Looking to the Future of Knowledge Management: 2020 Insight

Content management and content services

A foundational element to being able to get the right information to the right people at the right time is content man­agement, which is being enhanced with a new generation of content services. As a result, the market for content management is projected to growth at a brisk pace.

According to Fortune Business Insights, the global enterprise content management market size, which includes content work­flow, document management, imaging and capturing, web content management, record management, mobile content man­agement, digital asset management, and case management, was valued at $15.33 billion in 2018 and is projected to reach $43.16 billion by 2026 for a compound annual growth rate of 14%.

Content management facilitates the collection, management, and publishing of a broad array of information. But with a growing array of content such as images, documents, emails and more, and an increasingly diverse array of repositories, the term “content services” is becoming more popular.

There are a number of characteristics that differentiate content services from content management platforms, including a shift away from a single, centralized repos­itory to a more decentralized approach with access to information across multiple sources, a focus on faster time to value, and emphasis on using content proactively.

As in other areas of knowledge man­agement, in the future, AI and machine learning are expected to play increasingly prominent roles in content management and services, helping with processes such as classification, metadata extraction, and content identification in order to get infor­mation where it is needed, while also sup­porting security, governance, and lifecycle management.

Cognitive computing and AI continue to evolve

Ten years ago, enterprises struggled to sift through bottomless pits of data, searching for information in unstructured environments.

Valuable time that could have been allocated to more meaningful, productive tasks was wasted as companies sought a system that would allow them to conduct such tasks more efficiently.

Over the course of the past decade, digitalization has created an undeniable shift to bring businesses to where they are now. As companies create more varieties of content, large numbers of audio, video, and big data files have filled cloud and on-premise databases, making the pro­cess of finding relevant data even more difficult for employees.

According to Interact Source, time—the equivalent of one day per working week—is wasted by employees searching for information to do their jobs effectively.

Further, IDC data shows that “the knowledge worker spends about 2.5 hours per day, or roughly 30% of the workday, searching for information.” The challenges posed by searching in data silos are only increasing as enterprises accumulate more big data.

The key to solving this time and cost consumption dilemma may be in “cog­nitive computing,” which offers a new approach to enterprise search.

Cognitive search or insight engines combine powerful indexing technology with advanced natural language process­ing capabilities and machine learning algorithms in order to build an increas­ingly deep corpus of knowledge from which to feed relevant information and 360-degree views to users in real time.

Organizations should seek to “sim­plify, modernize, and automate,” with AI-infused KM. This will include three key processes. One is simplifying knowl­edge development and delivery using the power of AI to bring knowledge to users directly without them having to know how to find it and use it. Another is modernizing the use of knowledge through intelligent, context-driven delivery touchpoints and predictive ability. And the third is auto­mating the process of knowledge delivery by embedding and triggering it directly in tools and processes, and supporting con­tent development through automated tag­ging, linking, and structuring.

AI software platforms provide the functionality to analyze, organize, access, and provide advisory services based on a range of structured and unstructured information.

The technology components include text analytics, rich media analytics, tagging, search, machine learning, deep learning, categorization, cluster­ing, hypothesis generation question answering, visualization, filtering, alert­ing, and navigation.

Keeping an eye on governance

There are many platforms that provide basic and nominal web security, but not nearly enough when it comes to handling private documents. Businesses often deal with clients’ private information, so the bare minimum of security is not sufficient.

Dealing with the fallout after a cus­tomer data breach is something that businesses rarely fully recover from. According to a 2019 study by IBM and the Ponemon Insti­tute, in the event of a data breach, it costs a company an average of $150 per record that was lost or stolen.

The financial consequences of fines and legal fees are compounded by the loss of reputation, which will likely result in decreased business. That’s why it’s imperative to treat sensitive information, or anything that can lead to breaches, with the utmost care.

And it is no longer just the EU’s GDPR that is governing global organi­zations moving forward. In January of 2020, the world’s fifth largest economy implemented the most comprehensive privacy law in the U.S.: the California Consumer Privacy Act (CCPA), accord­ing to Heidi Maher, a certified informa­tion privacy manager and senior director of privacy and compliance at Epiq, who recently wrote an article on compliance for KMWorld.

