Does Power BI disrupt data governance?

Share on facebook
Facebook
Share on google
Google+
Share on twitter
Twitter
Share on linkedin
LinkedIn
Share on reddit
Reddit

In this blog post, Jen and Elizabeth explore the issues on data governance from the user lens.

It’s possible to view data lineage from a number of different lenses. From the IT perspective, like any other self-service tool, Power BI can be seen as a challenge when it comes to implementing a data governance strategy. Any data governance strategy requires a certain level of flexibility and access to a wider than average range of users in order to work to its maximum effectiveness. An IT department would do well to look at technologies such as the wonderful ASG Technologies, which has some mind-blowing data lineage and data quality features that are just far ahead of anything that the current author has seen in other technologies. Gartner agrees, noting ASG Technologies as a Leader in their current and previous quadrants. 

Users just need to see the results of data lineage; they are more interested in where the data comes from, and they may be involved in data governance as consumers and business domain experts rather than data and information governance expertise per se. This is where Power BI comes in. Because of this, Power BI has, in a way, disrupted the classic data governance model, but also gives us the opportunity to see data governance in a new, user-oriented light. 

Power BI requires the user to be able to strike a balance between a “self-service” system that allows end users to feel empowered to play around with data and make new discoveries, while also having procedures in place that ensure the ways that data is being accessed, used and shared appropriately, and GDPR regulations aren’t being broken. 

Power BI in itself imposes controls on consumption of its data, so operating within its parameters ensures a level of governance. However, the ‘freedom’ aspect of Power BI, while being one of its greatest draws for many people, comes with inevitable issues. Power BI also delivers information to its users very quickly, which in itself is a positive, but it can leave space for mistakes being missed, such as inconsistent or inaccurate data and multiple versions of the truth. 

Implementing a data governance strategy so users can see it in the Power BI system can help to overcome these issues. The best case scenario is to strike a balance between IT-managed, information-based problem solving and user-centred data exploration and experimentation, while leaving room to easily share ideas and collaborate.

There are in-built features in Power BI that will be helpful when implementing a data governance strategy. There are also plenty of things that users can do when using Power BI – whether that’s designing an engaging, interactive dashboard for customers, sharing reports with colleagues, or running calculations through their data for new insights – that will help make the process of putting a strategy in place easier, and will simplify upkeep once the strategy is in place. 

Firstly, Power BI has specific features which can support appropriate data governance. An example of this is the ‘My Workplace’ feature, which is a personal environment where the user can control how their information is shared. Power BI’s app Workplaces help users keep track of what information is certified, what can be made public and what is to remain private, and the extent to which public or semi-public information can be shared and circulated.

Users who are only viewing content in Power BI rather than creating, known as end users, do not need to have control over any of Power BI’s behind-the-scenes functions. This is where apps can be particularly useful, both for user convenience and good data governance. Apps can be used to allow end users to view data, without allowing them to see anything else on the creator’s end of the system that may be classified. It also prevents any user who isn’t in control of how the data is organised from changing the order or structure of the data or otherwise messing with it in any way that would disorganise it. 

On a similar note, and as is the case with any data platform, it is generally best for one individual per department to have oversight responsibility of the data being stored and used in Power BI. These individuals are usually referred to as data stewards. Anyone with data stewardship responsibilities will have a number of common roles and responsibilities, but when it comes to Power BI specifically responsibilities may include reviewing sources, fact-checking, double-checking reports before sharing, and determining who has access to what content.

Monitoring data use is a good data governance habit, and there are ways in which Power BI can make monitoring fairly simple. Usage metrics are an effective way of ensuring that GDPR compliance is being upheld, as well as general compliance with the company’s individual standards and policies – something that will be particularly important to keep in mind when there is a data governance strategy in place. Power BI lets users make use of built-in reports or live connections to the data source to review how data, dashboards and reports are being used. Usage auditing and keeping audit logs is also a good way of ensuring consistent GDPR compliance, and Power BI makes this easy by allowing it to be done in the Office 365 Security and Compliance Center.

The freedoms that Power BI allows may make the idea of implementing a data governance strategy more intimidating or confusing, but don’t let that put you off. When done right, an appropriate data governance strategy, paired with the autonomy and interactivity that Power BI allows, can be incredibly beneficial in a number of ways.

To read more about Power BI and Data Governance, here are some follow-up articles:

Data Lineage and Power BI: Why is it important?

The Hows, Whats and Whys of Data Governance

For more information, please get in touch and we can help.

Keep Exploring

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

© Data Relish. All rights reserved