In a recent AIIM survey, Over 70% respondents said that they find it easier to find information online than content on their company’s intranet. Many of us at some time or the other have wondered why our intranet search cannot work like the popular internet search engines.
The answer is simple and complex.
The simple answer is no company has a multi-billion dollar server farm to enable search on their intranet; making internet and intranet search comparisons unfair.
The complex answers lie in the fundamental differences between what and why people search on the intranet and internet. Search experts differentiate searches as discovery search and retrieval search- in lay man terms we search to gain knowledge about a subject or to find a specific object. Most internet search is discovery or knowledge search and most intranet searches are retrieval or object search.If I want to know more about a product, I am more likely to use a discovery method to find information sources pertaining to the subject of interest. Read the articles, listen to the podcasts and view the videos to gain the required knowledge. My expectations of finding the information quickly are very low, I am relatively agnostic to the information source, and I am ready to invest effort to collect information snippets and then string them together to build my knowledge base.
But when I am searching for a specific document it is because this search is part of my larger task and the delays in finding the specific document will lead to delivery delays, so my expectations for accuracy are absolute and I want to find the particular document instantly and not hidden in the 3rd page of the search result.
Apart from the fundamental usage difference there are definite technical differences emanating from the larger number of data types that intranet searches need to tackle, the federation of information sources, the lack of vested interest of authors to manually embed rich metadata with the content.
Traditional enterprise search engines relied heavily on metadata to index documents; and accuracy of the search depended on the performance of the crawler to extract meta-tags from content- file name, author, date, information source. Of late there has been increased adoption of content analytics to enable semantic and faceted search which has had a significant impact on the accuracy of search results. And accuracy of search improves dramatically with powerful content analytics technology.
Next week I will continue this discussion to talk about how analytics improves search.