Hacker News Reader

As an opposing datapoint, I actually had an unexpected A/B result when testing the recent S3stat redesign. Trial signups for the new version were better, but only by a percent or two. Not really enough to bother switching.

But I also tracked all the way through to purchases (which typically happen after 30 days of use), and I'm glad I did. After two months, the people seeing the new version were converting twice as well as the ones seeing the old version. As in, double the sales. As in, wow.

So I guess it's entirely possible that users do in fact respond to pretty design. The only thing one can do is test and see.

Having a "plain" website can be an advantage depending on your audience. I can say I trust places such as http://prgmr.com/ and http://tarsnap.com/ more than I would trust some "Delightful Xen Servers" or "Delicious Backups" startup-site.

prgmr.com is an excellently designed website. It is also a plain website. patio11 is talking about sites that are designed poorly, not plainly.

Or, for that matter, google.com, which launched a $340B business on a website that could've been made by a 12-year-old.

Ironically, Google actually has some really talented designers now, but the minimalist look has become part of the Google brand, and so all the designers now are constrained by user expectations of what Google should look like.

Funny, that doesn't match my, or, (apparently) many other HN user's experiences as we've watched Google's design march from "highly usable" to "the kind of thing Microsoft does"

I think that what you're seeing is the effect of design coherency on a product. You are generally far better off having one barely-trained designer put together a product than a dozen best-of-class designers. Why? Because a product designed by one inexperienced person will at least be consistent. They have their overall vision of what they want to achieve, and they can make trade-offs appropriately.

When you have dozens of designers that all want to leave their mark on the product, you end up with...dozens of features, none of which fit together quite right. Google is in that state right now. Microsoft is in a far worse state, where you have hundreds of PMs that each want to display their button as prominently as possible. This is what people make fun of as "design by committee".

Until 2010, Marissa was the sole gatekeeper of what could go onto the search results page. The result was that it was at least consistent, if minimalist and relatively unchanging. The other result was that she pissed off a whole bunch of other people by saying "no", brought Search to a virtual logjam, and made numerous talented people quit. That's the price of design consistency and a highly usable product for users; everyone who's not calling the shots ends up pretty disenchanted, and usually finds greener pastures. Steve Jobs played an analogous role at Apple Computer - it worked great while he was alive, but now there is a design vacuum that you're starting to see in their later products.

BTW, if you want to see what some of Google's design talent can do when freed from committees, check out the Doodles. In general there is just one artist responsible for each individual doodle, and it shows. (Some of the interactive ones have multiple people involved, but even then it's a small team with a single creative lead.) And the doodles generally have a reputation as being one of the most delightful parts of the site.

>>> When you have dozens of designers that all want to leave their mark on the product, you end up with...dozens of features, none of which fit together quite right. Google is in that state right now.

I think it is quite the opposite now. Google designers are TOLD to uphold similar design pattern on all Google products, which have the same features across all of them.

Seriously, why does gmail, google+, and google search should have a similar look? It just breaks some functionalities. These products are different and should not look alike. Pushing same patterns into users mouths, when it hinders the experience is an overkill.

Once I heard Marissa Mayer recount that she asked Sergey Brin where he got the idea of doing such a minimalist design for Google. Sergey's answer was: I don't do HTML.

There's similar quotes in an interview with Larry and Sergey in "Designing Interactions" by Bill Moggridge. Sergey explains (in a lot more detail than I do here) that they kept the front page simple because they didn't want to spend too much time on the front page. Only afterwards did they realise how powerful that was, and they decided to keep things off there as much as possible.

>Or, for that matter, google.com, which launched a $340B business on a website that could've been made by a 12-year-old.

... with an advanced understanding of graph theory, a research paper outlining a novel searching and ranking algorithm, a distributed cluster of servers crawling and indexing tens of millions of sites, and an inverted database of every keyword that appears on all of them. But yeah.

Pretty sure they just meant the web design. Not the application.

  > It has not been my experience that cutting-edge web design 
  > has made huge differences to sales in the sort of markets
  > I tend to operate in

With all due respect, but BCC never had "cutting-edge" web design, it was always (at least in the incarnations I saw) barely adequate design-wise. http://panic.com or http://tapbots.com or http://www.kaleidoscopeapp.com is what I'd call well designed. What's interesting they all also look like template design, but they have the polished look BSS is lacking.

Of course, maybe you have something other than BCC in mind, or you did try the design on par with the examples above and it did not convert well, then disregard this comment.

Exactly right, but also exactly wrong. patio11 made an understatement to get his point across - that design doesn't matter as much as people think. That you, respectfully, call Bingo Card Creator not even close to cutting edge just makes his argument stronger.

  > patio11 made an understatement to get his point across -
  > that design doesn't matter as much as people think

That may very well be true (and I think it is true in a sense). But I can also argue that without testing with truly polished design you only know that one "not-quite-there-yet" design has no advantage over other of the same kind.

Sorry, but the three designs you posted may have been well designed 10 years ago but now they just look old.

Then do you have counterexamples for sites you would say have a modern look?

Warning: Incoming flat design.

I personally think they look like trendy hipster design. Not much text explaining anything, just a load of icons I am supposed to understand without any context.

Seems to appeal to the iPhone generation.

I was thinking less BCC and more the consulting career when I thought that. Not going to point fingers, obviously, but you'd be surprised how little a $X0,000 redesign (from meh to almost-wow) did relative to e.g. changing the H1 on the front page.