The Strengths and Weaknesses of Different Cause Areas and Charity Ideas
- Joey Savoie

- Jan 23, 2020
- 1 min read

CAUSE AREAS
Unlike in previous years, we decided in 2020 to consider multiple different cause areas, which leaves more room for cause comparison. We think that generally, both entrepreneurs and donors have specific cause areas in mind when they attend or support our program. However, some have asked us for a sense of how the different cause areas, and more importantly, charities within them, compare. We think each area has its strengths and weaknesses and at this level, it's hard to reliably compare because many assumptions (both ethical and epistemic) need to be made. This article offers a starting point for such comparisons. First we show roughly how our 2020 cause areas compare, and then we look back at and compare our recommended interventions from research done in 2019. In 2019, our research focused on interventions for animals and within global health. As of 2020, we are considering the following four cause areas:
Mental health
Family planning
Animals
Health policy
2020 RESEARCH: COMPARING CAUSE AREAS
The table below shows our weighted factor model framing for our 2020 cause areas. Each area is color-coded from strongest to weakest.
Area | Direct cost-effectiveness | Relevant evidence | Execution difficulty | Non-captured externalities** | |
Mental health | Moderate | Some | Funding | Easier | EA movement |
Family planning | Low | Moderate | Logistical | Complex | Child, animals |
Animals | High | Low | Talent | Easier | Bar setting |
Health policy | High | High | Funding | Complex | Precedent |
* If the limiting factor cell is red, this means that the limiting factor will be met very quickly. Green means that the factor will be hard to meet. ** If the non-captured externality cell is green, this means that the externalities are large and positive. If the cell is red, this means that externality is small.
Another way to frame this is in terms of more specific key strengths and weaknesses.
Area | Strengths | Weaknesses |
Mental health |
|
|
Family planning |
|
|
Animals |
|
|
Health policy | Naive cost-effectiveness estimates show higher cost-effectiveness than standard global health interventions and maybe all other human-focused areas
|
|
2019 RESEARCH: COMPARING INTERVENTIONS
Area | Direct cost-effectiveness | Relevant evidence | Limiting factor | Execution difficulty | Non-captured externalities** |
Immunization reminders | Low | Moderate | Funding | Easy | Limited |
Tobacco taxation | High | Mixed | Policy windows | Complex | Precedent |
Iron and folic acid fortification | Moderate | Moderate | Logistical | Moderate | Moderate |
Area | Strengths | Weaknesses |
Immunization reminders |
|
|
Tobacco taxation |
|
|
Iron and folic acid fortification |
|
|
Area | Direct cost-effectiveness | Relevant evidence | Limiting factor | Execution difficulty | Non-captured externalities |
Dissolved oxygen for fish | Moderate | High | Logistical | Moderate | Moderate |
Food fortification for egg-laying hens | Low | High | Logistical | Low | Moderate |
Ask research | High | Moderate | Talent | Moderate | High |
Animal careers | Moderate | Low | Replicability | Low | High |
Area | Strengths | Weaknesses |
Dissolved oxygen for fish |
|
|
Food fortification for egg-laying hens |
|
|
Ask research |
| The impact depends strongly on the effectiveness of corporate and governmental campaigns
Impact relies on NGOs and organizations updating based on research
|
Animal careers |
|
|




The table format is helpful, but I kept wanting a concrete example charity in each cause area to anchor what “easier” vs “complex” execution looks like in practice. Like, what does “complex” mean here — regulation, stakeholder management, long time horizons, all of the above? Weird comparison, but it’s similar to how “this site” this site can give you a neat plan, yet the real difficulty is still the day-to-day execution when constraints pile up.
I like that you separate “direct cost-effectiveness” from “non-captured externalities,” because otherwise health policy tends to look unbeatable on paper. But I’m not sure how you avoid double-counting “precedent / bar setting” between the externalities and the execution difficulty buckets — feels like they bleed into each other. Funny enough, I’ve had similar “what’s actually being measured?” thoughts when people compare creative tools like imgg on output quality vs usability vs downstream impact on a workflow.
Interesting that family planning comes out “low” on direct cost-effectiveness here — I’ve seen people assume it’s automatically a top-tier global health lever, but the logistical/complexity side seems to be where the reality bites. Would be helpful to see an example of what would make you revise a score (new RCTs, better implementation partners, etc.). Slight tangent: the way categories shape what gets attention reminds me of directories that let you where to submit ai tool and how framing affects what people actually discover.
The “hard to reliably compare because many assumptions need to be made” line is doing a lot of work here, and I appreciate that you’re explicit about it instead of pretending the model is purely objective. Practically, I wonder how you’d advise a donor who’s deciding between two areas but only has limited time to dig into the caveats — do you default to the area with clearer evidence or the one with higher upside? It’s a bit like scheduling across teams where even something as simple as https://caesarcipher.org/converters/est-to-cst-converter can reduce mistakes, but you still need judgment calls around uncertainty.
I’m curious how stable these ratings are year to year — especially “relevant evidence” for animals vs mental health, since it feels like the evidence base can change quickly once a few big studies land. Also the “non-captured externalities” bit seems like the place where people’s values sneak in the most. This kind of tradeoff table is oddly like how I overthink puzzle games like BlockBlast where one constraint dominates everything once you’re a few moves in.