Developer

Metrics Dictionary

Build a business-facing glossary for metrics, dimensions, datasets, owners, synonyms, formulas, examples, and approval status before technical modeling.

Business vocabulary

Metrics Dictionary

Define business terms before mapping them to warehouse datasets. Use this for agreement, ownership, synonyms, and metric documentation.

Terms

Approved

In review

Glossary workspace

Business meaning first. Technical mapping can happen later in the semantic layer builder.

2 metrics ready to map

Metric

A business KPI with an agreed definition and calculation.

Revenue, Active User, Churn Rate

Dimension

A grouping or filter used to analyze a metric.

Date, Segment, Region

Dataset

A source table or logical source used by technical models.

Orders, Customers, Events

Policy

A rule that explains exclusions, ownership, or governance.

Refund policy, PII handling

TermTypeOwnerStatusDefinition

Dictionary export

{
  "metricsDictionary": {
    "terms": [
      {
        "id": "revenue",
        "name": "Revenue",
        "type": "metric",
        "owner": "Finance",
        "status": "approved",
        "definition": "Recognized order value after discounts and before refunds.",
        "formula": "SUM(orders.gross_amount - orders.discount_amount)",
        "example": "Revenue by order date and customer segment.",
        "synonyms": [
          "sales",
          "gross revenue"
        ],
        "tags": [
          "finance",
          "core"
        ]
      },
      {
        "id": "active_user",
        "name": "Active User",
        "type": "metric",
        "owner": "Product Analytics",
        "status": "review",
        "definition": "A user with at least one qualifying product event in the selected period.",
        "formula": "COUNT_DISTINCT(events.user_id) WHERE events.is_qualifying = true",
        "example": "Weekly active users by plan.",
        "synonyms": [
          "WAU",
          "MAU"
        ],
        "tags": [
          "product",
          "engagement"
        ]
      },
      {
        "id": "customer_segment",
        "name": "Customer Segment",
        "type": "dimension",
        "owner": "RevOps",
        "status": "approved",
        "definition": "Commercial grouping used for customer reporting and pipeline analysis.",
        "formula": "customers.segment",
        "example": "Enterprise, Mid-market, SMB.",
        "synonyms": [
          "segment",
          "market segment"
        ],
        "tags": [
          "customer"
        ]
      },
      {
        "id": "orders_dataset",
        "name": "Orders Dataset",
        "type": "dataset",
        "owner": "Data Platform",
        "status": "approved",
        "definition": "Order fact table at one row per order.",
        "formula": "warehouse.fact_orders, primary key order_id",
        "example": "Source for revenue, order count, and order date.",
        "synonyms": [
          "fact orders"
        ],
        "tags": [
          "warehouse",
          "commerce"
        ]
      }
    ]
  }
}

What is a metrics dictionary?

A metrics dictionary is the business agreement layer before technical implementation. It defines what terms mean, who owns them, what aliases people use, and how a metric should be calculated or interpreted.

How to build a metrics dictionary

1
Add business terms
Create metrics, dimensions, datasets, and policies in plain business language.
2
Assign ownership
Give every important term an owner so review questions have a clear destination.
3
Document definitions
Write the formula, example usage, synonyms, tags, and approval state for each term.
4
Review status
Use draft, review, and approved states to separate rough terms from governed definitions.
5
Export the glossary
Copy JSON or Markdown for documentation, analytics specs, or semantic-layer handoff.

Dictionary structure

metrics_dictionary:
  terms:
    - name
      type: metric | dimension | dataset | policy
      owner
      status: draft | review | approved
      definition
      formula
      synonyms[]
      tags[]

The dictionary keeps business meaning separate from technical implementation. Metrics can later be mapped to datasets, dimensions, and joins in the Semantic Layer Builder.

Reference: dbt - Semantic Layer

Examples

Business term	Dictionary entry
Revenue	Approved finance-owned metric with formula, synonyms, and reporting example. Prevents each dashboard from inventing a different revenue definition.
Active User	Product metric in review with qualifying activity rules. Captures ambiguity before it turns into conflicting reports.
Customer Segment	Dimension owned by RevOps with examples and aliases. Makes business grouping language consistent across teams.

Common use cases

Creating a metric glossary for business teams
Capturing metric owners and approval status
Documenting synonyms and common aliases
Preparing requirements before semantic-layer modeling
Exporting glossary docs for analytics handoff
Separating business definitions from warehouse implementation

Related tools

JSON-Formatter

Formatieren, validieren und minimieren Sie JSON mit präziser Fehlerzeilen- und Spaltenberichterstattung.

