Hunch’s says it “helps you make decisions and gets smarter the more you use it.” Sounds good to me, let’s dive in:
Request Invitation/Login Page: Good. Tells you what Hunch does big and bold. Obviously when Hunch is out of beta this page would go away because everyone would be arriving via Google search.
Create Account: 5 fields, 6 clicks. Excellent copy writing. The “and it won’t come too often” led me to actually check the newsletter box, and I never do that. One of the better signup forms I’ve seen.
Homepage (zoom): Needs to pick one function and make it paramount. Hunch should ask itself “what is the one thing we want a user to do at this point?” I’d kill the right column and increase the oomph of the list of “Topics”, that’s the most important part of the page. Hunch is supposed to learn so let’s assume the topics shown will become more interesting over time instead of feeling random. Is “topics” the right word, or would “questions” be better?
UX Hero trick: When designing a web page, step back five feet and squint. Is the page hierarchy clear? What jumps out at you? What gets lost? While squinting the logo and “Hunch Blog” dominate because they’re red and they’re diagonal so my attention bounces between them. Everything else gets lost. Meanwhile the topics list barely registers.
Topic/Question page: Rubber, meet road. When out of beta Hunch will live or die based on how well inbound Google search traffic sticks to this page. It passes the squint test. The arrow breaks the perimeter of the box so it calls out the next step well. Not sure how important “Watch this topic” is to Hunch. Based on its placement, I’m guessing not very. I’d like it if the answers to the question looked more clickable.
Profile/social media page: First of all, what the hell is a banjo? Tooltips (shown above) on other icons tells you what each means, but not so for banjos. Obviously they’re Hunch’s “experience points”, but how do you earn banjos? What does earning banjos do for you? I had to hunt for meaning on the FAQ page, and even then it took two paragraphs to tell me “banjos are experience points, yay for banjos” and not much else.
The following/followers area is meh for now. Either participation reaches critical mass, or it fizzles. Most faces I clicked on had empty profiles: no following, no favorited topics, no contributions. This is the UGC site “hump”.
Too see a profile page with a lot of resolution, look at Jeff Atwood’s profile page on StackOverflow. Hunch should ask itself “when we have huge participation, how will the design make the data useful?”
Overall: Hunch isn’t very useful yet (which Hunch actually mentions on its About page, nice touch). Its utility will depend on how user participation and Hunch’s “learning algorithms” work out. The obvious question is “how does it make money”, but Hunch feels like a “reach scale first, ask questions later” play.