Similar to GDPR, the CCPA empow­ers consumers—in this case, residents of California—to compel business to disclose personal information collected about them. Under CCPA, Californians can issue a request for information to a business, and that business must then dis­close the categories and specific pieces of personal information it collects. The busi­ness must also indicate where and why that personal information was collected, and with whom it was shared.

What’s more, requests for informa­tion by citizens can seek all data about them going back 12 months. California’s Department of Finance released a report which provided a broad view of the potential costs companies may face in order to become and stay compliant with the CCPA, noted Maher.

On the low end, researchers esti­mated that firms with fewer than 20 employees may have to pay around $50,000 at the outset to become compliant. On the high end, firms with more than 500 employees could pay an average of $2 million in initial costs. Collectively, the researchers estimated about $55 billion will need to be spent by companies to become compliant. In addition, total compliance costs for all companies, subject to the law, could range from $467 million to more than $16 billion over the next decade, Maher explained.

For companies managing CCPA requests, that means they can use content services platforms to connect data about an individual from all those different sys­tems in the business. This will allow them to respond to those requests as they come in, sharing the requested information back to the citizen in an appropriate format and in a timely manner.

Today’s most modern content services platforms will also leverage metadata and AI, which is expected to greatly assist firms in CCPA compliance. An intelli­gent, metadata-driven content services platform can automatically categorize assets that contain personal consumer information, and ensure it is properly managed according to various CCPA requirements—and whatever new regu­latory mandates emerge in the future.

Knowledge graphs rise

The ability for knowledge graphs to capture information and relationships and connect facts is showing potential for a range of use cases in fields as diverse as healthcare, cybersecurity, law, marketing, and banking and finance.

Knowledge graphs represent a col­lection of interlinked descriptions of entities—real-world objects, events, sit­uations, or abstract concepts.

Knowledge graphs built on top of semantic technologies, supported by machine learning technologies, can create a radical shift in metadata management. They can aid in information retrieval by semantic tagging, query expansion, fac­eted search, classification, or similari­ty-based recommender.

“Knowledge graphs are the future of taxonomy work,” noted Mike Doane, senior lecturer, Information School, Uni­versity of Washington, at KMWorld 2019. “They are the logical extensions of ontologies.”

Collaboration is critical

For many companies, the rise of inno­vative collaboration solutions supported by modern technologies is adding to their agility and efficiency.

According to a research report on the enterprise collaboration mar­ket published in September 2019 by MarketsandMarkets, the global enterprise col­laboration market size is projected to grow from $31 billion in 2019 to slightly more than $48 billion by 2024 for a com­pound annual growth rate of 9.2% during the forecast period.

Spurring the growth is increasing use of social networking websites and increasing usage of mobile devices for enterprise collaboration, according to the research. In addition, increasing demand for AI across the globe and a variety of emerging potential markets are gener­ating opportunities for vendors in the segment.

The survey report points out that enterprise collaboration solutions offer banking, financial services, and insur­ance (BFSI) companies a complete, ubiquitous communication and collab­oration environment, as the BFSI sector is becoming more innovative, customer-centric, and collaborative. In particular, the report notes, many companies are adopting solutions, such as unified com­munication, video and audio conferenc­ing, and intranet platforms, to improve employee efficiency, promote team­work, and simplify communication and collaboration.

What’s ahead

With the growing array of AI-infused knowledge management solutions, the goal is to put more accurate and timely information where it can be of most value. However, experts caution that the technol­ogy is not a panacea and human involve­ment will still be necessary.

There are still hurdles to overcome since corporate information resides across many silos, and AI is an immature technology that is still in the proof-of-concept stage, not ready to scale on a widespread basis, noted Unisphere Research's McKendrick. “Before we see ‘Alexa for the Enterprise’—whether by voice or text queries—organizations will need to step up their efforts to integrate and standardize their data asset."

And, noted  Murray of Applied Knowledge Sciences , “It is important for organizations to know what’s going on inside the ‘black boxes’ of AI, analytics, and machine learn­ing, how the knowledge was generated, why it was generated, and how to deter­mine its validity.”

Organizations must also understand that learning occurs in both directions, Murray said. “Humans teach the computers and computers teach the humans. But both can be wrong—so the cycle of learning is nev­er-ending.”