Tableau continues to be a Leader within Gartner’s Magic Quadrant Business Intelligence and Analytics Platforms in ability to execute, placing for the third time in this category in February, 2015. For good reasons, Tableau is a buzzword within many mid-to-large size firms, as addressed in my previous blog post, Tableau: Beyond the Buzzword – Part 1. Though Tableau may not be the answer to all of your BI and Analytic needs, Tableau can be a powerful tool to create value within an organization. By implementing the following best practices value can be increased, thereby maximizing the return on investment.
Best Practice #1: Utilize Calculations on an As-Needed Basis
One of the greatest feature of Tableau is its ability to create and manipulate measures within the application and apply them to drive powerful visualizations. The Calculated Field editor provides a simple interface for calculation creation, including suggestions and descriptions of calculations available. For example, some types of calculated fields that Tableau can create includes math operations, logic statements, aggregations, string manipulations, and data formulas.
New in Tableau 9, Level of Detail calculations allow modifications to the granularity or aggregation of the data within Tableau. These can be useful to create meaningful analyses when aggregated data may be misleading or confusing. For more information about Level of Detail calculations, see Tableau’s Whitepaper.
These custom calculations should be utilized with caution for two main reasons. First, multiple analysts creating various calculations could lead to vastly different results and conclusions from the dataset if measure creation is not governed by organizational standards. For example, within an organization, two analysts determine a calculation for “Sales Amount.” If there are any slight variances between the measures used by both analysts, this could lead to different conclusions by the analysts, causing confusion within the organization.
Second, table calculations increase load time because they are performed within the application. Utilizing multiple complex calculations on large sets of data can negatively impact performance of dashboards and worksheets. Instead, promote calculations that process in the database to improve performance. A data governance policy would designate an owner of the commonly used calculation so that its use is standardized between the analysts, reducing confusion and increasing trust in the analysts and Tableau.
Best Practice #2: Choose the Best Graphic to Meaningfully Provide Analytic Insight
One of the most popular and visually appealing features of Tableau is the map view and geocoding features. Tableau’s ability to assign colors and sizes to various countries, states, and cities makes this a powerful reporting option. These features can often drive users to jump right in to using the map views without understanding the granularity of your data, which can result in a misrepresentation in reporting. An important practice while using map views is the attention to the level of detail. In example 1.1 below, Illinois stores are seen as being profitable while Indiana stores are not. This could drive users to recommend focusing reform efforts for Indiana stores. However in example 1.2, if we add city to the granularity then you can see that the store that needs the most attention is actually located in Illinois.
These examples demonstrate the importance of the level of detail when telling the overall story of your data using map views.
Best Practice #3: Enable Dashboards to Tell “The Whole Story”
Tableau Dashboards are a great way to share analysis with others. Creating dashboards is an art, not a science, so there are many things to consider while utilizing these powerful tools. The following three best practices can be implemented in order to enable dashboards to tell “the whole story.”
- First, determine the purpose and audience of the dashboard and keep this in mind for the entire development process. This is essential, as trying to create one dashboard for multiple use cases can cloud the message. Defining a clear purpose at the start of the project will lead to better results.
- Second, ensure that the dashboard is a resource users can refer to when questions arise. Tableau’s White Paper, 6 Best Practices for Creating Effective Dashboards, explains that it is essential to have all of the data available and relevant to the decisions being made. If analysts do not have all of the data available in the dashboard, they will be forced to extract the data and create their own analyses elsewhere, which is ineffective and inefficient.
- Third, confirm that the dashboard is easy to use and understand. Tableau provides many specific suggestions for achieving this, but the key message is to create dashboards that can be understood quickly, ideally portraying the primary story in five seconds or less, have labels and instructions that prompt interaction, and are visually appealing with limited distractions. For more information about specific strategies, visit Tableau’s Knowledge Base.
Tableau Dashboards can be a powerful data visualization and analysis tool when implemented effectively. The following example demonstrates the power of filtering to drill down into your data. You can see some of the many ways that the data in your dashboard can change based on certain filters that are applied.
In summary, these three best practices can be applied to a Tableau implementation in order to maximize return on investment:
- Implement data governance policies in order to standardize metrics and calculations
- Choose graphics that accurately report the level of detail required for strategic decision making
- Utilize dashboards to drive powerful analytical insight
Overall, Tableau’s value can be most truly realized when organizational standards and best practices are put into place alongside a uniform set of data governance rules. Tableau’s analytical tools, including calculations, graphics, and dashboards can be used to unlock the true potential of the data within an organization.