It really depends.
For example if you are running a campaign with a $15/payout and you are getting $.20 clicks you need to get one conversion per 75 clicks to break even. I would run it atleast until you get to that point however, I'd strongly suggest running it to however many clicks equals 2 or 3 conversions.
If it isn't converting after X amount of clicks stop it for the day. MAYBE make some changes and then I would try the offer atleast 2 additional days (including a mix between weekends and weekdays) also spread it evenly throughout the day. Start it in the morning run it for a little while and then stop it, start it back up at lunch run it for a little while and stop it, then start it back up a little later and see how it goes. As long as you are getting useful data it's all good.
That advice is assuming you have a tight budget and want to test the niche/offer while keeping your potential loss as low as possible. I'd recommend probably spending 6-9x the payout testing. (more if its a zip submit, maybe less if its a $50/offer all though neither are probably the wisest offers to start out promoting as someone new.)
Don't be afraid to use some coupons too, you can easily find $100 coupons for YSM that should help out a lot. I know when I first started I probably used 10 of them and only got called on it once. I just said my brother made some accounts as well and I was making new accounts to test out different things and figure out how everything worked. No problems.