Amar Bhide in 1994 wrote an awesome textbook called the Origin and Evolution of new businesses, in which he took the INC 500, removed the outliers, and systematically interviewed and surveyed everyone who was left. One of his findings was that long-form, formal business plans were irrelevant to success.
I was curious, so I did a google scholar search and found this paper which has a good review of the literature. It cites several studies from different lines of research, and they’re all over the map. Some show a positive effect, some show a negative effect, and some show no effect.
This seems to support the null hypothesis. If there was an effect either way, you’d expect more studies to increase certainty one way or the other. The fact that there is no clear trend one way or the other seems to support the idea that any positive or negative effects are just random noise.
I was about 60% sure that business plans had no effect, I would willing to change that certainty to 58% now (it’s possible that there’s a small large effect that we simply can’t control for due to the nature of natural experiments)
However, one idea that did change for me after reading that review is that there may be a class of companies for which business planning is very useful – those companies in industries in which change is very slow. You might even be able to map out on a Wardley Map how useful business planning is to a company. If it’s all the way on the left of the map, long form written planning is less useful, and informal plans are more useful, whereas as it moves towards the right of the map, the reverse becomes true.