Why Causal Schema Exists

Schema.org provides excellent vocabulary for describing what things are—products, organisations, articles, medical conditions. However, it lacks the ability to express why things happen or how to reason about them.

Consider these common content patterns that schema.org cannot adequately represent:

• Multi-step mechanisms: "Drug X works by first blocking receptor Y, which reduces enzyme Z, leading to outcome W"
• Temporal progressions: "This condition develops through four distinct stages over 5-10 years"
• Multi-factorial causation: "Three factors combine to cause this outcome, with factor A being most important"
• Conditional recommendations: "If the property has characteristics X and Y, recommend treatment Z"

Causal Schema fills this gap with nine core types that work across any industry.

Origin & Evolution

Causal Schema evolved from HealthSchema, a medical-focused vocabulary developed for health content. When building HealthSchema, we realised that the causal reasoning patterns—pathways, progressions, factor interactions—weren't specific to healthcare. They apply equally to:

• Building pathology (how defects develop and progress)
• Automotive diagnostics (failure mechanisms and repair decisions)
• Agricultural science (disease progression and treatment timing)
• Infrastructure management (asset deterioration and intervention planning)

We extracted the domain-agnostic patterns into Causal Schema, keeping HealthSchema as a future health-specific extension.

Design Principles

Extension, Not Replacement

Causal Schema augments schema.org rather than replacing it. Use standard schema.org types wherever possible; add Causal Schema types when you need to express causal reasoning.

Single Namespace

All types—core and domain extensions—share the cs: prefix and https://causalschema.org/ namespace. This simplifies implementation and ensures consistency.

LLM-Optimised

These types are designed to help large language models truly understand causal content, enabling better reasoning, more accurate answers, and appropriate hedging based on evidence levels.

Evidence-Based

Properties support evidence citations and confidence levels, enabling transparent documentation of what's known and what's uncertain.

Domain-Agnostic Core

Core types work across industries. Domain-specific extensions (like the construction extension) add specialised types while inheriting core capabilities.

Who Maintains This

Causal Schema was originally developed by Creative Touch for real-world content markup needs. It is released as open-source under the Creative Commons Attribution 4.0 licence.

The vocabulary is designed to be community-driven. We welcome:

• Feedback on existing types and properties
• Proposals for new domain extensions
• Real-world implementation examples
• Bug reports and documentation improvements

Technical Details

• Namespace URI: https://causalschema.org/
• Prefix: cs:
• Current Version: 0.1 (December 2025)
• Licence: Creative Commons Attribution 4.0 International (CC BY 4.0)
• JSON-LD Context: https://causalschema.org/context.jsonld

Contact

For questions, suggestions, or collaboration:

• Email: contact@causalschema.org
• Website: CausalSchema.org