Structure vs Search: Curating Knowledge

As Facebook approaches IPO and Twitter becomes part of the general media landscape, corporate-sanctioned social media tools are slowly seeping into the workplace. Once you’ve made the cultural shift of getting people using tools like Yammer, it’s not much of a leap from being one of the cool kids to becoming another confused one.

People are starting to wonder how to blend the structured environment of a knowledge base with the more chaotic and time-sensitive social channel, or whether they must choose one over the other. Well, you can run both and frankly, you should. When email came along, many people were still using the telephone. When social media came along, many people were still emailing. And some are still using the telephone. The point being, it’s just another form of communication, which will become more common as the generations move through the workforce.

What does that mean for your knowledge assets?

The cogs in my mind have been turning for a while over how to manage the volume of dispersed knowledge that resides in enterprise social media formats. Back when I interviewed UserVoice’s Evan Hamilton, he predicted improved search algorithms playing a big part in the future of information management. When I read an article, recently, alluding to the death of structured knowledge, I almost had a panic attack. It’s true that many people will reach for search before navigating a hierarchical list of categories, but structured knowledge bases will never die. My happy place involves a little from column A and a little from column B—as this other article promotes. I can always find something on the interwebs that makes me feel better.

The problem lies in combining the column A and column B efficiently.

I’ve been doing some research on this and I had a conversation with Service Now’s Simon Morris. Service Now is a SaaS service desk tool with knowledge base and social add-ons. It’s on their Facebook wall, so hopefully he doesn’t mind me showing it here, too. Click on the Facebook wall for the context but this particular bit is what I’m talking about.

The fact that it can do what I want it to be able to do, with some jiggery, is encouraging. But there is no native “add to knowledge base” button in the social feed at this point, and this is what I want every enterprise social tool to be able to do.

Searching the stream for keywords and tags is useful but sometimes things flow through the stream that deserve a permanent home as part of a “chapter” of information on a topic. Enterprise chatter is certainly permanent but we consume it with a transient kind of attention. When some piece of chatter comes across your radar that you want to collate and add to the KB so you (and others) can consume as part of a collection of information you have to, effectively, copy and paste. It’s manual. As knowledge becomes more socialized again (like in the olden days, but through online communities), we’ll all become curators—collecting relevant info from the social stream and adding it into the structured environment. We have to get smarter about blending it in more effortlessly, so it doesn’t just confuse people and put them off.

In the meantime, suck it up and copy/paste so you can make the best of both worlds and I’ll keep prodding the tool vendors to make it easier. 😉


  • Hi Aprill, So – another feature we need for enterprise social feeds is the ability to curate a series of posts into a semi-coherent KB article.. right-click a Live Feed post and “Add to KB” so that we can convert conversations into a meaningful document in it’s own right.

    Or get users to use #hashtags which automatically curate content into a single KB so that it can be cleaned up and published.

    Great article

    • Yes, add this to the list of features in the pipeline. :) 

      I’m not so sure about automagically adding tagged content to a single article. That would just become a messy blob of text. It would be better to use the human intelligence to choose to click a button at the front of the task, than to clean it all up at the end. But then, would people remember to click it at all? Best to just have it automagically recognised as worthy, if someone had gone to the effort of tagging the comment in the first place? Hmm, perhaps best as a decision for the super-user to base around the culture of their environment. More stuff for me to think about, anyway.

      Thanks for keeping the conversation going.

  • Aprill, this is an excellent analysis of the knowledge management dilemma.  I thoroughly agree on the need for tools to be able to do more in the generation and production of knowledge content. 

    Helpdesk, service desk and other corporate databases already have all of the data that’s required for this.  It just needs the ability to collate, collect and summarize the information into a usable knowledge base article.

    Advances in text parsing algorithms, search algorithms and some good ol’ fashioned artificial intelligence and computer programming will allow software to produce this in the years to come, if not now.  (can you tell I’m a software developer?)

    Your point about all of us become knowledge curators is spot-on.  Bulls-eye!

    Good stuff.

  • Great post Aprill, I encourage everyone to read it!


Post a Comment: