[spiffytech]

A decision than increases metric X is not necessarily a good decision. Consider companies that jack up prices of gasoline and chainsaws after natural disasters. They make a boatload of money doing it, but they also accrue customer bad will.

I'm sure he gets a great sign-up rate from that highly-visible form, but if readers are irritated by it, what negative impact is it having for him?

[badclient]

False comparison. The biggest problem with your example is not lost customer good will but that you're breaking the law and risk getting fined/shut down.

No one says you don't have to be smart in picking the metrics you measure.

[studentrob]

what negative impact is it having for him?

That is a good question. Assuming there is some goal beyond signups (retention?), then if you cannot or are not measuring it, then that is the biggest problem.

[dpe82]

If his goal is sign-ups, why does it matter how irritated those who don't sign up may be?

[windsurfer]

Typically sign-ups are not an end goal, but rather a means to get money or involvement. How has Experts Exchange been doing on that front?

Focusing on metrics and simple minute behaviours while disregarding the big picture lead to evil user interfaces.