How important is site design?

I think MySpace is a great study. It's obviously terribly, horrifically designed... but it was at one time one of the most popular sites on the web. Not because it was pixel perfect, but because the social mechanics were designed very well.

Now look at Facebook. It's better designed from a visual and usability standpoint, and I think that gave them a little bit of an edge. But again, what really seems to have driven Facebook's growth is incredibly well designed social mechanics. Photo tagging is huge, for example. The News Feed is a huge engine for growth.

So, I think the answer is that it depends what you mean by "site design". A clean visual design, or even good usability may not be as important as a clean social design, or a clean API design.

As a designer you need to understand what are the key bottlenecks in meeting your goals, and design the heck out those.

Once you're Apple and you have a bajillion dollars, you can afford to polish the insides of your battery cases. But as a startup, you only get a few chances, and you have to spend your design effort where it really counts.

Design can mean different things to different people.

While you don't need to have a very flashy and attractive website, you do need a clean and usable user interface, and stuff like clean colors and typography.

It really depends right? Some users are looking for an experience, some users are looking for a service, and most users are somewhere in the middle. But depending on your product, your users may expect different things.

For example, if you have a site for coders, having a technical but not completely clean site would be okay as long as it functions (service works as advertised). Of course these same users would probably appreciate good site design more later down the road. On the other hand, if you're making a site like amazon for the typical internet user, an organized site with clear call to actions may be best (good user experience). But later if you change stuff it might freak users out if you're not careful.

Not sure if my input really helps explain anything other than that each site is probably different (I know, I know, I'm a genius...) but I guess what I'm saying is think about how the users are using the site and why, or even better, go out and meet them to ask them.

When you're looking at sites like reddit, hacker news and others which don't have a great/nicely looking user interface you would think it's not that important. But I think it also depends on the target group. For consumer-oriented, less internet savy people websites it is essential to have an appealing professional looking design, great usability, ease-of-use and optimised text.

So to answer your question I would release an early messier version to a closed group of beta users, ie. early adopters but before pushing onto the mass market I'd clean it up.

If you think your product has potential even with your misgivings about the site design, go ahead and launch. Release early and often, as they say.

As long as your design isn't dead-awful, such that it makes users feel your product is untrustworthy or poor, release now and iterate. You might want to consider getting feedback, either from posting it here as an Ask HN post, or by asking your friends/family/strangers. See what they think, then make an informed decision.

The perfect is the enemy of good. So is the bad.

It's hard to say. DrudgeReport and Craigslist seem to be getting on ok, but I personally would use okcupid.com over plentyoffish.com because the design is so kick-ass. I think if the functionality and usability of two sites is equal, a good design can give one the clear edge. We're visual creatures, and even a website can have increased perceived value with a shiny new coat of paint.