What is a Data Warehouse? 4 Types + Key Features

A data warehouse is an enterprise system that aggregates structured data from sources like CRM, POS transactions, and marketing into a unified repository.

Data Cloud platform in a ssot dashboard

Say hello to Data Cloud.

Data Cloud, the only data platform native to Salesforce, unlocks and harmonises data from any system — so you can better understand your customers and drive growth.

To build on this, let’s explore the key differences.

Key features Data warehouse (OLAP) Database (OLTP)
Main purpose Analysing aggregated data from various sources Processing transactional data
Use cases Sales analysis, evaluating marketing performance, financial forecasting Processing orders, managing customer details, updating stock levels
Data formatting Multidimensional data modelling focussing on data from multiple angles Unidimensional data, organising data into simple tables
Data architecture Data read operations, optimised to analyse a large volume of data Data write operations, optimised to update transactional data in real-time
Performance times Usually minutes, depending on the data being analysed In real-time, usually in milliseconds
Scalability High scalability Limited scalability
Useful for Business analysts, data experts, key decision-makers Operational staff, sales reps, end-users

Expanding on this, let's take a closer look at the key differences.

Key features Data warehouse Database Data lake
Main purpose Analysing aggregated data Transactional processing of data Exploring big data
Use cases Sales analysis, evaluating marketing performance, financial forecasting Processing orders, managing customer details, updating stock levels Running machine learning (ML models), exploratory data analysis
Data structure Primarily structured and semi-structured, though modern data warehouses can store some unstructured data Structured Structured, semi-structured and unstructured
Performance times Usually minutes Milliseconds (real-time) Minutes to hours depending on data size
Scalability High scalability Limited scalability High scalability
Useful for Business analysts, data experts, key decision makers Operational staff, sales reps, end-users Data scientists, business analysts, artificial intelligence and machine learning teams
data culture playbook connecting data and ai

Gartner named Salesforce a Leader in Customer Data Platforms. See why.

data cloud professional service consultant guidance image

Get expert Data Cloud guidance from Salesforce Professional Services.

With 1.3K+ certified Data Cloud consultants and 240+ implementations globally, we'll help you realise value quickly. Check out our guide to learn how.

FAQs

A database scheme is a blueprint that defines how your data warehouse organises and combines data. They indicate how data is stored in tables and how those tables relate to each other.

Star schemas consist of one central fact table that contains quantitative data, such as sales or revenue. This table is surrounded by dimension tables that have descriptive attributes. Star schemas are the simplest, most common type of schema in comparison to the snowflake schema, which is comparatively more complex and less widely used.

Extract Load Transform (ELT) is a data integration process used to prepare data for analysis. Unlike Extract, Transform, Load (ETL), ELT loads the data first before transforming it, leveraging the warehouse’s processing power to do so. ELT is better suited to cloud systems, whereas ETL is more common with legacy systems, where data needs to be transformed before it can efficiently be loaded.

A data mart is much smaller and more focused than a data warehouse. It’s essentially a specialised repository tailored to a specific department, like sales or marketing. It segregates important data for a particular team so they can access it and get targeted insights more easily than if they had to navigate the broader warehouse.