Data sources, staging area, data warehouse, data marts, user (reporting) layer.

Temporary data storage (Staging Area), central data warehouse, shared measures and dimensions, OLAP, reporting.

Temporary data storage (Staging Area), data mart, shared measures and dimensions, OLAP, reporting.

A hard-coded constant value that remains unchanged regardless of user interactions or report filters.

A static value that is computed row-by-row during data refresh and physically stored in the table.

A dynamic calculation that is evaluated at query time, aggregating data based on the current filter context.

A specialized table function used exclusively to return a list of values for iteration, not a scalar result.

ETL stands for "Email, Transfer, Log," a protocol used strictly by mail servers to move attachments between database and Power BI, having no relation to data analysis or databases.

Data transformation refers exclusively to encrypting sensitive data into unreadable text strings. It does not involve cleaning, filtering, or merging different datasets.

ETL (Extract, Transform, Load) connects to various sources (like databases or files), cleans and standardizes the data (transform), and saves it into a central destination (load).

A fact table is used to store descriptions (attributes) of an event or entity, whereas these attributes usually change slowly.

Measurable values (facts) must always be additive to prevent incorrect aggregations. Non-additive facts (e.g., ratios) should always be calculated only in the report.

A fact table should always contain atomic (lowest level) data to allow for flexible aggregation and to answer queries at any level of granularity.

Granularity describes the frequency at which the data warehouse is updated (e.g., real-time vs. nightly batch). The recommendation is to store multiple grains (e.g., daily transactions and monthly summaries) within the same physical fact table to allow users to switch between detailed and summary views without joining tables.

Granularity is the count of foreign keys present in the fact table schema. The recommendation is to design the fact table based on the primary key constraints of the source system, ensuring that the warehouse granularity strictly mirrors the operational system's primary keys to maintain data integrity.

Granularity refers to the number of dimensions linked to the fact table. The recommendation is to aggregate data to a higher level (e.g., monthly sales instead of individual transactions) during the ETL process to significantly reduce storage costs and improve query performance.

Granularity defines the specific business event or measurement represented by a single row in the fact table. The recommendation is to store data at the lowest possible (atomic) level of detail to ensure maximum analytical flexibility and the ability to roll up to any required summary.