When it comes to managing websites, I like to apply a fairly cut-throat approach to site optimization. Content and links which fail to perform are ruthlessly swept away in favor of those parts of the website which have performed or show potential. I like to think of this process of culling and promoting links as online natural selection or a Darwinian survival of the fittest. With today’s web analytics tools it is possible track and analyze practically every aspect of site usage, so revising a website based on behavorial data is often only a matter of taking the time to look at the data. A website subjected to this continual fine-tuning should evolve into a website which is more in line with customer needs, especially since the customers have essentially voted for the content and links which appear most prominently. I also have to admit that there is an inner sense of glee when sending non-performing links into extinction.

Unfortunately, this theory of website evolution has some strident opponents. It may sound like common sense or straight-forward logic to you and me (at least, it does to me), but there is a large and vocal community of people who do not subscribe to this line of thought at all. I am speaking about the advocates of intelligent website design, or course. These people can be found at levels of an organization, including highly experienced professionals in positions of significant decision-making capacity, as well as (gasp) e-marketing specialists in major marketing and advertising agencies. In fact, I’ve never been on a job or project which hasn’t featured a prominent respresentation of this faction. You will know when these people are in your midst, when your web analytics data is summarily dismissed and replaced by piles of indisputable market research, white papers, focus group results, and branding guidelines explaining exactly what should go where on the website. In my experience, arguing with these people usually proves an exercise in futility. Unless you are an experienced evangelist, your best chance lies with finding like-minded managers and building a base of support for your analytical approach. Or, you could try the good old “act now and ask for forgiveness later” ploy.

The point is this battle is too important to lose. Website real estate, despite widely-held beliefs to the contrary, is as precious as offline real estate. Visitors to websites are not constrained by physical distance to the competition and can leave your virtual real estate with a single mouse click. A website must engage the visitor within the first seconds and then work to help the visitor find what they seek. I don’t mean to dismiss strategic thinking out of hand. A compelling vision and well-conceived strategy are essential pre-requisites for success and should be based on solid market research, however, site optimization should not be driven by white papers and marketing egos. At that point, its best to put aside any delusions of intelligent website design and start competing on analytics. A lot of companies are already running their business on web analytics, and some are probably even using some form of natural selection when developing websites. Google has built an advertising empire based on a similar process for its AdWords. Are we soon to see websites which are programmed to evolve by themselves?