Teklens logo
DEEN

AI agents and technology

What is a knowledge graph? Definition and examples

Also: Enterprise knowledge graph

In short

A knowledge graph stores knowledge as a network: things such as customers, projects and decisions are nodes, and their relationships are edges. AI can then follow connections, such as which projects depend on a decision, instead of only finding similar text.

Why it matters: A knowledge graph shows what depends on a decision, not just where it is mentioned.

Why does Knowledge Graph matter for your company?

Semantic search finds text that sounds similar. But it does not know that an epic belongs to a project, the project to a market and the market to a strategy decision. A knowledge graph records exactly these relationships.

For leadership this matters as soon as something changes. With a graph you can ask: which projects, commitments and AI agents depend on this decision? The effort lies in upkeep, because a graph is only useful when its relationships are right.

Example: What depends on the US expansion?

Leadership decides “Strategy 2027: focus on the DACH mid-market, no US expansion”. The question comes at once: what is affected? A full-text search for “USA” returns a long list, from expense claims to the pitch deck.

A knowledge graph knows the relationships: a launch project, a customer commitment and a partner contract depend on the US market. Leadership sees straight away whom to inform and what to stop.

3 common mistakes to avoid

  • Starting with the whole company

    A graph of everything is never finished. Start with the things that depend on important decisions.

  • Capturing relationships only once

    Projects end and owners change. Without ongoing upkeep the graph shows a state that no longer exists.

  • Pitting the graph against search

    A graph and semantic search complement each other. Search finds content, the graph shows how it connects.

Step by step

  1. 1

    Name the key things

    For example decisions, projects and customers. You need no more at the start.

  2. 2

    Define the relationships

    Such as “project implements decision” or “customer holds commitment”.

  3. 3

    Fill it from the sources

    Derive nodes and relationships from Jira, Confluence and the CRM instead of maintaining them by hand.

  4. 4

    Test with a real question

    Ask: what depends on our last big decision? Is the answer right?

Bring one important decision.

In 30 minutes with Marc or Simon we show where your work drifts from the decision today.

Meet the Company Brain

Read on in the PM Lab

Articles that go deeper.

Build · Live ViewAI product operating model: from managing work to managing contextThe teams behind Claude Code, Codex and Cursor are not only coding faster. They delegate to agents, hand over whole features, build two prototypes instead of holding one meeting, and still hire product managers. Seven patterns from more than thirty interviews, one shift underneath: the scarce resource is no longer coding capacity. It is shared context and judgement. And why copying them blindly backfires.Read the article Define · PrioritisationBuild your own company brain: the enterprise AI playbook from Stripe's engineering teamStripe gave 10,000 employees powerful agents without turning them into AI engineers. The ten decisions behind Kai: one company layer, projects as governance, skills as company IP, sandboxes, and the infrastructure underneath. And what the same architecture looks like for a software product team.Read the article

Related terms

Back to the glossary