The end of 4e's advice on skill challenges is largely this -- it's always 3 failures and out, but the number of successes was reasonably varied. I usually play fast and loose with successes, getting, as you do, a feel by way of number of obstacles. I tend to not determine the obstacles ahead of time, or, if I do, it's a short bullet list of ideas that I can mold into the fiction rather than concretely prepped challenges. That way, I can maximally let the players drive the challenge, both in approach and in outcomes, by presenting challenges that follow the fiction more tightly. I know you like location based design, so this isn't really as useful there (and, to be honest, I don't often use skill challenges in my location design, something I might want to think on). I tend to use them for things like negotiations, or intrigues, or travel. Honestly, I think skill challenges for travel, especially open ended ones, are the bomb and should be presented as at least an official option way to do travel.