Propelling intelligent content services with AI and other strategies

As content services have steadily increased in their criticality toward creating scalable business applications, capable of democratizing information access, sharing, and collaboration, they have far surpassed their archaic, monolithic ECM system predecessors. Blending the phenomenon of artificial intelligence with content services is the next big thing, applicable to a variety of use cases that can optimize content processes.

Noz Urbina, founder of Urbina Consulting, and Chip Gettinger, VP of global solutions consulting at RWS, joined KMWorld’s webinar, Succeeding with Modern Intelligent Content Services (and AI), to discuss what makes a content service truly intelligent, as well as how that applies and intersects with various other technologies and industries.

Urbina began by explaining the top market drivers for content services across industries and departments:

• Next best action based on all available data
• Becoming omnichannel, an audience-centric, integrated approach to value delivery
• Better metrics to measure performance
• Greater personalization pushing more meaningful content
• Generate larger ROI and scalability
• Balance both global and local needs with consistency

These needs reveal the necessity for an omnichannel, service-based infrastructure that manages content as a business asset, separate from format and channel, with scalable, efficient processes.

To propel the idea of content as a business asset, Urbina urged audiences to consider the strategy of “components,” or a pool of modular content assets that replace the idea of containers for a more integrable, dynamic, and reusable system. Whether the objects are videos, infographics, slides, or images, they should all exist as components to empower a more intelligent content system.

This “component” strategy leads us to intelligent content services, or a service that can be called from multiple applications or channels, where ML and AI can support the processing and manipulation of both structured and unstructured data. Examples of these services include auto-tagging, semantic search, ranking, recommending, insight engines, and many more.

Some intelligent content services work to transform hundreds of thousands of pages of content in legacy format, typically manually maintained and ridden with broken links, as restructured content ready for multi-format output with bad links removed or flagged for human review—known more commonly as Conversion-as-a-Service.

Like Conversion-as-a-Service, other intelligent content services drive the automation and optimization of content processes, including auto-tagging, content strategies for personalization, and omnichannel-driven “next best actions.”

Urbina explained that managing content assets across repositories, equipped with next best action recommendations and omnichannel delivery, is a valid form of intelligent content services that emphasizes the use of next best action tools—solutions where systems use AI and data, which can be from various sources, including other AIs, to determine what’s best to do in each situation. Next best action services can:

• Link content topics to relevant people, with contact options, to support better internal and external knowledge journeys
• Leverage personalization tech driven by ontology relationships and CRM data to recommend content items related to an audience’s subjects’ potential interest

Auto-tagging—which Urbina defined as an iterative process that improves with time to apply existing taxonomies to new and existing content for easier identification and comprehension. Terms are mined from an organization’s existing corpus of content, combined with any existing taxonomies, and are structured and applied automatically to content. Auto-tagging can further be fine-tuned with human feedback.

Gettinger transitioned the discussion to focus on the significance of semantic AI, a process of adding intelligence to address data fragmentation across external content and data sources. An intelligent semantic layer translates fragmented data to be more efficiently filtered into web delivery and search, contextual delivery, and documents.

An intelligent content platform, such as Tridion, manages the delivery of data once it's been translated by a semantic layer. Tridion offers auto-tagging, one of the coveted intelligent content services, with content components. This provides:

• Greater findability and actionable insights through intelligent semantic search and knowledge graphs
• Knowledge-as-a-Service which offers dynamic delivery of data components rather than documents, equipped with content and existing as structured and modular
• Intelligent Content Hubs that provide a portal for collaboration, empower self service, and maintain a single source of truth

Semantic AI can also help to amplify the process of targeted recommendations, another form of intelligent content services. With the Tridion Dynamic Delivery Experience (DXD), users can configure concept schemas and boost relative weights, drive recommendations for headless delivery, deliver targeted content components at scale, and harness low-code, multichannel delivery.

For an in-depth review of intelligent content services, use cases, and examples, you can view an archived version of the webinar here.