By Cruce Saunders
Although Knowledge Management (KM) as a chartered enterprise practice has been around for 40+ years, its biggest value will be realized in the years ahead. The combined impact of engineering content sets, graph technology, and Content-as-a-Service architectures will change enterprise knowledge as we know it. The effectiveness of knowledge will remain impaired until operating models and methodology move beyond the traditional, siloed content workflows.
Today’s world of rapidly multiplying content assets, along with new channels for publishing, calls for new ways of thinking and working. The future of knowledge is contextual. Knowledge no longer works for consumers in document form; it’s moving toward modular forms and connected to a customer’s unique circumstances and contexts. Thus, the future is dependent on enriching critical knowledge assets into structured, semantic forms, then connecting those assets to Content-as-a-Service and artificial intelligence (AI).
Every day at [A], we see the shifts toward content intelligence inspiring major transformation efforts. Most organizations are moving toward new models for managing their owned spheres of knowledge. The changes in enterprise knowledge and content management are here and accelerating.
Knowledge Expresses an Organization’s Reason for Being
The KM function focuses on the “what” and “how” of a company’s products and services. Enterprise knowledge investments focus on an organization’s “reason for being”—the organization’s products, services, and value-add to the marketplace.
As a result, many company functions integrate with KM systems, including customer support, training, onboarding, documentation, and self-service. Depending on the organization, KM systems might intersect far beyond the traditional areas into marketing, public relations, strategy, market research, and thought leadership functions.
Ultimately, a KM System (KMS) is any type of information technology (IT) system that stores and retrieves knowledge in order to improve understanding, collaboration, and process alignment. Most traditional KMSs are the primary system that a call-center or service-center representative will use to interact with end-user customers.
Within a customer service team, for example, a KM solution provides specific benefits to different stakeholders.
- Service Agent: To find information via a single source, provide guided chats for easy and consistent service, and work with built-in e-learning
- Editor: To craft custom workflows, create custom-built approval processes, create guided scripts and chats, address complex topics with the ease of a Powerpoint flowchart
- Manager: To ensure consistent service from all agents, centrally manage scripts and guides, onboard new agents faster, and generate custom reports
Due to legacy technology limitations, however, many traditional KM solutions have been focused on centralized content and data objects stored in relational databases. Getting at the knowledge requires more-or-less regular, Boolean searching patterns and linear lists of results. Calls to retrieve the identified objects often happen at the document level, rendering many KM efforts little more than large-scale document management systems. As a result, even in the face of large KM investments, knowledge has remained both siloed and difficult to retrieve. This status quo demands change, and it is, in fact, now changing and evolving actively every day.
The Modular Future for Knowledge Expressed in Content
Compared to the evolving omnichannel and content technology landscape, the document-oriented and schema-siloed approach to KM provides decreasing utility. Instead of finding more and more value from capital allocations into KM, with no changes, organizations will continue to find less and less value in those investments.
At [A], we see KM with the development of modular, intelligent, shared content assets. These assets can then—through content-as-a-service—empower diverse internal and external learning and experiences. The modular content being pursued by marketing and customer experience teams is very much related to the modular knowledge assets being called upon by customer support teams. KM can, and should, be a key driver in the gradual enterprise movement toward content intelligence. KM is closely related to Content Services.
If KM is the collection of systems, tools, technology, and processes associated with managing an organization’s knowledge, a Content Services Organization is the collection of systems, tools, technology, and processes associated with managing an organization’s modular content assets. This may seem like doublespeak, but the distinction between knowledge and content is significant.
Knowledge precedes content. It is the raw material of content.
Content expresses knowledge. Content is derived from knowledge.
Therefore, when we cultivate knowledge assets, we ultimately give rise to smooth customer and employee experiences and improve the entire value-stream supporting customers and employees.
Trends Driving Content-Intelligence Transformation
The shift has already happened: The migration away from single-silo, KM applications is a foregone conclusion. Now organizations are rapidly progressing toward an AI-supported and extended organizational “knowledge sphere.” Shifting knowledge capital into structured, semantically rich forms—and connecting those assets to machine learning and other AI enhancements—is the path of the future.
At [A], we note the following three trends that should inform any progress on KM strategy in the near future:
- The rise of Intelligent Content: Enterprises are moving to component-based content for both KM systems and customer experience (CX), to fuel the assembled, contextual use of content.
- The rise of Semantic Services: Enterprises are adopting more modular and graph-based semantics that can be made available in service-provider relationships to multiple systems and applications across teams and platforms.
- The rise of Content-as-a-Service (CaaS): Enterprises are moving away from a siloed publishing model and toward the ability to deliver modular content assets via API to many downstream users and applications.
Content intelligence continues to grow across enterprises as new approaches to content get introduced, demonstrate value, and continue to expand adoption.
