Sweta Bhagat, ServiceNow
April 15, 2025

Howdy, folks! Ever heard of Coach Ted Lasso? The coach who knew nothing but taught us everything. When I first watched Ted Lasso, I wasn’t just entertained but was inspired. Ted didn’t need to be a football expert to build a winning team. He focused on structure, relationships, and trust. And that got me thinking, what if we applied the same approach to structured content and knowledge graphs?

Just like Ted Lasso turns a scattered group of players into a cohesive team, knowledge graphs organize scattered information into a structured, meaningful system. Let’s take a page from Ted’s playbook and see how his mindset can help us manage knowledge better!

Ted Lasso and the Power of Connections

For those who don’t know, Ted Lasso is a series based on super-optimistic American college football coach, Ted Lasso. He is hired to manage AFC Richmond, a fictional English professional football team based in London. The twist? Ted has no experience coaching football. But that’s not what makes the show special. The real magic comes from the relationships between the players, staff, fans, and even Ted’s quirky yet inspiring coaching style. It’s these connections that drive the team’s journey and growth, both on and off the field.

Much like Ted’s approach, knowledge graphs are built on relationships. They link different pieces of information together, making it easier to understand how things are connected. These connections help us find the right information and make better decisions, just like Ted’s relationships help his team improve.

DITA: The Foundation of Structured Content

First, let’s talk about Darwin Information Typing Architecture (DITA), an XML-based framework for structured content. Imagine DITA as the playbook of a football team, where each module represents a player with a specific role aimed at achieving a set goal. DITA ensures modularity, consistency, and scalability, making content easier to manage, repurpose, and adapt for various outputs.

DITA is also AI-ready, thanks to semantic tagging, metadata, and self-contained topics. As a result, AI can retrieve and assemble relevant content dynamically, delivering the right information at the right time.

Knowledge Graphs: Building Connections

Think of knowledge graphs as your content’s version of a well-connected team. They help AI understand relationships, retrieve accurate information, and make smarter decisions. Kind of like team roster, showing how each player representing a piece of content is connected, who’s the star striker, who’s mentoring whom, and how they all contribute to success.

And hey, just like Ted Lasso wins over his boss with those famous homemade biscuits, a well-structured knowledge graph sweetens the deal for AI-powered search and content discovery. When your information is properly connected, it’s easier to find, use, and trust. Turning scattered content into a winning strategy.

Knowledge graphs rely on two key components:

  • Resource Description Framework (RDF): This structures data using triples (subject-predicate-object), similar to how you would define the dynamics of a team. Think of it as: “Ted Lasso (subject) coaches (predicate) Jamie Tartt (object).”
  • Ontologies: These structured, machine-readable representations of knowledge ensure AI doesn’t just store data but understands how concepts interconnect. Kind of like Ted as a coach ensuring every player understands their position on the field.

Why Are Knowledge Graphs a Game Changer?

In a world where intelligence is cheap, meaning becomes everything. Knowledge graphs create a network of insights that make AI smarter and content discovery seamless. They turn raw data into a connected, meaningful system that improves search, recommendations, and automation.

Lassoism: Connect, Adapt, Believe

Now the Ted Lasso show isn’t explicitly a knowledge graph, but this thought experiment helps to understand how knowledge graphs organize information. This brings us to Lassoism, an approach inspired by Ted’s leadership style that combines optimism, collaboration, and strategy to enhance knowledge management.

Just like Ted builds strong relationships among his players, knowledge graphs connect information for better discovery. When combined with DITA, you can create a dynamic ecosystem that adapts, evolves, and provides meaningful insights.

Imagine a football team as a basic graph. In Ted Lasso, Roy Kent is the seasoned veteran, and Jamie Tartt is the talented but impulsive player. In a basic graph, you’d only see that they are teammates, with players as the nodes and their interactions forming the edges.

But a knowledge graph takes it a step further. Instead of simply noting that they’re on the same team, a knowledge graph highlights how Roy Kent mentors Jamie Tartt. This deeper understanding helps AI with search, recommendations, and automation. Let’s look at the following graphic to see how this deeper connection works in action.

 

But AI needs to go beyond just static knowledge and always retrieve the latest, most relevant information. That’s where RAG comes in.

RAG: The Ted Lasso of AI

Retrieval-Augmented Generation (RAG) enables AI give better answers by first looking up the latest information from reliable sources and then using that information to create a response. This method enhances accuracy, minimizes hallucinations, and ensures AI-generated content remains relevant and useful.

Ted Lasso didn’t start as a football expert, but he knew how to learn, adapt, and bring out the best in his team. That’s exactly what RAG does. It identifies the right information, builds connections, and keeps improving. Let’s break down the evolution of RAG through Ted’s journey:

 

 

  1. Vector-Based RAG: Early Ted, relying on basic stats. At first, Ted focuses on basic data, like win-loss records, player stats, and surface-level knowledge. Similarly, early RAG methods retrieve information based on simple keyword matching without deeper understanding.
  2. Graph-Based RAG: Ted learning team dynamics. As Ted gets to know his players, he understands how they interact and play together. Graph-based RAG works the same way, mapping relationships between concepts to add more context and meaning to responses.
  3. Knowledge Graph-Driven RAG: Ted refining his playbook. By this stage, Ted is no longer just reacting. He’s strategizing. He connects past games, player strengths, and team dynamics to make smarter decisions. Likewise, knowledge graph-driven RAG organizes and links information for more precise and meaningful retrieval.
  4. Document Object Model (DOM) Graph RAG: Ted at his peak. Now, Ted’s approach is fully integrated. He combines team psychology, tactical plays, and data insights to create a deep understanding of the game. DOM Graph RAG does something similar by blending structured content, metadata, and AI to provide the most accurate and context-aware responses.

RAG keeps evolving, just like Ted. New techniques, such as HippoRAG, mimic how our memory works and help AI gather insights from different sources over time. A great coach remembers every lesson and uses it to shape what’s ahead.

From Content to Intelligence: The Winning Workflow

 

Transforming structured content into an intelligent knowledge system involves:

  1. Authoring: Creating modular DITA topics with rich metadata that are flexible and reusable.
  2. Structuring: Organizing topics in DITA Maps, aligning them with an ontology.
  3. Automation: Building and enriching a knowledge graph with taxonomies and implementing RAG.
  4. Utilization: Enabling contextual AI-driven search with SPARQL-powered queries, a query language used for retrieving and working with data stored in RDF format.

This structured approach ensures content isn’t just stored, but can be discovered, modified, or further refined by AI. You can always alter this workflow to bring in manual review and editing workflows.

The “Believe” Effect in Action: The Richmond Way!

Ted inspires his team to believe in themselves, and a well-organized content system does the same by enabling users trust the information they need. When content is properly managed, it’s not just available but also useful, leading to higher adoption and better insights.

So, here is the takeaway. Just as Ted Lasso builds a winning team by fostering trust and strong relationships, a well-structured content system brings information together in a way that users can rely on. Believing in connections and applying Lassoism while using DITA and knowledge graphs sets you up for success in building a strong content ecosystem.

In Ted Lasso’s words, Believe. And if you really want to embrace his spirit, bake some biscuits for the boss because relationships matter. Whether on the field or in content management, success comes from the connections we build. Let your knowledge graph help you connect the dots, adapt, and always improve.