Most of us are familiar with the cliché apples and oranges, which aptly describes the difficulty of trying to compare two things that really aren't comparable. But it's even more complicated when you're considering a data governance tool.
Organizations turn to data governance software to help ensure that data delivers business value. Selecting a tool can become challenging, however, when you're looking at features that are attained by wildly different tools and vendor philosophies across several distinct product categories. In some cases, several vendors may provide the same functionality in their products -- but through very different means.
As a result, you must be very clear on what you want to do with the data governance tools, as well as the capabilities that you need. In addition, evaluation and selection depends not just on features and functionality, but also on how you will add business value with these tools. For example, if you have a requirement for a glossary, make sure it will actually be used before you acquire a data governance tool with that function. Regardless of how many bells and whistles a product has, if you can't demonstrate visible value in a short period of time, getting continued funding and executive support for your governance initiative will prove difficult.
Prioritizing the features of data governance tools
To help you decide which software features are most important to your organization, the following functional areas are listed by most likely to least likely in terms of meeting requirements for new or recently launched data governance programs:
Information and data artifacts. All data governance programs must manage key artifacts, such as data elements, data models and glossaries, which are crucial to supporting data consistency. This is usually the first functional area addressed by governance programs. Even if it isn't the first, it's a mandatory function.
When you're researching governance tools, look for software that can identify and track common create, read, update and delete activities for data elements. Also, look for a tool that has data discovery capabilities that enable you to scan and identify data elements, plus data and metadata values. Another useful feature is the ability to manage relationships between data elements through hierarchies or taxonomies. Lastly, look for a data governance tool that enables you to classify data based on its use or relevance -- for example, whether it's regulated, life-limited or used as a reference item or a fact.
Data management elements. Remember that data governance is oversight of data management, which means you'll need to track data management activities and standards. This is the second area of functionality that will generally be high on the list of priorities for organizations. Look for governance tools that:
- Provide data quality management, including quality rules, profiling and reporting.
- Provide data domain and master data lifecycles, reference data.
- Perform data movement, data lineage views and positioning.
- Address process discovery and reporting.
- Capture change history of modifications made to metadata objects, the business glossary and lineage models.
- Support cascade changes, which enable you to remove a data element in an upstream source and have it auto-removed from downstream sources.
- Expose metadata, lineage and business glossary data for use in third-party products and custom applications.
- Maintain a drill-up and drill-down display of lineage information.
- Print and exchange visual representations of data lineage.
Other artifacts. These are documents or other text-based materials, including manuals, charters, product specifications, digital media and emails, that are stored permanently for subsequent use or review. While most governance programs address structured data stored in rows and columns within database tables, many documents also require oversight. This set of functionality is rarely addressed early in a data governance program, but almost all organizations that develop a robust, operational program will need to address unstructured data and document management.
Look for products that provide metadata support for document classification and document lifecycle management. Also, look for a tool that provides the ability to create, read, update and delete, as well as eliminate redundant, obsolete and trivial data.
Governance operations. Again, this isn't usually the first functionality sought out in a tool, but it soon becomes an obvious need, and will mean looking for a tool that can do some or all of the following:
- Assign and manage governance roles and responsibilities (e.g., who the data stewards are).
- Set up the various structures in a governance program, such as working groups, councils and forums, and track the activities of their members.
- Define and monitor service-level agreements, issues and activity statuses.
- Enable change approvals (e.g., policies and standards), escalations and audits.
Workflow management. As stated earlier, collaboration is a core, necessary theme for the successful operation of data. For example, getting a policy through many cycles of review and approval can be onerous without the use of workflow and deliberate collaboration. Workflow can be used for accelerating and operating master data management processes, overseeing data quality or managing and escalating data governance issues. It's also effective for building and sustaining the data governance program -- for example, when the members of a governance council need to work together to approve new documents and processes. Workflow capabilities also aid in cross-team coordination and efficient decision making.
These types of tools generally do the following:
- Define and manage workflow for collaboration and cooperation among stakeholders.
- Manage permission levels for users to make changes to glossary definitions, standards, policies or other components of data governance.
- Support separate approval processes and workflows for custom data attributes, policy annotations, business definitions and comments.
- Track the progress of work as part of data governance activities.
Business alignment. Data governance is more of a business initiative than a technology project, so you need to be aligned with the business -- especially if senior management doesn't see the value of effective governance upfront. But while business alignment is a critical success factor, it isn't usually the first functionality organizations look for in a data governance tool. In fact, most of the time a spreadsheet will suffice.
Some of the alignment features listed below may seem similar to data management functionality, and they are often found in the same tools, but they're more for connecting data governance processes to other aspects of the business. Such features include:
- Documenting hierarchies of strategic plans or business processes.
- Monitoring business strategies and plans.
- Calculating the business value of data or contributions to achieving business goals.
- Tracking metrics of progress and effectiveness with an accompanying scorecard.
Narrowing the list of tools to consider
Once you've defined the required functionality, you can decide which category -- or categories -- of data governance tools to explore. To meet all your needs, you might have to buy a combination of tools from different categories: traditional data management tools, data quality software and governance program and policy management platforms.
There are several other selection criteria that you should also consider to help you differentiate among data governance products.
First, how are they offered? Does your organization prefer tools that are deployed on premises or cloud-based tools? Do you need mobile capabilities? Also, what kinds of connectors to existing software in your enterprise do you need? Many vendors offer connectors, but they vary widely in style and completeness.
Are you looking to govern big data environments? Some governance tools work well with big data, while others don't. And what kind of reporting and dashboard features do products offer? The ability of tools to report on the content of their own repositories can be surprisingly lacking.
Get preliminary pricing, as many of the tools are offered as options to core tool sets. You may get the functionality you want, but it may come at an eye-watering price. Also, examine the vendor's training and support materials. Many vendors have adapted older tools to data governance, but they haven't updated their training materials or ability to support data governance.
Advice for building a data governance framework
A strategic approach for governing your organization's data
The need for effective data governance models