Content Intelligence Will Power Artificial Intelligence
Against the content intelligence backdrop, consider the adoption of artificial intelligence (AI), and the ways it is being employed to extend and amplify human knowledge.
Given the vast stores of information, it’s natural to seek automation. Vendors promise that content can be automatically indexed and combined across disparate platforms through natural language processing (NLP), entity recognition, and even created into new forms with natural language generation (NLG).
Any AI service that works against unstructured content, however, runs into significant error rates. Machines have a hard time disambiguating complex human communication. As any IBM Watson customer can attest, training machine models on unstructured content is a hugely human-capital-intensive endeavor with highly imperfect results.
The path to coherent AI lies through structure and semantics. Transforming content sets into structured, semantically coherent services has been sidelined as “too much work” for years, but that is changing in the face of AI consumers. It’s becoming clear that structuring content first takes no more actual effort than trying to train an AI on an unstructured corpus of content and achieves much better accuracy. One method, structuring, looks difficult at the beginning. The other method, training, looks impossible in retrospect. But the good news is that the mistakes have been made (to the tune of many billions of dollars). We can stop making the same mistakes now. Trying to train recurrent neural networks (RNNs) on unstructured content sets to generate reasonable human responses is ultimately a waste of time. The path to AI is the path of structure. Regardless of the approach, content takes hard work. Never underestimate knowledge. Never underestimate content.
Major barriers to the discovery and usage of organizational knowledge is eliminated when these same technologies are used against structured content sets with human supervision to tag and organize information using a growing semantic model, automatically associating contextual semantic metadata.
Eventually, AI will come to play a pivotal role in constructing and comprehending the insights within rich, ontologically-sound knowledge graphs that power everything. Knowledge management will get amplified in effectiveness by machine-assisted learning, accelerating enterprise knowledge value, and availability to human and machine consumers.
The long-held ambition of proactively delivering “the right information at the right time” to knowledge workers is well underway. As content intelligence and artificial intelligence combine, new world systems emerge that flow intelligence through networks of people and processes to uncover insights and solve problems in real-time.
Architecting for flow with Content-as-a-Service
All the different systems that content and knowledge has to move through currently require manual transport, transposition, and transformation. As systems and channels increase, the more effort and cost gets involved. All of this friction has been driving Content-as-a-Service initiatives across the enterprise. Sometimes this starts as interest in a headless content management system. Headless content implementations, and larger CaaS programs, in turn, have become a key driver in the evolution of shared team awareness around the needs to architect teams, systems, and processes in new ways.
The goal for everyone: Service-oriented exposure of content and knowledge assets to support an increasing, and perhaps an infinite, number of customer needs and channels across many enterprise systems and teams.
New disciplines support new realities
At [A], experience leads us to strongly believe that the creation of new practices around knowledge itself is what will make this new era of knowledge asset reuse plausible. The disciplines of structured content and engineering of content—combined with the development of standard, shared semantics around content and the building of new operating approaches—underpin an organization’s ability to embrace the new world with intelligence.
Information 4.0 as an analog to Industry 4.0 just makes sense. But these shifts cannot happen within companies with the archaic approach many enterprises still take to authoring and managing content and knowledge.
The future of CaaS and the optimization of KM requires new team roles and organizational structures. [A] advises the creation of a cross-functional Content Services Organization (CSO), incorporating three essential practices: Content Strategy, Content Engineering, and Content Operations. A Technical Communication article titled “The Enterprise Content Services Organization” provides more detail about the CSO, as does a whitepaper published on simplea.com. Suffice it to say, the knowledge management revolution depends on people and processes, organizational design and system design. New approaches and new architectures are necessary for capturing and managing structured related content sets across groups.
Key Ingredients for Success
The inevitability of change is certain, that much we have been taught by a year of drastic change in 2020. Knowledge makes things better – not just the existence of knowledge, but its availability, curation, and application in the right moments that matter.
The availability of collected, semantically ready content into systems via Content-as-a-Service endpoints is the inevitable destination for any organization seeking dynamic interoperability.
Knowledge Management is not just a class of software or a single department, but an enterprise-wide effect built from multiple causes, most of which deal with the shape, structure, semantics, and annotation of the content itself, as well as the operating model by which knowledge and content are managed every day. The instructions for transformation initiatives should come with the label, “Some restructuring required.”
Knowledge Management goals should be extended to include the long-term intelligence of content assets, building content and semantic models, orchestration and governance, content supply chain, and architectural innovation beyond short-term expediency.
Do not rely on any single integrator or a vendor-led implementation. The real solution is far more complex and ecological. Interdependent challenges can only be addressed with a systems mindset that prioritizes the intelligence of the knowledge and content assets over the expediency of their delivery.
Saunders, Cruce. “The Enterprise Content Services Organization.” Technical Communication 66, no 2 (May 2019).