Everything about founder culture trains persistence. Almost nothing trains the opposite judgement, which is required just as often.
Because the same evidence supports both conclusions. Slow progress means it is working slowly or it is not working. Nothing in the data distinguishes those.
And because every story we celebrate is about someone who continued.
The founders who persisted and were right write books. The ones who persisted and were wrong are not available for comment.
The base rate is invisible, which makes persistence look better than it is.
Everyone who continued and was right tells you so. Nobody surveys the ones who continued and were wrong.
The longer something has run, the harder stopping becomes, and the years invested have no bearing on whether the next year is worth investing.
Knowing this does not remove the pull. It only makes it recognisable.
Deciding in advance what would constitute evidence of failure, while you are calm and before you are attached.
Written down, with a date. Not a target you will rationalise — a specific condition you agreed would mean stop.
Describe the situation to someone uninvolved as though it were somebody else's. Their reaction is usually immediate and usually correct.
The information was available. Proximity was preventing it.
Capital and attention released from something that is not working are the inputs to the next thing. A founder who stops at two years has eight years of runway that the one who stops at eight does not.
I do not have a clean method for this. Neither does anyone I have asked. The best available is a condition written down in advance and a person willing to hold you to it.
Read next
The long unglamorous middle →