I’ve faced a number of issues recently with Tableau report performance not being as good as required. This article is my attempt to combine the different ways to improve the speed I’ve learnt from other sources. The intention is to update this as I learn new techniques to achieve better performing Tableau reports. Read more…

(If using Tableau 9+ check this updated article on using LOD calculations to create bins from a measure.)

This post is a continuation of a previous article showing how to create bins from a measure introducing a situation where data blending is a solution. Credit to Richard Leeke for supplying the solution.

In the previous article we created bins from a calculated field and discovered that when we use these bins they can re-calculate and give us results that we were not expecting. To recap I had some data showing an enquiry count per listing per month. I want to sum these enquiries for each listing to put each listing into a bin dependent on the total number of enquiries it had received over the time period. Read more…

This post about how to create bins from a measure in Tableau was originally written in the days of Tableau 7. Now things have evolved and it is far more straightforward, the updated article on using LOD calculations to create bins from a measure is here.

For this post I have to give a huge thanks to Richard Leeke who found the ‘Tableau only’ solution (as opposed to pre calculating the data) for this problem. As a quick overview for what I was trying to do using Tableau, I wanted to create a calculated field of which the result would be used to create bins. The calculated field is a measure, not a dimension, but the same rules apply.

The post is quite long and complex hence it’s broken up into multiple parts – the solution using data blending will be detailed in the next post, creating bins from a calculated field.

The test data has 3 columns: Month, ListingID and EnquiryCount – in other words it showed the enquiry count per listing per month. I wanted to calculate enquiries per listing over the entire time period and use the result of this calculation for the bins. The sum of these enquiries for each listing id defines which group they belong to – i.e. 1 – 10, 11 – 20, etc. In other words if ListingId 1 had an EnquiryCount of 10 in Month 1, 2 in Month 2 and 8 in Month 3, ListingId 1 received 20 enquiries in total so would be in the bin 11-20. Once I know which bin each listing belongs to I want to see for each group what % of total enquiries came in month 1, month 2 and month 3. For ListingId 1 50% of enquiries were received in Month 1, 10% in Month 2 and 40% in Month 3. Read more…