Take control of your Tableau analytics with data governance and content management

Brian Franklin

Brian Franklin

Chaos is the natural state

On a recent trip I visited the Hoover Dam, and it occurred to me that the Colorado River and its system of dams are analogous to the large amounts of data organizations and the data governance programs put in place to maintain and control that mass of data. Without the dam, generating electricity from the river would be inconsistent and unreliable. Similarly, a data governance program and content management bring order and security to large amounts of data.

data governance

As part of a modern BI solution, Tableau Desktop allows you to rapidly analyze data through insightful visualizations. Tableau works well with data from traditional governed sources, such as data warehouses, as well as non-traditional sources external to your enterprise IT infrastructure. If your organization does not currently have a formal data governance program overseeing these sources, or you are not getting the full benefit from your program, considering reading Eniko Tucker’s post, “Creating a Successful Data Governance Program.”

As the usage of Tableau permeates through an organization, the first questions that will be asked of your Tableau analytics are, “Where did the data come from?” and “How did you calculate the values?” Data governance and proper content management controls will lend credibility and confidence to the reports and dashboards you produce using Tableau. People and processes need to be in place to provide oversight, security, and protocols for publishing. This is not to say that every visualization you create in Tableau should go through review, but rather that the level of governance should be dictated by the degree of risk and visibility associated with the information.

Challenges

  • Accuracy: The correctness of your data and metrics is paramount to the quality of your Tableau reports.
  • Dependability: Your analysis is deemed dependable if it is repeatable, reliable, and available to others.
  • Maintainability: Tableau data sources and workbooks should be centralized, organized, and backed up to ensure they are available when needed.
  • Security: Access to your Tableau resources should be restricted to authorized users.

Take control

The only thing worse than no business intelligence is bad business intelligence. Your data is coming from different sources, so you will want to assign ownership for your different data entities to the appropriate people to act as data stewards, guiding usage of your data. Vital to this role is ensuring the data is accurate and consistent across sources. As part of this process, your Tableau metrics should be reviewed to ensure they conform to standards. Gaining insights into your data to drive business decisions is at the core of Tableau, so you want to be certain that your calculations are accurate.

Once you have created meaningful visualizations with Tableau, you will want to reuse them, and others may want to leverage your work for their own purposes. Centralizing your data sources and workbooks in Tableau Server ensures that your work is available whenever and wherever you need to access it. Connections to your enterprise data can be defined and managed within Tableau by IT resources and made available to authorized users, adding consistency and security. Workbooks should be grouped together based on functional areas, subject matter, or any other method of organization that is appropriate for your needs.

Tableau Server offers robust security to help keep your data and workbooks secure. Users, logging in through secured authentication, can be organized into a variety of segmented permission groups: project-based, workbook-based, data source, or even user groups. Data itself can have user permissions, only allowing users to see the data rows that they are granted permissions to. A diagram below provides a visual representation of permissioning on Tableau Server.

Tableau security heirarchyGovernance is critical to success

You will likely see rapid adoption of Tableau analytics in your organization as it demonstrates its ability to improve decision-making. Integrating Tableau into your data governance strategy is vital to ensuring its accuracy, dependability, maintainability, and security. If you have any questions, feel free to reach out to Slalom’s network of experts.

About Brian Franklin
Brian Franklin is a Solution Architect with Slalom Consulting in Dallas, Texas with a background in finance and over 15 years of technical experience, specializing in designing and developing insightful business intelligence solutions. His combined business experience and technical expertise have contributed greatly to many successful projects.

5 Responses to Take control of your Tableau analytics with data governance and content management

  1. Reblogged this on BI Tools and Technology and commented:
    This is great because it addresses the important reasons for having data governance. Because Tableau can essentially be used by the masses, it makes it much more likely that data can be grabbed and interepretted and defined by the user, not the enterprise. Even though a tool makes it easier for end-users to use the data, doesn’t mean the data should no longer be properly governed. It is important to remember Tableau is a tool used to visualize and analyze the data as it has been defined, not to recreate data depending on each user’s interpretation.

    • I completely agree. Different business users often have separate needs related to the same data. Many hours have been wasted by business customers and IT staff trying to chase down differences found between reports created by different people. This need to maintain varying perspectives of data highlights the necessity for effective governance to help ensure the accuracy and reliability of workbooks across the enterprise.

  2. Lars Kertabski says:

    We’ve run into governance problems at the past 4 companies I’ve worked at with Tableau. One of the biggest problems is Tableau sales and the company representatives that visit with us suggest there is no issue with data governance. Then we read blogs like this that suggest that the basic principle behind Tableau being “Data for the People” can introduce the spreadmart and governance issues that actually can and do appear. In the end it is all up to the company ultimately, but it is extremely difficult when Tableau is being sold into your organization at multiple levels and departments, sometimes with very little IT knowledge, and different sales reps. When sales reps feel you’re going to take them to task, extend the buying timeframe, introduce IT, and question them hard on governance, how best to do it and so on, they ‘land and expand’ to other departments with budget and sometimes never tell anyone. Scaling a governed solution when you realize that you walk into a high level meeting with 2-3 departments being represented and people have 3 versions of the same report is scary. I really wish Tableau was more forthcoming like this blog about being aware of governance from the very first license sold. It is very hard when again, many times it is sold without IT leaders barely knowing. Only after several depts and managers and ultimately a Tableau server comes in does IT really start to get noticed.

    These things need to be defined by the enterprise. With a tool that can quickly spread, particularly with desktop only, and load of extracts and excel files, enterprises need to understand how to control this before license #1 even lands.

    • Tableau awareness is rising rapidly, and IT organizations and the business customers they support should work together to prepare themselves for proper enterprise-wide implementation of Tableau and other self-service tools. While a data governance strategy is driven by business, the IT organization must be an active facilitator to implement an effective plan. As my post states, in the absence of order, chaos will reign. The comments above reinforce the need for a enterprise data governance strategy that covers Tableau and other self-service tools.

  3. Pingback: How to improve the performance of your Tableau dashboards | The Slalom Blog

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: