One approach to these problems is to blame it on the search engine, buy or build a new one, and start again from scratch. While this approach can feel like progress, it avoids two major underlying issues that may doom it to failure.
The first issue is that search is dependent on people looking for what they want in a sensible manner. Search teams deal with this by trying to promote search tools and then trying to ensure that metadata structures and search weighting match user behaviours. However in certain process- and procedure-driven environments, we can predict the information that people will need – e.g. contact centres, back office processing. Rather than rely on users finding things, we can proactively serve it to them.
The second issue concerns the underlying quality of the content. most enterprise content repositories are like teenager's bedrooms. They are not kept as tidy as they should be. Things go on inside them that shouldn't. And their users can be thoughtless and distracted. Implementing enterprise search is like turning on a light in such an environment. All your illusions of order are swept away by the blunt, ugly reality that sits in front of you. However turning on a light is not enough to clean up the room - or keep it clean. If your underlying content is poorly written and poorly managed then enterprise search simply allows your people to find bad stuff quicker. Unless findability strategies are linked to tools and techniques that maintain the quality of content, they will fail.
How do you achieve these goals? Well, you'll just have to come to IKO 2016 to find out!