3 ms·
Sure, but statistically, if each idea has a 10% chance of working, you're significantly more likely to have one good idea if you test 10 ideas (1 - 0.9^10 = 65%
by nostrademons 10y ago
Sure, but statistically, if each idea has a 10% chance of working, you're significantly more likely to have one good idea if you test 10 ideas (1 - 0.9^10 = 65%) than if you test one idea (10%). And even if you work really hard at that idea, so that its chance of success goes up to 20-30%, you still would be better with the 10 quick tests.
In practice, that usually means do the minimum you can to verify the key assumptions of each idea, back out and try something different if any key assumption doesn't hold, and if you find something that has a good chance of working, stick with it.