The trick is to make a list of all the ways it could fail and test these first. Have two personas, one that assumes it will fail and one that is enthusiast. Set

The trick is to make a list of all the ways it could fail and test these first. Have two personas, one that assumes it will fail and one that is enthusiast. Set the tasks priorities so the enthusiast one starts coding tasks that disproves the failure hypotheses of the skeptical one. If you can't get past one, don't switch to the "easier" stuff. Put the project in "hibernation" mode and switch to another one until you get another insight, usually from reading a totally unrelated thing, that will unblock you.