Regarding the average view duration %: I think it makes sense for longer videos to have somewhat lower percentages. Fewer people are willing to sit through a long video, compared to a short one, so it is logical to have more people fall off throughout the video. But if you measure the average view duration in terms of minutes, not percentages, your results are really impressive: AI 2027 and MechaHitler have AVDs of ca 10 minutes and 13.5 minutes, respectively.

There is no need to do public advocacy if you feel this is not something you would enjoy/be good at. If at a later stage you feel more comfortable talking about effective giving and veganism with others, that's great. But not mandatory.

If you can just keep donating a part of your income regularly to effective charities, do that. That is probably the most useful thing you can do right now, far more effective than arguing with people on Reddit.

Digging wells in Niger seems surprisingly cost-effective. Thanks for writing this analysis, really good work!

Some questions I had:

1. If we assume that each well serves an average of 1,200 people, does this mean that none of these 1,200 people had access to clean water before this well was dug? Or are there cases where an existing well was already supplying clean water, but it was located further away? In other words, can we attribute all of the credit for giving these 1,200 people access to clean water to this well and this well only?
2. How much water can one well supply per day/hour/minute? Let's say we want to provide at least 20 liters per person per day. That would be 24,000 liters per day to supply 1,200 people. Can the well handle such a level of demand?

FYI, the Institute of Economic Affairs, where Snowdon works, is known to have received funding from alcohol, tobacco, sugar, gambling, and oil industry lobbies.^[1] The UK's Charity Commission recently opened an investigation into IEA.^[2]

One can, indeed, make the argument that an outright ban on smoking would violate the rights of people who have been smoking for their entire lives.^[1]

But most citizens would probably agree that at least some level of government intervention is justified for tobacco, when you consider that smoking is responsible for the deaths of 7 million people every year.

It does seem necessary to get governments to spend more of their own money on health, indeed. Do you think it would make sense to fund charities to try to convince governments to invest more in health (perhaps by also helping them increase their tax revenues, via increasing tax collection efficiency)?

Thanks for sharing this analysis, very interesting.

Haven't yet decided whether I approve of this name change. But it is certainly good news for the Nigerian economy (and maybe also the Republic of Congo) :) I just snapped up the coefficientgivi.ng domain name (inspired by shortened URLs like spoti.fi and youtu.be). Let me know if you want it (for free of course)!

Given that you are American, it is quite logical to first focus on America. But this tax incentive looks like something that could also be implemented in other countries around the world. Maybe this is an idea that AIM's Charity Entrepreneurship program could consider incubating?

“Scriptwriting is our current biggest bottleneck” Can you elaborate a bit? Is the hard part deciding what the messages are that you want to deliver or rather how to deliver them?