I expect some push back from some of my colleagues on this topic. I say, sometimes it is appropriate not to launch. If we do not have a reasonable understanding of the risk versus reward for the project, and there is some deficiency in the project (say missing some key testing) then launching is probably not a good thing. I can think, for example, of the costly recalls, and rework, and perhaps litigation and customers not happy wit our organization as fairly negative and harmful events should we launch when we perhaps should not.

I have at least one project manager friend or mine, that says the business is in business to make money, and not launching is a very bad thing indeed. It is better to get the product out then not to. We can never truly know the risk versus reward with a high degree of certainty anyway. Many times companies have gone through due diligence and found the product fails unexpectedly in the field.