Rob Wiblin

And maybe there's others who would rather try something new and try to improve the model.

Is that it?

Yeah.

Yeah.

What do you think of the idea that the end game ought to be that, I guess, the state level departments of health in India basically absorb your staff and your project and just start doing it because I suppose they have the revenue raising capacity and maybe the economies of scale more broadly to do this in a sustainable way?

It's interesting that you say it seems like the more effective altruist-minded donors are more interested in just delivering, just like scaling the project that you're doing, whereas it's maybe the less effective altruist-minded donors who are more interested in this handoff to the government.

You might think that it could easily be the reverse, where the people with the effective altruist mindset are like, well, wouldn't it be much better if we could just fund this for a little while and then have the government take it over and they could deliver it to more people and we wouldn't have to pay for it anymore?

Do you know what's going on there?

Yeah.

Okay.

So the motives and interests of the funders is one issue.

Yeah.

Are there other challenges to reaching a full scale?

Yeah.

So you're saying a potential trap or a potential problem would be if someone tries to replicate these RCTs and finds that in this second case, the SMS messages or the ambassadors don't work or the effect size is a lot smaller.

And then what would you say to your funders?

Maybe you want to be able to point to work that you've done assessing your own impact in exactly your context because you are measuring it and be able to say, well, we do have a good sense of what impact we're having at least.

I guess a long crazy view might be this has been tested quite a lot already.

There's been a big study done by the very best people who do this kind of work, and they found that it worked.

The effect was pretty big.