i fixed my anxiety with NLP, now i see every failure for what it is, a step closer to getting the campaign profitable (or one more way not to run it).
you need a thick skin to run IM like with acting, stocks, modelling most of the time you're in the failboat. brutal truth is nearly all campaigns fail at first (sometimes many times) before they turn the corner. worst thing you can do is pull out early, unless of course the results were that bad (a la 93000 clicks boy) that there really is no hope. spend, analyze, spend more, repeat until profit. its not rocket science. pull out early and not learn anything, then you're fucked and out of pocket.
hope you sort it.