The first and most important step towards digital maturity is creating a watertight roadmap that guides the organisation from its current state through each stage of the journey towards developing a fully-functional and optimised data solution. Data governance, the most fundamental part of that process, should be central to an organisation’s planning right from the very start.
Engage at the Enterprise Level
Data governance requires enterprise-level buy-in, rather than just evaluating data assets and requirements at a departmental level. This means that the IT department cannot operate a successful data optimisation project without full approval and engagement with the board, through C-suite executives, down to departmental directors and managers. Each step along the way needs to be mapped, resources in each individual department need to be evaluated, and project planners need to spend time assessing which data is of value, which is most critical, and how it will tie in with data from other departments as part of the holistic enterprise data optimisation program.
However, developing such a map is challenging, and it is challenging for the same reason that data governance itself is challenging: Data is often stored across many different systems, from traditional, highly structured data warehouses to next-generation cloud-based sources and everything in between. Sources are often stored in different physical locations and in different formats. This means that to even begin to catalogue the available data, data governance stakeholders need a way to gain visibility across all of the different data sources in the organisation.
The Needs for Enterprise Data Governance
Organisations, then, need intelligence about each individual data point – who owns it, what its lineage is, how it relates to other data points, and so forth. To provide proper governance over data, organisations need insights into the data itself. They need a complete view of the data’s history.
A data governance plan needs to define who is able to access which data point, the speed at which data needs to be presented to each user, and the level of quality that the data is required to demonstrate.
Data Virtualisation: A platform That Enables Enterprise Data Governance
From a technical perspective, data governance, as described above, is possible, to the extent that an organisation’s metadata – descriptive data about the data itself, such as where the data is stored, and who is allowed to access it – can be made centrally discoverable within the organisation. Data virtualisation is a technology that makes this possible.
Data virtualisation uses a novel approach to integrating and managing data. Rather than moving data from one location to another, data virtualisation provides real-time views of the data across its original locations. Data virtualisation is able to perform this function because it contains the critical metadata necessary for accessing each data source, and it is implemented as an enterprise layer between all sources and all consumers. The data virtualisation layer itself contains no actual data; however, the metadata enables it to easily connect each user with each required data point, as needed, and on-demand.
By serving as the universal source of metadata across an organisation, data virtualisation also provides a solid foundation for data governance. From a single interface, stakeholders can implement data governance controls across the entire organisation. Because of the metadata at the heart of data virtualisation, not only does data virtualisation provide a unified view across all of the disparate sources in the organisation, but the view is provided securely, in that the only individuals who can access each view are those with the necessary rights; for all other users, the data is dynamically masked. Finally, with data virtualisation, systems administrators can easily see who has accessed which data, and for how long. All of this functionality is of prime importance when governing data at the enterprise level, across multiple departments and multiple types of users.
The Critical First Step
Your first step in the data maturity journey, in conjunction with developing a strong data governance plan, is to implement a data virtualisation layer. This enables organisations to sort out all data and seamlessly present it, in real-time, to users who need it.
Denodo is a leading data virtualization platform, which is the core technology enabling modern data integration and data management solutions.