Search

How to speed up your Tableau

With many businesses using Tableau as a data visualisation solution for presenting back their data in a useful way, I have identified the steps that should be put in place to make sure you are getting the most out of the software and have the fastest possible response time to your insights. Below I have listed out best practice methods and tips which will see an uptick in performance and mitigation of annoyingly slow load times. Tips & Best Practices · Set timestamp to datetime fields YYYY-MM-DD [hh:mm:ss] as this has the advantage of being an industry-standard date and time format, this format is frequently used in databases and languages alike. It is a ubiquitous datetime field that most systems can translate between and reduces friction with data types and naming conventions. · Minimise the number of JOINED (and UNION) tables as Tableau will spend a large amount of time running queries that execute these joins, either reading the data to match between tables or executing the query. If these tables are joined within the database and materialised as a snapshot view, there will be a reduction in job runtime. · Delete or remove unused, unnecessary, or redundant columns, tables, worksheets, or data sources. Reduce the available search area of data and allowing the system to ignore excess information and focus only on what is important for that specific job. · Reduce nested calculations, essentially calculations within or referencing other calculations. Maintain calculations within the database to relieve strain on Tableau and avoid additional work in-platform before reaching the dashboard. · Use INCLUDE filters instead of EXCLUDE as this will reduce the scope of data you are searching for in specific queries or when populating dashboards. Be more specific upfront with the data you need for each dashboard view. · Minimise number of fields – use the ‘hide all unused fields’ option which will do as it suggests and removes columns from the data source. · Memory-Mapped Files/Multi-Index queries that speed up the process of querying a select number of fields when grouped together and frequently referenced. · Create temporary tables and views to then manipulate and query instead of the main table. In summary, if you apply any combination of these tips then you will see a boost in your Tableau performance and be able to make business-critical decisions in a fraction of the time it previously took. Compounding all of these will send you on your way to crunch numbers and deliver actionable insights faster than your competition. For more details on how to implement these changes get in touch with me via staylor@beinthelead.com.