Nice summary, but be careful with FB ads. It can easily suck tons of money for literally nothing. Even if you pay per click, you get clicks from absolutely irrelevant people (bots?), with zero engagement.
My heuristic is that I budget 2x expected LTV per channel for the initial test. So if I think my customer could be worth $100 over their lifetime, I will allocate $200 to reddit ads, Facebook, etc. If I don't get any movement in the funnel (clicking signup; actually signing up; paying; etc...), then I abandon the channel until I have substantially updated the product or messaging. If there is some movement, then I devote effort to optimizing the funnel, starting from the ad creatives toward each segment of the signup flow
