> Also, when you get down to brass tacks, if Bob has zero engagement with the product, Bob will not pay for the product, and thus the downside risk of annoying Bob is to a first approximation zero.
Yet Bob might make a blog post and get significant exposure on a very well known technology site where many industry leaders participate - where people like me might say: "hmmm do I really want to deal with FooCorp spam when I can use similar competing services which do not seem to annoy people with spam"
There is always risk to annoying customers. You just never know how it will come back and bite you.
The counterargument to this is that FooCorp is probably not optimizing its sales funnel to catch the kind of people that read technology sites. Statistically speaking that is a marked minority.
You make a good point that you don't want to annoy user so much that they rise up against you with torches and pitchforks, but I believe patio11's point is that the indignance of a few HNers does not compare to increasing engagement/conversion of trial users by e.g. 0.5%.
Yet Bob might make a blog post and get significant exposure on a very well known technology site where many industry leaders participate - where people like me might say: "hmmm do I really want to deal with FooCorp spam when I can use similar competing services which do not seem to annoy people with spam"
There is always risk to annoying customers. You just never know how it will come back and bite you.