8. Faster Filters


Filters can cause significant delays when generating views. In this lesson, we will learn how to optimize filters for performance - with a particular focus on quick filters.

To explore more Kubicle data literacy subjects, please refer to our full library.


Quick filters and performance

- Dimension and Measure filters are known together as Quick Filters
- Quick filters can cause considerable performance issues for Tableau, when dealing with large datasets
- Aside from filtering the data source, a number of techniques can be used to optimize quick filters

Optimizing quick filters

- Try replacing quick filters with action filters
- Use the wildcard option when you have a long list of options (e.g. email addresses)
- Avoid using Only Relevant Values option


In previous courses, we've looked at a variety of filters. From extract filters, through to context, dimension and measure filters. When in the view, dimension filters and measure filters are used the most often. And together they're known as quick filters, and they can create severe performance issues if not used correctly.

In Tableau, it's often tempting to add lots of quick filters to the right-hand side. Because they're very easy to create, and they can give you some very interesting insights really quickly by updating your charts for various measures and dimensions. However, for larger data sets, lots of these filters can impact performance, and really frustrate users who are expecting instant updates.

When you want to use quick filters, there's a couple of ways of improving performance.

The first of which is to avoid quick filters altogether, by using different sheets as filters. For example, I can select revenue by salesperson and use as filter.

And now I can remove this quick filter.

And as you can see, I get a very quick update, whenever I use this filter. I can also do it for my second sheet. So when I want Pennsylvania, I get Pennsylvania on the right-hand side for each salesperson.

And this allows me to remove this quick filter.

The reason that these filters are so fast is because they don't generate database queries, unlike quick filters which do.

Another performance issue related to quick filters is cardinality. High cardinal columns have lots of values that are very rare or unique. For example a list of email addresses or unique identifiers.

And this particular example where we have states, we can obviously have up to 50 values. Which is in very high from a cardinality perspective, but also, could affect loading of large datasets. If we create a filter on state, which I'll do, we can see all of these check boxes.

And Tableau has to create multiple queries for each of these options that are created. If we have very large datasets, and these queries are taking a long time, a faster option is the wildcard match.

So when we select wildcard match, and just want Florida, then FL will do. And now I can only see Florida in my chart of choice. Again, this is not the best user experience, but when there's performance issues at play, wildcard options might be something that you explore. To make sure that this filter works for both charts, I simply hit the dropdown, Apply to worksheets, hit selected worksheets, make sure that both are ticked.

I now have revenue just for Florida.

In the next lesson we'll move on from filters, and now look at optimizing calculations.

Advancing in Tableau
Optimizing for Performance


My Notes

You can take notes as you view lessons.

Sign in or start a free trial to avail of this feature.

Free Trial

Download our training resources while you learn.

Sign in or start a free trial to avail of this feature.

Free Trial