A data management platform (DMP), also referred to as a unified data management platform (UDMP), is a centralized system for collecting and analyzing large sets of data originating from disparate sources.
A DMP creates a combined development and delivery environment that provides users with consistent, accurate and timely data. At its simplest, a data management platform could be a NoSQL database management system that imports data from many systems and enables marketers and publishers to view the data in a consistent manner. A high-end DMP might combine data management technologies and data analytics tools into a single software suite with an intuitive and easy-to-navigate executive dashboard.
A key role of a data management platform is to collect structured and unstructured data from a range of internal and external sources, and to then integrate and store that data. These platforms also analyze and organize data to provide insight to data-driven parts of the business, such as marketing and advertising campaigns.
Data incorporated into a data management platform can be first-party data -- coming from an organization's own applications, systems, websites and products -- as well as second-party data from partners and other associates. In addition, DMPs use third-party data to fill in holes in a company's own data and partner data.
The roster of data management platform vendors focused on marketing and publisher clients is long, with some bigger name vendors getting into the technology more recently through acquisitions. These offerings include Adform DMP, Adobe Audience Manager, KBM Group's Zipline, Lotame, MediaMath DMP, Neustar Identity DMP, Nielsen eXelate, Oracle BlueKai, Salesforce DMP (formerly Krux) and Turn Digital Hub for Marketers.
Data protection morphing into data management
Beginning in 2016, data protection vendors began extending their offerings with broader data management capabilities. Veritas laid out a plan to build a data management platform that combines its Veritas Resiliency Platform, which orchestrates the recovery of virtual machines in multivendor hybrid clouds; Information Map, which aggregates data and presents it visually; its Velocity copy data management system; its newly launched policy-based storage management component called Access; and HyperScale for OpenStack, the compute piece of Veritas' broader platform.
Commvault is doing something similar, expanding into data management with its Data Platform data protection engine, which has an integrated indexing system. The company plans to open Data Platform to third-party application developers to build an ecosystem of data management capabilities on the backup software. Commvault is also moving into managing data in the cloud.
Products referred to as data management platforms are also emerging in other parts of data management. Komprise, for instance, analyzes file storage and uses policy-based automation to move inactive file data to object-based storage targets that run locally or in the cloud.
Rubrik is aiming to offer a full-scale cloud data management platform with orchestration and reporting. Rubrik Cloud Data Management combines backup, recovery, replication, search, analytics, archival and copy data management on a single platform. Rubrik Alta protects workloads natively in Amazon Web Services, Microsoft Azure and private clouds.
The Cohesity DataPlatform includes a scale-out cluster of tiered secondary storage. Like Rubrik, Cohesity also refers to its architecture as converged or hyper-converged secondary storage.
Why you use a data management platform
Traditional data management platforms are most often associated with products and development projects designed to help marketers, publishers and ad agencies. They are used to turn offline data, as well as data from online sources, such as sales and CRM systems, web analytics and mobile channels, into information that can be used to support media purchases and ad placement. However, with data protection systems expanding into data management, and as big data has taken hold throughout business, DMP capabilities are being applied to those systems to manage data and facilitate data analysis and activation.
In the marketing context, DMPs give businesses control over their customer data to make better decisions about marketing campaigns and media buys. Data management platforms aggregate customer data from various sources and then analyze, organize and segment it into different customer or "audience" types, looking at factors such as location, income, browsing behavior and past purchases.
Advanced algorithms use the segmented audience data to identify potential customers -- with profiles similar to an organization's existing customers -- that are likely prospects for marketing campaigns. The data is then used to tailor marketing and ad campaigns to better target those prospective customers. Once campaigns have been run, a DMP assesses their performance across different audience segments and media channels, and helps refine marketing and advertising efforts.
In effect, a data management platform takes a large amount of data and makes it actionable. It provides a business with a deeper understanding of its customers and gives data-driven direction to media buying decisions, campaign targeting and ad optimization.
Demand-side platform vs. data management platform
Still in the marketing context, data management platforms work with demand-side platforms (DSPs). These platforms are real-time bidding systems that connect media buyers with multiple ad and data exchanges, and automate programmatic advertising bids and purchases in real time.
A DMP feeds segmented audience data into a DSP, which uses that data to target ads at the right audience. The DMP analyzes the results of that targeting, providing feedback to the DSP on which audiences are responding well and which aren't. The demand-side platform then uses that information to further optimize its bidding and ad targeting efforts.
Open source data management platforms
The EDB Postgres Platform from EnterpriseDB is an open source operational data management platform that brings together multiple components to manage structured and unstructured data in a federated model. It's based on PostgreSQL, an open source relational database management system and makes integration across different databases possible, enabling users to combine unstructured data with structured, transactional data.
On the big data front, vendors offer advanced data management capabilities combined with open source Hadoop distributions to manage and analyze large volumes of data. SAS Data Management, for instance, is a suite of tools that provide access to data on legacy systems and Hadoop. It also transforms, integrates, cleanses, delivers and governs data on premises and in the cloud. IBM BigInsights combines enterprise capabilities with Hadoop components into a single platform. It lets users manage and analyze large volumes of structured and unstructured data.
The future of DMPs
New technologies are changing the data management platform landscape. Internet of things applications and inexpensive sensors are producing more data than ever for DMPs to manage. Machine learning technology is being integrated into data management systems and used to move data from management platforms to its destination faster than ever.
Longer term, "fast data" architecture will make it possible to quickly process large volumes of data, and simultaneously data management is evolving to enable businesses to analyze information instantly. New streaming data products are being built to handle large amounts of complex, diverse and often unstructured data. These streaming platforms work with data in motion, evaluating it the instant it arrives. While such systems aren't yet financially within reach of most companies, they signal the future of data management platform technology.