Chapter 9: Tableau Server Best Practices

In this chapter, you will learn about best practices when publishing content to Tableau Server and refreshing data, and ways you can optimize the performance of your data and workbooks. Understanding these principles will help reduce the space used on the server, improve the refresh speed of your data, and increase the efficiency and performance of your published content. Following these guidelines will help you in discussions with your company's Tableau Server administrators and improve the user experience of those interacting with your published content.

Note

This chapter assumes some familiarity with developing calculations and data visualizations in Tableau Desktop. Additionally, this chapter is not meant to be an exhaustive list of all best practices but contains some of the most useful advice that the authors have learned through research and personal experience.

As you read through this chapter, it is important to remember...

Examining row-level security

After publishing a workbook to Tableau Server, all the users who are granted access to that workbook can view all the data within those views. But what if you want different users to have different access to the data being provided? That is where Row-Level Security (RLS) comes in. In Tableau, RLS allows the dashboard developer to restrict what data a certain user can see when interacting with a workbook or data source. This is accomplished by applying a filter that defines security policies that stipulates which "rows" of data a user can see when they sign in to Tableau Server. RLS helps provide better control over the data a person can see based on the restrictions placed on their username when they log in to Tableau Server.

Let's imagine RLS in practice. Pretend you work for a large U.S. insurance company. The company divides the country into four regions in which it provides insurance coverage: North, East, South, and West. The company...

Leveraging naming conventions

It seems like nobody thinks much about good naming conventions until after it's an issue. We've all looked for content or a file that we knew was located on Tableau Server or a shared drive but were unable to find it, at least not without a lot of work and some serious self-doubt. In fact, you were probably looking right at what you wanted and failed to recognize it because the name it was given meant absolutely nothing to you. However, if the name had been clear and easy to understand, it would have popped right out to you as your eyes hovered over it.

This is because names are most helpful when they make sense to as many users as possible. A well-thought-out naming convention helps reduce the effort required for your users to read, understand, and find content. A name should help frame your work in a way that describes it and connects it to other related content. To maximize the end user experience on Tableau Server, we recommend taking...

Utilizing published data sources

In Chapter 1, What is Tableau Server?, you learned that a published data source is a data source that has been created and published to Tableau Server. Once on the server, a single published data source can be used by multiple workbooks. Published data sources are useful when your organization has important metrics that will be used in multiple reports or visualizations housed in different workbooks. If each workbook has its own data source, it not only takes up more room on Tableau Server and consumes more server resources when each extract refreshes, but it can also cause data variances if the workbooks are not refreshed at the same time. For example, if one workbook refreshes on Monday and another refreshes on Wednesday, but both workbooks visualize the same business metric, then end users are likely to be confused and contact you with questions when they notice differences in the data presented by the two workbooks. To avoid this, you can create...

Refreshing your data

As you learned in Chapter 1, What is Tableau Server?, an extract is a subset of your data that has been optimized for Tableau using the Tableau Data Engine. When you go to publish a workbook to Tableau Server, data extracts and live connections will be embedded in the workbook. Unlike published data sources, embedded data sources can only be used in the workbook in which they are published.

Choosing a data connection type

If you choose to embed a data extract in your workbook, and you publish it to Tableau Server, you can create extract refresh schedules so that the data will refresh automatically according to the selected date and time. You can even create multiple refresh schedules for a single workbook or data source, such as every month on the first day of the month and every Wednesday.

Note

You can create a data extract in a workbook and then publish only the data source to Tableau Server to create a published data source. Just like any data extract...

Improving performance

When end users are interacting with content on Tableau Server, such as a dashboard, it is important to consider the user experience, much like you would when building a website. This means considering how clearly the information is presented, how easy the tool is to navigate, and how quickly the information can load with each click.

According to UX Planet (https://uxplanet.org/how-page-speed-affects-web-user-experience-83b6d6b1d7d7), 0.1 to 1 second is the ideal load time for a web page, and 10 seconds is the longest a user is willing to wait before performing other tasks or leaving the website. Other sources suggest that you only have 1-2 seconds before a large portion of the audience abandons the page. Although your business partners or end users may have more patience waiting for a dashboard to load because of their need for the specific data than the average person browsing the internet, it is still important to keep user experience in mind when developing...

Summary

In this chapter, you learned how to implement RLS to manage who can access specific data within a data source or workbook. You also learned our recommendations to create meaningful naming conventions, to select the appropriate data connection type for each situation, and to understand some of the considerations when selecting a data extract refresh schedule. Finally, you learned some of our recommendations for improving the performance of your data sources and workbooks and have a few suggested resources for further learning.

In the next and final chapter, you will find additional resources to continue learning about data visualization and Tableau.