Making the Data Hub Work: Five Steps to Merge Data Management and Governance
by Kevin Petrie | September 24, 2020
The first and second blogs of this series examined the vision, tradeoffs and evaluation criteria of the data hub. Suppose you weighed the pros and cons, and decided your environment needs this type of platform. What now?
To recap, business managers and BI analysts use the data hub to play a hands-on role in data management and governance. They apply their domain expertise to integrate, transform and explore data related to customers, supply chains, employees and other entities. Selected and implemented well, the data hub eliminates silos and spreads the glue of governance, master data management (MDM) and data quality.
However, enterprises must balance competing demands to achieve this vision of convergence. They must reassign work but still increase productivity; change tools but maintain governance controls; and standardize but still handle custom work. Enterprises that strike the right balance pay close attention to people and process, more than shiny new tool features. They let the business drive the technology rather than vice versa.
So what does an effective data hub strategy look like? Data leaders can help ensure success by taking the following five measures.1. Enlist the Chief Data Officer as your sponsor
The CDO, now present at most enterprises, must evangelize the need and vision for the data hub across the enterprise. Their charter of aligning IT, business and data strategies – and balancing distributed self-service with central governance – matches the core objective of a data hub. CDOs also can be strong advocates because they understand the need for creative thinking, and new tools, to streamline data consumption. Without strong CDO sponsorship, business users will treat data hub initiatives as optional, fail to convert fully, and thereby reduce their productivity.
2. Align with existing data modernization and consolidation efforts
Effective data hubs must integrate easily with both legacy systems and the modern cloud-based platforms to which enterprises are migrating. A well-integrated data hub helps business managers and BI analysts derive master data that presents a comprehensive view of the business. It also simplifies the role of data stewards by eliminating the risk of siloed master data. Data hubs that integrate with only parts of the business, on the other hand, can contribute to silos that undermine the productivity of both data consumers and data stewards.
3. Consider a federated analyst organization
Business managers and BI analysts must embrace the data hub to generate its intended results in terms of productivity, governance and standardization. Consider creating a dotted-line organizational model in which IT trains business users and mandates their usage of standard practices, templates and data models. Give looser rein to data scientists and developers to experiment with custom scripting and innovation. Once their new data models are ready to go into production, collaborate with those teams to integrate the appropriate governance controls. Without some method of federating activities, analysts go different directions and undermine the data hub’s control.
4. Elevate the data steward role
Data stewards benefit greatly from data hub initiatives because they reassign the tactical work of master data creation, curation, etc. to business managers and BI analysts – assuming they have the support of CDO sponsorship and a federated org structure. But data stewards also need to elevate their role from firefighter to policy-maker and educator. They should guide data consumers about how to manage data in governed ways. If they don’t, the data hub loses its most informed champions.
5. Chalk up a quick win
As with all technology initiatives, the data hub will best earn and retain executive support by demonstrating quick value. Consider starting with a bite-sized project, focusing on a departmental need, narrow data set, easily-structured model, etc. Once you set and achieve an easily measured goal for that project, you can secure the necessary goodwill and budget to extend your data hub footprint to a sequence of larger and more complex data sets. Don’t try to boil the ocean.
Ultimately data governance and therefore data hubs comprise a living, breathing strategy that needs careful attention. Enterprises that enlist their CDO, align with data modernization efforts, train data consumers and empower data stewards – taking one achievable project at a time – can roll out sustainable data hub initiatives. They can reconcile the tradeoffs of a new technology and strike the right long-term balance for successful citizen stewardship.
Below, a suggested webinar that examines the problem of data silos, the proposed solution of a data hub, and its intended benefits. It assesses the inherent tradeoffs of converged platforms, key evaluation criteria and finally guiding principles for successful data hub initiatives.