And, pretty routinely, every time I post a link to my economic social news site, http://newsley.com. I get around 70 clicks. Frankly, the people that come to Newsley from HN give me the best feedback and are my best users. In fact, a large portion of my first 1000 visitors came from HN. I get a lot of bounces, but i also pick up a couple of regular users every time. Granted, the demographic that I'm looking for actually intersects with HN's demographic somewhat.
I can also vouch for leaving links in the body of comments. As long as your link is on topic to the blog post, I really don't have any problem at all dropping a link in a blog comment. I think it can often add to the discussion.
If you do the math, that is approaching $0/hour. It is simply not a good way to acquire customers, nor should it be expected to be.
What I was saying is that leaving commnents on blogs helps you get the initial traction, both in terms of exposure and press coverage. That's it. Once you are past that stage, I agree it's an inefficient use of your time to spend all day combing blogs looking for relevant posts to comment on!
It's important to take a step back and look objectively at all feedback, especially if it comes from a demographic that is not your core customer.
You comment on blogs to engage the existing early adopters of the niche you are targeting to start the process of customer development.
You hit up friends to practice pitching (don't underestimate this) and getting some up front feedback on the cheap to help refine your idea.
That surely cannot hurt, but the return on investment of posting comments on a site is minute. I mean, I am addicted to commenting on HN and other sites daily and it is a lot of fun, but the notion that this is actually a good use of time for my startup is wishful thinking.
I would never expect to get a single customer from this. I think getting an article on TechCrunch is of debatable value... but the traffic you'd get from a comment underneath an article from people checking out your profile link is miniscule.
Regarding telling your friends -- again, it doesn't hurt, but it is thinking really small and is not sustainable. That might be a good strategy for winning a minor student council election, but the ROI is going to be utterly negligible. Basically, you could spend a week phoning everyone you know and convince them to check out your site. If you're lucky, this would roughly equal the amount of traffic you would get by having a moderately influential person spend 30 seconds writing a tweet about you.