It's an interesting post but I don't think the statistics actually support the conclusion. The 1st blog post referred 1'500 people -- we don't know how many actually visited the blog post. The second post had 3'000 visitors. Without knowing the bounce rate from the first post we can't compare these stats.
UPDATE: The other point to consider is that this is purely post-hoc reasoning. I can think of at least one alternative explanation -- the people who arrived from the first blog post were already motivated enough to click another link so they were more likely to purchase. Without doing some kind of experiment to differentiate these hypotheses we can't make a judgement one way or the other.
It probably sounds like I'm raining on the author's parade. That's not my intention. I really like that he's put these figures online and is trying to analyse. I hope I'm just showing that the issues are more subtle than they seem at first.
Hey Noel thanks for the feedback! For your first point about the blog post both of the stats that I referred to when it comes to visitors 1500 vs 3000 are both the number of people that were referred to the site from each blog post. So I wasn't talking about the traffic stats for the blog posts themselves, I was only talking about stats that I know for both of them which was how many times they referred visitors to DomainPolish.
I think that your second point is interesting. What do you mean by more motivated to click? I can tell you that the bounce rate on the home page was a fairly consistent 30% for both posts, which leads me to believe that people just trusted the site more when they read a review from someone else and felt more comfortable buying.
I think part of the winning formula in this case was that it's relevant to HN - (HN readers are the demo for DomainPolish). I wrote the article because I thought it was a cool service, I never expected anyone to read it.
I wrote the whole post in 15 minutes, it was a stream of thought that turned out to be relavent to the community.
Haha although I never explicitly planned for that to happen (definitely not smart enough for that) it seems to have worked out nicely. There will be more soon. Stay tuned :)
Social proof is very important. Even subtle aspects can have significant impacts. I think social proof ties in with another area that I think may even be more important: the impression of longevity.
With both my own and others' projects, I've noticed that if you don't give an impression that you've been around a while or that your project/product is in active use or production, you'll find it significantly harder to gain traction. I think this is why the "trick" of using screenshots, videos, and so forth, tends to work well, since people get to see something that, supposedly, is in action, rather than a floaty promise.
If you're launching a podcast, an e-mail newsletter, or something where a sign of activity will boost signups, make sure you get a "few in the bag" and can do your public launch at episode 5, issue 7, or whatever. Likewise, with a business like that described in the post, make sure you can get some early reviews, some testimonials, and some blog posts in the bag so that you look like you're going places, even when you aren't.
I think that this is a great point. You want your project to feel like its been around forever and will continue to be around for a long time. That inspires confidence in your users to buy from you and/or recommend you to their friends. Without longevity they can't be confident enough in you to invest themselves.
Simply puting some images of human faces on your site near the call-to-action button helps. The Facebook widgets that show who has "liked" your site work great for this. Our eyes are drawn to pictures of faces and a group of faces next to a button says "these people tried it, what is wrong with you?"
Some more:
* Pictures of faces
* A testimonial (one face and one sentance)
* A number indicating usage (4,243 downloads today!)
* Expert reviews (TechCrunch loved it!)
It is social proof IMHO. A social proof of one, but a social proof nonetheless - social proof coming from a fellow HN user might add some authority to the mix:
"social proof: People are more willing to take a recommended action if they see evidence that many others, especially similar others, are taking it."
UPDATE: The other point to consider is that this is purely post-hoc reasoning. I can think of at least one alternative explanation -- the people who arrived from the first blog post were already motivated enough to click another link so they were more likely to purchase. Without doing some kind of experiment to differentiate these hypotheses we can't make a judgement one way or the other.
It probably sounds like I'm raining on the author's parade. That's not my intention. I really like that he's put these figures online and is trying to analyse. I hope I'm just showing that the issues are more subtle than they seem at first.