Adaptive Websites: The Next Wave?

First of all, I apologize for such a long layoff. The Master’s Project ended up eating all of my time for the past few months.

Anyway, what are these adaptive websites you speak of? A lot of people are using web statistics to change and adapt their websites. Google statistics has become virtually ubiqitous in today’s information superhighway. My Master’s project is on machine learning and my independent study is about using statistics to help Flash games adapt to players. These got me thinking about using statistics not just as a redesign tool for designers or developers, but to have the website itself adapt to the users.

We’ve seen some baby steps in this direction. For example, Amazon tends to track what users look at and suggest other products whenever shoppers return to the popular online store. Some websites change their featured content based on popularity. Tons of websites allow users to customize the content that they are fed. However, what if we made this automated and made wholesale changes geared toward the individual user while all the spinning cogs are kept backstage?

Let’s use blogs as an example. We could use tags and categories and combine that with time data associated with a page to figure out what articles a user enjoys reading. We could then display those types of articles at the top whenever that user visits the site. Imagine, if you went to a grocery store and every item on your list was already stacked in a cart ready for you to check out. That’s essentially what is happening here, except the system is also creating your list for you.

Yay, content. That’s great. Wait. Don’t some websites already do something like this? The answer is yes and the usefulness for it is still there. However, it’s not truly innovative to just have adaptable content. It can be used in innovative ways for domains for which it is not currently used, but the idea itself is nothing new. However, what if presentation of the content was adaptable? Let’s not confuse this with presentation that is customizable, such as modules found on such sites as iGoogle or NetVibes.

The dynamic presentation of content can actually enhance the user experience for individual users. For example, take a look at the ESPN home page. We’ll go with an extreme example just to illustrate the point. What if the website could hook into your webcam, use some computer vision algorithms to track your eyes and then enlarge the areas where you typically look? This would make the areas that you pay attention to more pronounced and start to hide the areas that are irrelevant to your interests. Now, I realize there are certainly privacy issues and such with webcams, but similar actions can be performed with readily available web statistics.

This approach may not really be tapping into any new data, but rather using that data for more than just a background analysis and input into the next step in the iterative process. This data, now becomes an on-the-fly tool in aiding the user in getting the most out of their online experience. And from my point of view, as a User Experience Designer, that’s what we are truly trying to reach. The policies of Universal Design are basically compromises in order to try and fit a design into as many users as possible. Yet, with a design that changes to fit each user then we’ve essentially reached the Nirvana of Universal Design.

So, what do you think? Will websites move away from being customizable to being more adaptable? Do you think we have to tools currently to make this effective? Leave your thoughts in the comments.