7. Optimizing the View

Subtitles Enabled

Sign up for a free trial to access more free content.

Free trial


Changing your view can improve the performance of your Tableau workbook. In this lesson, I'll run through some tips to bear in mind if your views are taking a long time to load.


Optimizing the view - reduce the number of data points

- Tableau recommends using less than 1,000 marks within a view
- Using very large numbers of marks (or data points) can slow down performance
- To solve this problem, segment the dataset with a filter and split analysis over multiple sheets

Optimizing the view - accessing the Describe Field

- Provides you with helpful information on a dimension or measure
- Much faster than dragging measures and dimensions into a text table
- Especially useful for datasets with a large number of columns
- Accessed by right-clicking on a dimension or measure


In previous lessons, we have looked at ways to improve performance while loading data. In this lesson we're going to look at some options to improve performance within a particular view. Let's start by knowing a few things to keep in mind as you build your workbooks, so that it will perform faster. After you load data for the first time, and go to the worksheet, you're often tempted to drag columns into the shelf to see what data exists within this particular shelf. With larger datasets, this can create problems. And if I want to do a quick check on what exists within a dimension or a measure, I right click, and go to describe.

And this gives me a nice summary of the data. It explains whether it's discrete or continuous, a dimension or a measure, its location, the type of data, and if it contains a null.

This technique is particularly valuable when you have lots of dimensions and measures due to the number of columns in your dataset.

To close this dialog box I'll just click in the bottom right-hand corner.

The nature of the visualization that you create also affects the loading time.

Let's say I create an X Y scatter with quantity and sales.

And this only contains one data point due to aggregation happening on both of these measures.

Now let's add a level of detail, say customer.

I know of many more data points on my view, and this will take longer to load. Although in this case, I still don't have too many data points.

Instead of customer ID, now let's add transaction ID, which includes every row in the database.

And this will mean 32,000 data points on my sheet. When I click add, you can see that this is not easy to view, but also for large datasets such as this can create delays when building your view. So instead, I'm going to flip back to customer ID.

Whenever you have a situation where the number of data points or marks is beyond a thousand, it's always worth asking, can I improve the visualization to make it more meaningful, and with the faster loading time for my manager? In the previous example, we looked at a chart with too many marks or data points in the view. This can also happen with text tables. Let's flip this chart to a text table, and again, I'll remove customer ID, and put in transaction ID.

And for large datasets such as this, it can create delays, and make the dataset more unwieldy for users as I scroll down.

When you have a large dataset such as this, try to perform a filter on a particular column rather than including the whole dataset within the sheet. For example, I may only want to include the top 1000 transactions. So to do this, I'll create a filter, go to top select top 1000, and the field which will be transaction ID. I'll then press okay.

As you can see, this is now much faster, when I scroll up and down the dataset.

As you may have noticed, Tableau actually tries to stop us adding a large number of data points into our view, and create a warning whenever that number is greater than 10,000. I'll show this again by removing transaction ID from the filters, removing transaction ID from the rows, and now adding this value again to create the warning. Whenever you see this warning, always try and abide by Tableau's recommendation, and filter your dataset, or do not add this particular column to your view.

The remaining ways of improving the performance of your view involve filters which I'll explain in the next lesson.

> Tableau Tableau for data visualization
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