An inflammatory statement, I know, but let us look at the evidence that leads me to believe that it is a correct one.

The first and most obvious reason is the user’s ability to find stuff in their work place. In December 2013, SearchYourCloud completed a survey exploring this. All responses came from enterprise users, 200 from the USA and 200 from the U.K. The results were shocking. Not because I was unaware of what they would be, but because of how easy they were to predict. The same survey 4 years earlier yielded the same responses, “20 percent of an enterprise worker’s time is wasted looking for stuff.”

IDC (News - Alert) has done similar surveys and found that on average knowledge workers spend 8.8 hours a week searching. That is a staggering $14,209 out of an average income of $75,000. This means that on average, a user will waste an entire day each week at the office struggling to complete searches for information they need to carry out their work.

Those surveys alone should be enough proof for most people, but that is not the whole story. Enterprise search is a complicated and scientific affair. Well, that is what those vendors implementing solutions would have us believe, and, in their world, they are not incorrect. This complication means that the user has become secondary to the system, with little or no effort put into finding out exactly what it is a user requires from a search application. Instead, most suppliers have been falling over themselves to provide a “Google (News - Alert) Like” interface. All very well and good, and it has to be said that in the “Web world” Google does an amazing job searching. However, finding the information from a range of different sources so that you can carry out your work is a different type of application. This has led to large, costly implementations with a very small user uptake.

The main reasons for this come from the perceived wisdom many years ago that it would be a great idea to place all of your “stuff” in one great bin. A literal translation may be required. "All of your indexed data - your emails, your files, your CRM results and your doc management put into a large separate data silo." This index would then be used to search against. The problems that come from this set-up are huge. The main ones being :

• Users are completely ignored.
• The system was generally 24 hours behind the information, making search results outdated.
• These systems need large management cycles to make sure the indexed data was working.
• They require users to meticulously use only those information points that the system has access to, as adding a new search point is costly and takes time. This means that the user is not able to bring on new information sources or file stores—i..e. Cloud Stores.
• Lastly, and not insignificantly, they generally require the content to be tagged and classified. If this process is managed incorrectly, then the user will search and search, but find nothing.

This out of date and old-fashioned methodology cannot work in today’s connected world. Now we have far more information points than ever before. Most of these sources are being updated in real-time and users want to access that information in real-time too, not just when the index engine has caught up. It really is the Age of Federated Search. There are so many applications out there for consumers that do an amazing job at federating information: about traffic, travel, food and many, many more. Why has this not moved over into the enterprise?

My own personal view is that traditional vendors find it just too hard to offer such features. They have their core offerings and they will stay with them. Users—well they are there, but they do not sign the PO. This is a harsh view I know, but nobody has given me any alternative explanation as to why Enterprise search is so bad and does not seem to be improving in the mainstream, despite the cost implications on business and the stress caused to users.

About the Author:Simon Bain is the company founder, CEO and chief architect of SearchYourCloud’s software solutions.