Proportion of Searchable SharePoint Assets, information management, symantic searchEven though the use of SharePoint in the enterprise continues to grow at a phenomenal rate, recent research sponsored by Smartlogic shows that it is still creating problems for many users. In particular, of those questioned, more than half say that finding the right information is a major headache.

The research was carried out earlier this year by MindMetre Research for Smartlogic across more than 500 information professionals in the US and Europe. While the survey group appears extensive, the research does not indicate what versions of SharePoint the respondents are using, nor does it mention when the surveys were carried out.

While the timing is probably not too important, it should have really looked at what versions the respondents were using as Microsoft would undoubtedly argue that many of its components have been extensively revised since SharePoint 2003.

Indeed, with SharePoint 2013 only really starting its life, the results might be a lot different if the survey was done again a year from now.

The research aimed to reflect the situation in enterprises now and assess levels of satisfaction with current deployments. For the sake of transparency, it is important to note that Smartlogic produces advanced search software for finding enterprise information.

SharePoint In The Enterprise

That said, the results are interesting. In a nutshell, it found that many enterprises are not happy with the search capabilities of SharePoint. This has the potential to become a serious problem for enterprises and, from a business perspective, it could cause problems for Microsoft too.

According to Microsoft’s own figures, 78% of Fortune 500 companies now use SharePoint in some capacity while 20,000 new users were added each month between 2006 and 2011. This means that at this point in time, 100 million users have access to it globally.

This is the starting point for the research. The report, SharePoint: Shows huge promise --Could do better, noted three principal reasons for investing in SharePoint:

  • Information Retrieval: It says the ability to find information and eliminate information duplication is the main reason for deploying SharePoint.
  • Effective collaboration: 60% of those surveyed said the ability to collaborate was one of its most crucial functions.
  • Compliance: 42% consider SharePoint key to meeting regulatory requirements and minimizing data breaches.

SharePoint Expectations

If SharePoint is being widely deployed across enterprises, the survey indicates that satisfaction levels with search performance is quite low. Many users said that it does not fully support the business it was deployed in, while many others said they had problems finding internal information.

While many senior directors in enterprises said it did not meet expectations, the proportion of information professionals that said the same was higher. When asked what kind of performance expectations they had for SharePoint, 67% replied that they would be disappointed if it did not achieve 90% of performance expectations.

SharePoint search satisfaction

In the majority of cases, this performance threshold dropped below the 90% mark. In fact, 80% said that it failed to meet expectations. This breaks down into 46% who said it fell a little short of expectations, and 34% who said it fell well below them.

SharePoint Perceptions

The research points out that SharePoint is primarily an enterprise collaboration platform and not a search tool. However, because it is deployed for many other functions than collaboration, expectations are not always going to be met. It also shows that many information professionals believe they are getting less out of SharePoint enterprise search functions than their peers.

Even worse, a quarter of SharePoint users believe that the SharePoint search function is offering them access to less than half the information contained in the enterprise. It also shows that most users think that information workers in other enterprises have better search capabilities.

The result is that 62% of information professionals working with SharePoint said that they were, at least to some extent, unhappy with SharePoint and the return-on-investment as a considerable amount of content remains difficult to find. It also suggests that the relevance of search results is not always guaranteed.

SharePoint ROI Disappointing

Is Semantic Search The Answer?

The problem with search is that for accurate results users need to be able to search in context. In a global enterprise with billions of pieces of information, a simple search query can result in millions of irrelevant answers containing the original search query.

Conversational search results based on metadata or labels are unable to provide context to the terms in the query. Semantic search solves this problem.

At this point, it is important to recall that Smartlogic provides semantic search technology. However, it is not the only product offering this capability, HP-Autonomy IDOL being another obvious vendor in this space.

That aside, it is clear that as SharePoint is being used to manage more and more information across an increasing number of silos, semantic search offers one solution to issues arising from growing information loads. Semantic search attempts to understand searchers’ intent by taking into account context and likely intended meaning so that irrelevant search responses can be filtered out at the very earliest stage of the search.

It does this by creating a vocabulary that brings together all the key terms that an enterprise might be using to do its business. It also creates a standard for information classification that enables users to place documents, or other information assets, in context.

This semantic model can drive a metadata classification process, adding a second level to "findability." It also solves the problem of inconsistent and inaccurate manually created metadata. By adding software that understand the semantic model, it is possible to scan documents, put them in context and accurately tag them. To ensure that this classification is an ongoing process, the system’s functionality needs to include taxonomy and ontology management, automatic classification and semantic processing.

While there is little doubt that semantic search will improve many systems, there is a question as to whether or not enterprises are prepared to invest more money in a technology to work with a system that has already required substantial investment. There is also the possibility that Microsoft may take heed and improve search in future versions of SharePoint as the ability to retrieve information becomes increasingly important.