to investigate a possible renewal grant in the future and also slightly decreasing the grantee’s ability to engage in long-term planning).
• We might decline to fund a grant that’s right at our cost-effectiveness threshold in favor of preserving funds for opportunities later in the year that we expect might be more cost-effective (passing on a good opportunity we won’t be able to come back to).²These are all generalized examples of real adjustments we are likely to make to 2024 grants currently under investigation by our research team.

When a donor asks what our research team would do with their additional funding, it’s hard to identify any one thing with precision. And when we are asked at the end of the year what we left unfunded that someone could go back and cover, there isn’t one crisp answer; instead, there’s a trail of prudent adjustments that add up to a tremendous number of additional lives we might have saved.