What is Social Search? Is it simply having the ability to tag a result you found helpful and being able to share that with other users? Or is it more than that? Having the ability to tag results you like so it boosts the relevance for the next person enriches the search experience, but I would expect this capability to be available out-of-the-box for any platform that deals with information access. While being able to tag a useful result is, well, useful, there is a lot more to the story when you use the term social in combination with search.

When I think of social search, I think of the ability to pull content together from multiple social media platforms and combine it with internal content to provide my users with a full view of the information landscape.

"Social" Content

Nearly every blog, news page, and search results page (whether Google or Twitter) is RSS enabled. Having to build complex crawlers to go out and parse website pages is becoming a thing of the past. It's an expensive process that is prone to problems.

Using RSS feeds to augment internal content lets you combine your in-house data with external sources to provide your users or site visitors a complete answer when they ask the question "Why?"

If your users need to answer questions like "Why is that competitor having a string of wins against us in the field?" they would typically search within your company and get back documents, perhaps related to competitive overviews or wiki pages. However, shouldn't they also get organized results pertaining to recent news, or blog mentions, or what is currently being said about the topic or company on Twitter?

All of this information can easily be procured via RSS.

Let's look at a search for Attivio.

What does Attivio say?

Starting on the Attivio website, you see RSS-enabled feeds:

• Blog

• Newsroom

• Events

That's a good start.

What's being said on the internet?

Bing and Twitter searches are all RSS-enabled too. Adding an RSS feed URL to your platform should be as easy as copying and pasting into a field on a webpage.

While setting it up is not exactly straightforward, Google search results can be subscribed to. Here is a site that explains how to do it.

So now that we've assembled some great external sources of information and made them available, your search platform should be able to seamlessly add new content from these feeds and have them available for your users in near real-time.

What are we saying internally?

Every search platform on the market can handle the unstructured content found in the files scattered across your servers, but can they also:

• Connect to your CRM system?

• Pull in emails (and their attachments)?

• Integrate with your corporate wiki?

Well, they should because there are valuable assets in all of those areas as well.

Pulling it all together

Now that all of this content is available, the platform needs to be able to present results to users' questions in a meaningful manner.

Having a system that can dynamically generate facets based on the contents of your index OR at query time is a powerful feature. Dynamic facet recommendation alleviates the need for maintaining an ever-growing taxonomy and allows you to on-board new sources without having to manually map the content to particular categories.

Here is where tagging a result is truly valuable – being able to rate, tag, or vote on a result and have the platform boost or lower the relevance of that result for the next person adds a lot to the end-user experience. What's important is that the platform supports this tagging in real time. Having a platform that supports real-time updates to an index, without the need for re-indexing your existing content is an absolute requirement for social search.

True collaborative search is found in a platform that enables more than just tagging a result, though. You should be able to save an alert as a query, tell the application to update an account in your CRM system with a link to a particularly interesting result, on-board new sources without involving a system administrator and most important, find the answer you're looking for.

My idea of meaningful social search includes results from external sources & internal sources, organized in an easy-to-use interface and, yes, with the ability to tag content I like or dislike with a rating or a comment.

TrackBack URI for this entry

Comments (2)