Source of Truth

What is a Source of Truth?

A source of truth (also called system of record) is the authoritative source for a particular data element in an integrated system. It answers the question: "Which system's data should be considered correct if there's a conflict?"

Key characteristics of a source of truth include:

• Authority: Recognized as the definitive reference for specific data
• Ownership: Clear responsibility for data quality and maintenance
• Integrity: Contains the most accurate and up-to-date information
• Governance: Subject to established data management policies

Why Source of Truth Matters

When integrating multiple systems, establishing clear sources of truth helps:

• Prevent data conflicts: Clear rules for resolving inconsistencies
• Improve data quality: Focus data governance efforts on authoritative sources
• Simplify architecture: Clear data flow patterns from source to consuming systems
• Streamline troubleshooting: Know immediately where to investigate data issues

Common Source of Truth Patterns

Single System of Record

One system is the definitive source for a specific entity or attribute:

Customer Data Source of Truth: CRM System
↓ (one-way sync)
FileMaker
↓ (one-way sync)
E-commerce Platform

Domain-Specific Sources of Truth

Different systems are authoritative for different aspects of the same entity:

Customer Contact Info: CRM System
Customer Purchase History: E-commerce Platform
Customer Support History: Help Desk System

Temporal Source of Truth

The authoritative source changes based on the lifecycle stage:

Lead Stage: Marketing Automation (source of truth)
↓ (conversion)
Opportunity Stage: CRM System (source of truth)
↓ (purchase)
Customer Stage: ERP System (source of truth)

Derived Source of Truth

Data derived or calculated from multiple primary sources:

Inventory Levels: Warehouse System (primary)
+ Reserved Items: Order System (primary)
= Available Inventory: Inventory Service (derived source of truth)

Source of Truth Challenges

Multiple Writers Problem

When multiple systems can update the same data:

• Last-Writer-Wins: Simple but can lose important changes
• Merge Strategy: Combine non-conflicting changes (complex)
• Manual Resolution: Human intervention for conflicts (reliable but slow)
• Version Vectors: Track update lineage across systems (complex but complete)