Smarter Content Delivery for Smarter KM
I recently spoke with Simon Trussler, director of Iknow LLC, to discuss the continuing challenges that many organizations face with findability and usability of enterprise knowledge. Iknow is sponsoring APQC’s 2019 Knowledge Management Conference, and Simon will be speaking as part of our High-Tech Solutions to Knowledge Problems track.
Most organizations have built enterprise content repositories and installed search engines. Why is it still such a problem for employees to find what they need?
It’s partly because content volume keeps growing so fast. It’s become so easy to create documents and add them to repositories that it threatens to overwhelm the organization’s ability to digest and use it. In most organizations that we talk to, users complain that enterprise search gives irrelevant results, content managers are overwhelmed, and executives are increasingly aware of the opportunity cost of the knowledge that’s hidden in their reams of unstructured data. At the same time, expectations are only going to grow—employees interact daily with Google search, chatbots, and voice apps in the world outside of work. In comparison, enterprise search feels like it’s falling behind. People are spending too much time looking for documents.
The “digital workplace” is supposed to address this, isn’t it?
Yes, the digital workplace concept promises users the capability to see only the most useful content for their daily work tasks, whether they browse, search for it, or have it delivered via push technologies. However, many organizations are struggling to deliver on this. They lack the fine-grained tagging of content required to deliver exactly what each user needs, and they know that they need to make the output more usable and navigable.
So what are the key steps to improve this situation?
We’ve talked at previous APQC sessions about the critical importance of taxonomy, classification, and curation of content – these are really the scaffolding you need to deliver on the promise of findability and usability. Fortunately, the tools for these keep getting better, and the break-even point to consider auto-classification of content seems to be getting lower in terms of content volume. Without accurate tagging of content, search engines struggle to return meaningful results, partly because there’s so much ambiguity in how we refer to things. Auto-classification is really the only practical way to handle this for repositories with more than, say, 50,000 documents. But even in smaller organizations, you can benefit by freeing up manual tagging time to focus more on curation, highlighting the best content for each topic. Once you have accurate content tagging, findability immediately improves because of features like tag-based relevance tuning and taxonomy-based search suggestions. But it also creates big opportunities for improving usability of knowledge.
It helps in several ways. One of our favorites is developing topic knowledge pages that are populated dynamically based on taxonomy terms. SharePoint’s managed navigation and term-driven pages functionality makes this relatively easy to set up, once you have the taxonomy and tagging backbone in place. There is still a lot of usability thinking to do, in designing the pages so that content is presented in the most readable and useful way, and deciding what additional filters and in-page navigation might be needed to get people more quickly to what they need. But once the basic template is in place, the sites can be rolled out very quickly for any topics that you choose.
Beyond these, what are the next steps for improving usability and findability going forward?
There are several options, with either a broader or a narrower focus. In terms of breadth, well-developed taxonomies with many synonyms and cross-topic relationships can really help users discover important related content that they may not have thought to search for. In the public domain, Wikipedia is a prime example of this—leading you naturally to related concepts—but there’s no reason why we shouldn’t leverage this type of topic discovery in the enterprise sphere too. In the narrower direction, we're talking about “chunking” up content into more focused blocks that answer specific types of question; for example, dividing an operating procedure document into separately tagged sections focused on each of the major steps. That does mean a more structured content authoring process, and probably an investment of effort in preparing previous content, but it could save a lot of time and money going forward by helping people get directly to what they need.
For more information on this topic and insights on how to address these challenges, please join us at the APQC Conference in Houston from April 29 to May 3.Tweet