I was reading some of my favorite blogs and sites on the advent of Web 2.0, and the characteristics that distinguish web 2.0 from 1.0. One characteristic of the Web 2.0 launches in 2005 is that most launched - and remain - in beta. You should never, ever launch a new Web site in beta. Period. All the functionality should work as designed and as intended, erorr free. As the saying goes, you only have one chance to make a good first impression. Your users are not guinee pigs, they are not your QA team, and they don't work for you. It seems sites are launching just to say "Hey, look what a great idea I have, and someday it might actually work!" I wish beta sites had a sign up list that read "Please contact me when the Web site is out of beta" because I'm not returning until it actually works.
I think there is some confusion on what the main purpose is for releasing an item into beta. Your primary business, your core functionality, your "Web site" taken as a whole should never be in a state of public beta. The purpose of a beta should never be to test functionality, but whether that functionality is desired and will be utilized by your visitors. If you want to launch a new feature or subset of features to your Web site, and you want to know if those features will be adopted by your visitors, then by all means release those features as beta. However, the underlying functionality underpinning those features should be ready for primetime and work flawlessly. Otherwise, how would you know if a poor adoption of a feature is due to lack of interest, or simply because the feature does not work well?
Take Amazon for example. When they rollout a new product line, e.g. apparel, jewelry, beauty, etc, they release those new
stores into beta. However, the underlying funtionality of the store, from browse and search, to adding an item to a cart or submitting a review, work flawlessly. What they are really testing is the level of interest in those stores and whether customes are willing to purchase from those store at a velocity that justifies them supporting and sustaining those stores. If so, they will eventually take those stores out of beta, and if not, they can shut them down. There is more to it than that of course - getting the right mixture of Amazon offering and third party offerings, etc, but those are internal metrics that don't affect user experience.
When a company launches a site in public beta, they are communicating one of two things (they never specify, leaving it to the user to discover on his own): either they are testing the overall concept - the main purpose of the site existing - will have relevance and resonate across the web, or that the underlying functionality is still in development. If it's the former, there is no need to designate the site as in beta, as all Web site launches are testing the validity of the concept, making the designation moot and pointless, and if it's the latter, they are acknowleding they released a half-baked, ill-performing web site. Unfortunately, many of the 2005 "2.0 Web sites" fall into the latter, and their business will suffer as a consequence.
