ooh i really like this discussion. one thing i'd add - i've always thought resilient systems-building and good governance work that can protect against many kinds of risks has been super neglected by EA (and also by the world). i'm fairly sure at this point that it's a result of (in addition to what's already been named):

• EA's hard focus on marginal impact - systems-building work requires many things to shift before you see impact, so it's hard to justify one marginal additional person or dollar working on it
  • corollary to this is that EA is extremely focused on shifting behaviors and opinions of elites, rather than building mass movements or reaching many people or changing cultural norms or creating better structures. obviously this isn't true of all EA work but relative to e.g. anti-fascist and mutual aid movements it's a huge difference in theory of change
• pure vibes, which i think is obvious in some of the other comments too

incorrect order of operations, you should be selecting for a future wife who is also down to donate 10% of her income

yeah not sure unfortunately!

TLYCS's own pledge is progressive with income, for what it's worth https://www.thelifeyoucansave.org/take-the-pledge/ 

https://www.powerfordemocracies.org/research/our-recommendations/ !!

omg I have already left my EA comms job but I would be down to hang out and yap in there if that's welcome!!!

as far as I can tell the answer to this type of question is always that someone did a napkin calculation 10 years ago and decided that either (a) lots of funding within an arbitrarily-defined "cause area" means everything within that cause isn't neglected, or (b) affecting a large pool of funding isn't tractable enough and therefore not worth spending EA resources on, and then because of path dependency in the development of EA as a community of practice it's now just hard to gain traction or interest in cause areas outside of the EA canon

1000000% this is the messaging that resonates best for Giving Green donors, also precisely for the reasons you named. We did a bit of qualitative user research and persona work a few years ago on this, and it still continues to be a core theme with new donors (who range from EA-adjacent to classic environmentalist).  I don't have anything useful to add, just here to corroborate!

edit: ok I have one useful thing to add, which is that, especially because of the type of work we fund, we explicitly don't use "certain impact" or "best charity" framing. but "certainty in giving" is different from "certainty in impact"!

I once saw on the Forum that someone had scraped the 990s from a bunch of EA and AI safety* orgs and put all the salaries in a spreadsheet, with names - it wouldn't be that hard to go from that to at least an estimate of what you're looking for, for the highest-paid employees. I can't find a link to the post anymore, and want to respect that they might have taken it down with good reason, but given it's public information, if some enterprising data-wrangling Forum-poster wants to dm me for it I'm not opposed to sharing the link...

*I do have a loose intuition that besides the grantmaker/grantee divide, the AI/not-AI divide within EA is driving some of the bizarre funding and salary dynamics

When early digital experiments don’t show ROI, many orgs seem to conclude that the channel itself is misaligned, rather than that execution, resourcing, or the evaluation window were insufficient. Given small budgets and high standards of proof, it’s not surprising those early attempts fail — but that doesn’t tell us much about the counterfactual of sustained investment.

yeah I basically think this is the problem, and agree that some level of investment would yield a return, but small orgs can't just keep putting in time and money for hypothetical return at some undetermined threshold! we're not for-profits that can take out loans or get VC money to sink into big upfront acquisition costs :') 

again, if any funders are interested in funding EA digital marketing experiments for audience growth, I'm all ears... I'd like to see a case study of what level of investment is needed for smallish orgs to see a return, especially for fundraising asks.