For about a month, my son has been throwing incredibly powerful and intense temper tantrums. They’re totally unmanageable and you basically just have to ride out the day. Some days are good. Some days are bad. This summer, we had a particularly rough go of things during our two and a half week vacation, and during one of the weekends, my wife and I discussed whether or not we should just pack up and go home or refuse to bring our son back next time. My wife made a great point and suggested that since it was Friday, maybe we should wait until Monday to judge.

She was right. The weekend was not perfect but it was manageable. So the question becomes, “when do you judge whether or not something has failed?” It’s a really great question. Sometimes you jump to conclusions too quickly. You haven’t given the project or whatever it is enough time to actually get started and as such, it hasn’t matured and come to fruition. Sometimes you wait too long, you stay in and when the cards are finally dealt you find out that you’ve failed miserably and wasted time, money or both. Trying to figure out when to judge whether or not something was a success is incredibly difficult.

The truth is, I don’t have any idea how to do it. I truly don’t. I do have a few ideas though.

First, trust your instincts. If you think that things have gone south, there’s a very good chance that things have gone south. But also try to put things in perspective. Have things been worse? And in that case, did everything work out ok in the end? Did you survive that? Then there’s a pretty good chance that you’re going to survive this.

Second, ask. If you’re building a startup and you want to know if things are where they should be, find someone who has experience in that field and just ask. Most people that have built something incredible would love nothing more than to tell you how to build something incredible. And they’re also more than willing to tell you when you’ve done something that you shouldn’t have. In general, people don’t want other people to fail.

There’s a great saying that I really love. “When the horse dies, dismount.” That’s the idea. When something is no longer working, walk away. Not before it’s actually failed and not after. At the ACTUAL point where it all goes sideways. Find that point. That’s when you pack it in.