Quite possible that I'm overanalyzing it, but only A/B testing would say for sure. Either way, thanks for reading the post and taking the time to question wheth

Quite possible that I'm overanalyzing it, but only A/B testing would say for sure. Either way, thanks for reading the post and taking the time to question whether rearranging navigation options has an impact on conversions, which pages get the most traffic, etc. My view is based on my experience. When I first visited the site, I was ready to sign up. But then my eyes glanced at the navigation bar, and the blog was linked first. So I clicked on 'Blog' instead of signing up. As much as I was ready to sign up, I also find it very interesting to read startups' blogs, so I clicked through there and only really came back to sign up because I was planning to write a blog post about Buffer's UX. I would have bounced otherwise. Moreover, the blog's masthead + logo link to the subdomain's index (i.e. the blog itself), so while there's a prominent 'Sign Up' element on the blog (credit to the Buffer team on that), my initial attempt to go back to the home page (and perhaps sign up) was met with a return to the blog. I don't think most users have much patience, and it's quite possible that people visit, get distracted by the blog, and then bounce, especially when a first attempt at going back to bufferapp.com doesn't pan out as expected. Even if the order doesn't matter at all, and there's no impact from 'Blog' being first, my point is also about subdomains, as I sort of hinted at above. So when I say the current UI directs traffic away from the main site, I mean to say that if I click on 'Blog', almost all of my interaction options from there send me to somewhere else on the blog.bufferapp.com subdomain where my options are less about those on the main site (i.e. sign up, explore pricing, etc.), and more about reading posts, commenting, clicking through to the founders' Twitter pages, etc. For all I know, that interaction may somehow increase brand affinity and the likelihood of positive word of mouth for Buffer. But in the short run it seems more likely that it will distract from at least one metric: registrations.