Picking the Right Category of KM Software
Done right, KM can deliver tremendous value to customer service and support. Yet for all the benefits, the software category is marked by confusion—many very different solutions purport to be "KM." This article lays out the categories of knowledge software to help you pick the right one for you.
Knowledge is in people's heads, and it's hard to capture, even assuming they wanted to share. Knowledge is scattered in many places; it can be stale, and hard to find.
To address these challenges, KM tools need to provide one-stop shopping for answers, wherever they are. Capturing knowledge must be easy. And managers need analytics to continually improve the knowledgebase and team performance. Different solution categories take very different approaches to these requirements:
1. The straightforward solution: CRM KM modules. CRM vendors make modules that store knowledge articles in a database, provide authoring screens and implement workflow for approving articles.
An advantage is that the module is integrated with CRM. However, challenges come from the fact that databases aren't an especially good fit for the dynamic nature of KM. This results in odd limitations, like character-count limits, fixed templates and rigid workflows. Worse, because search is limited to the database table, a single search box can't reach all the relevant knowledge.
CRM modules are best for those with relatively basic requirements, who prize the simplicity and low cost of a pre-integrated module—they're typically deployed for low-complexity customer service.
2. The technical solution: Structured knowledgebases. When academics addressed the problem of using knowledge to answer questions, they focused on knowledge representation—how the information was stored. In the human brain, a question and the context in which it's asked triggers a rich wave of associations. Vendors worked to make knowledgebases act the same way using decision trees, inference engines and case-based reasoning (CBR) systems.
These systems store knowledge in a very particular way in order to retrieve it in a human-like way. To add knowledge, the user must also do "knowledge engineering," for example by updating a decision tree to guide users to a new document, or by tagging specific symptoms so CBR can match them to a query. This makes it much harder to author content.
Accordingly, these systems are best when there is a relatively small repository of fairly static content-auto repair, for example.
3. The ubiquitous solution: Content management. CM systems are available to nearly every knowledge worker. Because of a superficial resemblance between KM and CM, and because essentially every IT department has a CM solution, there's often pressure to use CM for knowledge. This doesn't work well, because the "similarities" don't hold up on closer inspection.
Service and support knowledge comes in small, actionable, structured chunks: answers to specific questions, or steps to resolve a particular problem. CM systems, by contrast, manage longer unstructured documents covering many topics. KM users have a question, but they don't know the answer they're looking for. CM users, on the other hand, are looking for a specific document.
Finally, for KM, measurement matters. It's important to report on knowledge activity, quality and outcomes. Even if you build analytics in CM, the measureable elements aren't all there. The great thing about using CM for KM is that it's possible to start now with a tool that's already in place. But sustaining a knowledge program requires significant and ongoing investment to fill the gaps.
4. The general purpose solution: Search engines. Most enterprises have search engines, too, and they're agnostic about how or where knowledge is stored. That's good news for people creating information, since they don't have to do it in any particular way, and also for people searching, as a single search box can return relevant content from any source.
IT organizations often suggest using a search engine for KM. But while search is the most salient part of KM, it's certainly not the only part. Search engines don't author or store knowledge, nor do they report on its creation. In fact, the most KM functionality lies outside search. Search engines can't deliver knowledge management by themselves; they need to be integrated with other solutions.
As with content management, search engines can be initially appealing because most organizations already have one. But they require significant development and ongoing investment in integration to deliver full KM.
5. The complete solution: Full-featured knowledge management. You probably saw this coming: a category that covers the shortcomings of the others while touting most of their advantages—the integration with incident tracking provided by CRM modules, the answer retrieval of structured knowledgebases, a content repository like CM systems, and access to content in any repository as provided by search engines... along with simple, structured authoring and reporting.
Users need answers, not long documents. So, solutions in this category also use knowledge representations. With specialized ontologies and automatic classification, they tag content without requiring human "knowledge engineering." This combines the benefits of search engines (automated context indexing) and structured knowledgebases (search that answers questions.) These tools have knowledge repositories, authoring, packaged CRM integrations; and analytics, supplying all the needed capability out of the box.
This comes at a cost: even the best pre-packaged CRM integration takes some effort to implement, and there is the initial investment and ongoing maintenance of ontologies. So full-featured KM becomes an ongoing project and a part of everyday business processes with training, measurement activities and maintenance.
As a result, these systems are most appropriate for larger service and support organizations with more complex KM needs and highly dynamic content.
When looking at full-featured KM applications, consider:
- Is the product going to be supported by its company for the long haul?
- Is it fast enough?
- Do the analytics provide insight with ad-hoc reporting and dashboards?
- And is it cost-effective to implement and operate?
No one solution is right for everyone. The categories outlined here should help you think about your approach to the market, and allow you to have productive conversations with IT.
Knova Knowledge Management from Consona is a full-featured KCS Verified solution, especially designed to handle even complex queries across channels. We work with the world's most demanding, high-volume service and support organizations, ensuring our products meet their high standards. Learn more about Consona and find a detailed whitepaper on this topic in the resource library at crm.consona.com.