CSV Combiner & Analyzer

Drop multiple CSV files, validate matching headers, merge the rows, analyze columns, and download one clean file.

Ontology Builder

Sketch entities, relationships, owners, and fields, then export a clean JSON ontology for data and product teams.

Semantic Layer Builder

Model datasets, dimensions, measures, joins, and SQL previews for analytics teams before committing to code.

Frequently asked questions

How is this different from the Semantic Layer Builder?

The Metrics Dictionary defines business meaning, owners, aliases, and review status. The Semantic Layer Builder maps those definitions onto datasets, joins, SQL expressions, and export formats.

What should be a metric?

A metric is an aggregated business KPI such as revenue, active users, churn rate, or gross margin.

What should be a dimension?

A dimension is a descriptive grouping or filter such as date, region, plan, segment, or product category.

Do I need technical table names here?

Not necessarily. Keep this tool business-facing. Use the Semantic Layer Builder when you are ready to connect terms to tables and SQL.

Last updated 2026-06-18

{ "metricsDictionary": { "terms": [ { "id": "revenue", "name": "Revenue", "type": "metric", "owner": "Finance", "status": "approved", "definition": "Recognized order value after discounts and before refunds.", "formula": "SUM(orders.gross_amount - orders.discount_amount)", "example": "Revenue by order date and customer segment.", "synonyms": [ "sales", "gross revenue" ], "tags": [ "finance", "core" ] }, { "id": "active_user", "name": "Active User", "type": "metric", "owner": "Product Analytics", "status": "review", "definition": "A user with at least one qualifying product event in the selected period.", "formula": "COUNT_DISTINCT(events.user_id) WHERE events.is_qualifying = true", "example": "Weekly active users by plan.", "synonyms": [ "WAU", "MAU" ], "tags": [ "product", "engagement" ] }, { "id": "customer_segment", "name": "Customer Segment", "type": "dimension", "owner": "RevOps", "status": "approved", "definition": "Commercial grouping used for customer reporting and pipeline analysis.", "formula": "customers.segment", "example": "Enterprise, Mid-market, SMB.", "synonyms": [ "segment", "market segment" ], "tags": [ "customer" ] }, { "id": "orders_dataset", "name": "Orders Dataset", "type": "dataset", "owner": "Data Platform", "status": "approved", "definition": "Order fact table at one row per order.", "formula": "warehouse.fact_orders, primary key order_id", "example": "Source for revenue, order count, and order date.", "synonyms": [ "fact orders" ], "tags": [ "warehouse", "commerce" ] } ] } }

How to build a metrics dictionary

Add business terms

Create metrics, dimensions, datasets, and policies in plain business language.

Assign ownership

Give every important term an owner so review questions have a clear destination.

Document definitions

Write the formula, example usage, synonyms, tags, and approval state for each term.

Review status

Use draft, review, and approved states to separate rough terms from governed definitions.

Export the glossary

Copy JSON or Markdown for documentation, analytics specs, or semantic-layer handoff.

Dictionary structure

metrics_dictionary:
  terms:
    - name
      type: metric | dimension | dataset | policy
      owner
      status: draft | review | approved
      definition
      formula
      synonyms[]
      tags[]

The dictionary keeps business meaning separate from technical implementation. Metrics can later be mapped to datasets, dimensions, and joins in the Semantic Layer Builder.

Examples

Business term	Dictionary entry
Revenue	Approved finance-owned metric with formula, synonyms, and reporting example. Prevents each dashboard from inventing a different revenue definition.
Active User	Product metric in review with qualifying activity rules. Captures ambiguity before it turns into conflicting reports.
Customer Segment	Dimension owned by RevOps with examples and aliases. Makes business grouping language consistent across teams.

Frequently asked questions

How is this different from the Semantic Layer Builder?

The Metrics Dictionary defines business meaning, owners, aliases, and review status. The Semantic Layer Builder maps those definitions onto datasets, joins, SQL expressions, and export formats.

What should be a metric?

A metric is an aggregated business KPI such as revenue, active users, churn rate, or gross margin.

What should be a dimension?

A dimension is a descriptive grouping or filter such as date, region, plan, segment, or product category.

Do I need technical table names here?

Not necessarily. Keep this tool business-facing. Use the Semantic Layer Builder when you are ready to connect terms to tables and SQL.