Single Source of Truth

Short answer: A single source of truth is a practice more than a product. For each important piece of information, such as order volumes, headcount or the customer list, the company agrees where the official version lives. It also writes down what that item means. That source can be one system, or a reporting layer such as a data warehouse that combines several systems. People still use many tools, but they take official figures from the agreed place.

The term is closely related to system of record, and people sometimes confuse the two. A system of record is the authoritative system for one type of data, such as the order system for orders or the staff records system for employees. A single source of truth usually means the wider arrangement. It often brings data from several systems of record together so teams can use it consistently. Signs that a company lacks one include teams reporting different order numbers, several versions of the customer list in spreadsheets and meetings spent reconciling figures. Achieving it depends on agreed definitions, named owners and rules about where people create and edit each item of data.

The team lists the measures and records that matter most. It decides the system of record for each one and writes down the definition and owner. Pipelines bring data from those systems into one reporting layer, often a warehouse. The team retires spreadsheets that copy the official figures, or labels them clearly as working copies. It publishes the definitions in a business glossary or data dictionary. Only the owner can approve a change to a definition.

A distributor sees three different stock counts for its best-selling product. The warehouse system shows 1,200 units, the online store shows 1,150 and the sales team's spreadsheet shows 1,300. Each gap has a cause. Open orders already claim 50 of the 1,200 units, so the store shows 1,150 as available. Nobody has updated the spreadsheet for two weeks, and it still includes 100 units that have since shipped. The company makes the warehouse system the official source for stock. The store and the sales team both read from it, and the store labels its figure as available stock. The team retires the spreadsheet, and sales staff stop promising stock that has already left.

Automations and reports work only when they read the same records. If an automation reads one customer list and a report reads another, staff see conflicting results and lose trust in both. When leaders spend meeting time debating whose number is right, decisions slow down and confidence drops. Decisions on staffing, stock and service depend on figures that people trust.

The idea is rarely literal. Most companies keep several authoritative systems, each for its own type of data. Trying to force everything into one large system can be costly and disruptive. Bringing data together does not fix poor data at the source. Some figures legitimately differ by purpose, such as stock on hand and stock available to sell. The aim is a clear definition and label for each measure.