Generally speaking, all else being equal (and don’t tell anyone I said this) the release date of a particular version of my product is not that important to me. In the long run no one remembers if a release was late. What everyone remembers – the customers, technical support, sales engineers – is when a release was incomplete or buggy. That’s what will lose you a customer, and it’s how the product team loses the respect of their colleagues.
Therefore, as a product manager, I’m have to be more concerned about the customer’s experience than the company’s short-term desires, especially about release dates. The company might want a release to come out on a particular date, but if it will jeopardize an important customer’s experience, I can’t allow that to happen. And in the end, the customer is more important. Losing a customer is very costly, while shipping a few weeks late to make sure a product is actually ready is not very costly. It does impact the schedule of the next release perhaps, but less than the fire drill of the emergency fixing of bugs you actually shipped.
I’m interested to hear your thoughts on this – do you sacrifice schedule for quality, or have you managed to figure out how to drop features for time?