This Tableau Checklist & Guide is designed to help you build efficient, high-performance dashboards while following best practices. It ensures that your dashboards are structured, optimized, and easy to use. It would also help you fine tune your dashboard for a better performance.
What’s in This Document?
This document consists of two main sections:
The Checklist
A structured list of best practices for building Tableau dashboards.
Covers key areas such as data source, visualizations, dashboard design and performance.
Helps you stay on track and avoid common mistakes.
The Guide
Detailed explanations for each checklist item.
Provides insights into why each step matters and how to implement it effectively.
Helps you troubleshoot potential issues and improve dashboard performance.
Additionally, the outline on the left allows you to quickly navigate through different sections of the document.
How to Use This Document?
Navigate Using the Outline
Use the document outline (on the left) to jump to different sections quickly.
Follow the Checklist (Feature available in downloadable pdf)
Use the checklist as a step-by-step guide while building your dashboards.
Mark off items as you complete them to ensure best practices are followed.
Refer to the Guide for More Details (Feature available in downloadable pdf)
If you need clarification on a checklist item, refer to the guide for a deeper explanation.
You can get to the guide for that particular item by clicking on the point and following the bookmark. This section provides additional context, best practices, and optimization tips.
Make a Copy & Use It for Your Work (Feature available in downloadable pdf)
Duplicate this document and use it as a reference while working on your dashboards.
Ensuring each item is checked off will help improve performance, usability, and consistency.
Tableau Checklist
1. Data Model and Source Optimization
Limit data size by applying data source filters
Use Hyper Extracts
Prefer native drivers over ODBC/JDBC for better query performance
Avoid custom SQL unless necessary; optimize at the database level
Utilize Tableau’s logical layer relationships instead of joins
2. Calculations and Aggregations
Materialize String calculations
Make groups with CASE statements
Test the most frequent outcomes first in conditional statements
Reduce complexity (Nested calculations and redundant logic)
Reduce Granularity of LOD or Tableau Calculations
Use IN instead of OR
Use MIN or MAX instead of ATTR
Minimize slow functions like COUNTD
Materialize date conversions and calculations
Even basic statistical calculations can become expensive at scale. This walkthrough explains how correlation and other statistical measures are computed in Tableau and how to use them efficiently.
3. Filters and Parameters
Minimize quick filters
Avoid filters on high-cardinality fields
Use context filters for high-cardinality filtering
Replace unnecessary filters with parameters
Use the apply button
Set your Date filters to continuous
Ranges are better than itemized filters
Use Action filters to reduce query load across data sources
4. Worksheet and Visual Optimization
Limit mark counts
Visuals are better than texts
Consider using text formatting instead of shapes
Reduce image size
Use transparent background PNGs instead of JPGs
Avoid overuse of polygon marks
Use the page shelf sparingly
5. Dashboard Design
Use fixed-size dashboards
Reduce the number of views per dashboard
Drill into detail, rather than loading all data in one load
Hide worksheets not in use, until needed.
Clean-up containers.
Break up complex dashboards
6. Performance Analysis
Use Performance Recorder to identify slow queries, calculations, and visualizations.
Optimize the slowest views(impact of filters, data source, and calculation efficiency)
Warm up server cache by accessing dashboards after publishing.
7. General Tips
Avoid overcomplicating dashboards with unnecessary features or excessive interactivity.
Leverage Tableau Prep for data cleansing, reshaping, aggregations.
Regularly clean up workbooks by removing unused fields, sheets, calculations.
Tableau Guide
Understanding the performance of tableau dashboards:
Tableau Dashboards have 4 Core elements:
Data
Calculations
Worksheets
Dashboard Layout
When your end-user decides to load a dashboard:
Tableau writes queries processed by your data source. Tableau is, at its core, a visual query generator. Much of the work Tableau does is translating your designs into queries your underlying data source can understand. Those query results are then used by Tableau to perform additional calculations on the data (such as window calculations) as well as the internal calculations Tableau may need for your dashboard.
