When something is new, we need to approach it in an exploratory manner. We need to experiment and try things out.
And so it has been with the Web. That period is now over. We need to move from seeing our Web sites as a series of projects to managing them as a well-planned process.
The quality of search on many Web sites is awful. Most organizations would simply not countenance having their reception area in the mess that their search engine is in. Yet far more people pass through their search engine every day than through their reception.
One of the reasons that organizations manage search so poorly is that they see search as a project. Someone decides a search engine is needed. A specification is prepared and software is bought and installed. Then, the team moves on to the next project.
In all likelihood, the search engine has not been properly optimized. It is certainly not being optimized on an ongoing basis, based on observed search behavior. Nor are people trained to write quality metadata when they publish content. Nor is the quality of this metadata monitored. So the quality of search gets worse and worse.
I've done workshops all over the world. Everywhere I go I ask people what the killer application for the company intranet is. “The staff directory,” comes the reply almost every time.
And what is the problem with the staff directory? “It's out of date!” is the chorus.
Again, getting a staff directory was treated as a project. The software was bought and installed. Nobody, however, thought about how it was going to be kept up to date. It was like the staff directory was seen as some marble statue that, once installed, would remain fixed forever.