What is Data Integration?

John Spacey, updated on July 26, 2023

Data integration is the practice of combining data from a variety of sources. This is typically done to support processes, user interfaces and reporting that require data from heterogeneous systems.

It is common for organizations to have their data spread amongst systems that may number in the thousands, each with a distinct data model. It is equally common for critical business data to end up in documents, emails, text files and other unstructured data repositories. As such, new projects may be unaware of existing data and replicate it leading to duplicated effort, data quality issues and needlessly complex business processes and technology stacks.

Data is often isolated along political lines in an organization. In other words, competing managers may not cooperate and tend to develop duplicated data repositories. As a result of these issues, data integration is a significant challenge for many organizations.

Many data integrations are intended to meet the requirements of a single project and may be designed as point-to-point transfers of data. With time these integrations can grow to number in the thousands and are often difficult to maintain. In other cases, data integrations are architected to be reusable services with a common data model. This is typically a more desirable solution but can represent a major initiative that is commonly underestimated by project teams.

Overview: Data Integration
Type	Information Technology
Definition	Services that provide access to data to support technologies such as automation, user interfaces and reporting.
Examples	A hospital that accesses an emergency patient's national health records to see if they're allergic to anything. A sales report that uses data from a customer relationship management, billing platform and marketing automation system. An ecommerce engine that displays products from thousands of partners in a single user interface.
Related Concepts	Process Integration