Your worksheets/visualizations are rendered based on the query results and calculations before being placed in their final layout. This process, when broken down, can be summarized into just four components that impact performance:
Query Time
Calculation Time
Rendering Time
Layout Computation
Some ground things which are true in all dashboards built in tableau:
If data source is slow, then it will be slow.
Fix: To use tableau extracts
If it slow on Tableau desktop, then it will be slow in Tableau server.
Newer versions are generally better
Data Model and Source Optimization
Tableau is, at its core, a visual query generator. It takes your designs and translates them into multiple queries to your data source. Well-designed data sources can give your production dashboard a fast foundation.
Tableau offloads as much of the calculation and aggregation work to the underlying data source as possible. This gives you a lot of flexibility in handling the processing through data source optimization and is another reason why Hyper Extracts are such an important part of your toolkit for performance.
Data sources in Tableau Tableau has an extensive ability to connect to data across many different platforms. For the purposes of talking about performance, we’re going to broadly group these connections to the following types:
File-based data sources – such as Excel and CSV;
Relational database data sources – such as Snowflake, Oracle, Teradata and SQL Server, as well as specialized analytic appliances such as HP Vertica, IBM Netezza, etc;
OLAP data sources – such as Microsoft Analysis Services and Oracle Essbase;
No SQL data sources – such as Hadoop;
Cloud-based data sources – such as Salesforce, Google, etc.
For scenarios involving external market or financial data, connectors like FactSet require careful configuration to avoid performance bottlenecks. This example shows how to set up and optimize a Tableau Web Data Connector for reliable, scalable use.
For most connections, Tableau can either query the data live or create a Hyper Data Extract (a .hyper file). When querying live, that means that Tableau will generate queries to the underlying data source as you build visualizations and dashboards. Since data sources come in all shapes and sizes, slow and fast, the performance of your data source will have a huge impact on your dashboard. If your data source is slow, the Tableau Hyper Extract is your secret weapon for performance, the majority of environments would benefit from using extracts.
Tableau Consultants – Enterprise-grade services for data transformation, governance, and actionable executive dashboards.
A Tableau Hyper Extract is:
A persistent cache of data. It is written to disk and reproducible.
A columnar data store – this is a format of storage that is particularly good for analytics
Separate from your underlying data source. All Tableau queries will go to Hyper instead of live data connection. This has the added benefit of helping reduce resource contention in your underlying data source.
Refreshable on a schedule. You can completely regenerate an extract or incrementally add rows of data to an existing extract via Tableau Server and Tableau Online, or do more complex updates via the Hyper API. Most production use cases require full extract refreshes for a number of reasons.
In-memory and Architecture-aware – Hyper is an in-memory technology but is not constrained by physical RAM available, unlike many similar technologies.
Portable – extracts are stored as files so can be copied to a local hard drive and used when the user is not connected to the corporate network. They can also be used to embed data into packaged workbooks that are distributed for use with Tableau Reader;
Often much faster than the underlying live data connection.
If you are using a live connection in Tableau and you suspect your underlying data source could be causing performance problems, creating a Hyper Data Extract is often one of the easiest ways to improve performance. Beyond being a very capable analytic database, leveraging Hyper gives you additional ways to improve the speed of your data that you might not have access to with other systems either because of the nature of those data sources or because of the administrative rights needed. It’s also worth mentioning that Hyper is workbook-aware in certain circumstances. When it is embedded with a workbook, Hyper Extracts will automatically create summary tables and other artifacts to help speed up your dashboard. This is a level of customization that would be hard (if impossible) to achieve in other data sources.
1. Limit data size by applying data source filters
Tableau performs better with smaller datasets. Applying data source filters ensures only relevant rows are loaded, reducing query size and memory use. For example, if your analysis focuses on the last two years of data, filter out older data in the source or during extract creation.
Aggregating data (e.g., monthly instead of hourly) reduces the number of rows, making queries faster and extracts smaller.
Go to the data source pane -> Click on Add under the Filters -> Add required filters and press OK