Data warehouse planning

The data warehouse planning page in the DataWorks console is where data warehouse architects and model group members design the structure of a data warehouse. From this page, you define data layers, business categories, data domains, business processes, data marts, and subject areas — the foundational building blocks that shape how your data is organized, transformed, and consumed across the enterprise.

Key concepts

DataWorks data warehouse planning is organized around two parallel frameworks: a technical layer framework that defines how data moves through transformation stages, and a business-driven framework that mirrors your organization's structure.

Data layers

A data layer is a logical tier in your data warehouse that reflects a specific stage of data processing. By default, DataWorks divides a data warehouse into five layers:

Data flow rules: Data flows forward through the layers — ODS feeds DWD, DWD feeds DWS and DIM, and DWS feeds ADS. Reverse dependencies are not permitted. DWD is the only layer that reads directly from ODS; DWS and ADS must not reference ODS tables directly.

You can customize this default set — add layers, rename them, or remove unused ones — to match your organization's methodology.

Business categories

A business category is the highest-level division in the business-driven management framework. It represents a broad area of your enterprise, such as Finance, Marketing, or Supply Chain. Business categories are the top of the hierarchy:

Business Category
  └── Data Domain  ─── Business Process
  └── Data Mart    ─── Subject Area

Data domains

A data domain groups related business processes under a business category. Where a business category defines a broad area, a data domain narrows the scope to a specific functional unit within that area — for example, "Order Management" within "E-Commerce."

Business processes

A business process is the atomic unit of data production within a data domain. It represents a concrete, measurable activity that generates data — for example, "Place Order," "Return Item," or "Process Payment." Each business process maps to one or more fact tables in your dimensional model.

When defining a business process, consider the following:

• What measurable event does this process record? (the grain)

• What metrics does it produce? (quantities, amounts, durations)

• What dimensions provide context? (who, what, when, where)

Data marts

A data mart is a focused subset of the data warehouse, optimized for a specific department, product line, or analytical function. Data marts belong to the business category level and are organized separately from data domains.

Subject areas

A subject area is a collection of business subjects used to categorize data in a data mart from multiple analytical perspectives. Where a data domain organizes data production by business process, a subject area organizes data consumption by analytical theme — for example, "Customer Profitability" or "Inventory Aging" within a supply chain data mart.

How it works

Setting up data warehouse planning in DataWorks follows a top-down sequence:

1. Define data layers — Establish the technical tiers that govern how data moves from raw ingestion to serving.

2. Create business categories — Set the top-level business divisions that reflect your organization's structure.

3. Add data domains and data marts — Under each business category, define the functional areas (data domains) and department-level analytical stores (data marts).

4. Define business processes — Under each data domain, create the specific data-producing activities that generate fact tables.

5. Create subject areas — Under each data mart, group analytical themes that describe how end users query the data.

Once this structure is in place, DataWorks model designers use it as the governance backbone: every model, metric, and dataset is registered under the appropriate layer, domain, and business process, making lineage tracking and impact analysis accurate across the warehouse.

What's next

• Business planning — Create and configure business categories, data domains, data marts, and subject areas in the DataWorks console.

• Data layer management — Add, edit, or remove data layers to match your warehouse architecture.