Did you play professional basketball? I didn’t either. You know who else didn’t? Daryl Morey. In 2007 he became General Manager of the Houston Rockets and in 10 years at the helm has yet to have a losing season.
Some Tableau dashboards have performance issues. Commonly people will assume it’s because they have too many rows or columns of data, but that’s not always the primary issue. Your dashboards might be hindered by memory, data source type, computer memory, or a number of other factors. Check out the video below to learn about some of the common issues plaguing Tableau dashboard performance and how you can rectify them!
A couple weeks ago I was teaching a course and received a question from a student. Her question was “How can I use Tableau to only show the most recent 3 transactions per customer?” I thought I’d have an answer for her quickly, but I was wrong.
My first thought was, let’s just use the “RANK” function to accomplish this. We’ll use it as a Table Calculation to determine the most recent transactions by customer. I was feeling confident until I saw this:
A student recently asked me how she could create dynamic, color-changing labels based on whether a field passed a threshold. My first response was “Tableau can’t do that.”. My second thought was “How can I make Tableau do that?”
Most data isn’t stored in a manner that is optimal for analysis. It’s frustrating when Tableau doesn’t like your data structure. If you are new to Tableau or working with a new data source, you might run into issues getting columns and rows to display the way you want.
Tree maps are a data visualization used to communicate hierarchical values in a systematic way with nested rectangles. A lot of the tree maps I see look something like this:
I don’t know about you, but I don’t find this to be particularly informative or compelling. I prefer to use tree maps as a way to highlight a few relevant data points. Notice in the dashboard below how I use a tree map to highlight the top 10 items sold.
We have done hundreds of business intelligence (BI) projects. We often come into the middle of projects that are headed for failure and stakeholders are looking for a magic bullet. There is no magic but we do have repeatable processes that assure success.
First, how do we measure success for our BI projects?
If you’re like most people in the world of analytics, you probably didn’t get a formal education on the topic. Sure, we all know a few data scientists, but most of us came from other backgrounds. When semi-familiar terms are tossed around, it’s easy to nod your head along with the crowd while in the back of your mind wondering, “does that mean what I think it means?”.
I was given a challenge by a student recently. They wanted to color territories on their map by a region, but they wanted the intensity of the color to represent the amount of sales that occurred in an individual territory.
Adding a highlight action can be a really nice way to help your end user recognize relationships between data on separate worksheets. However, sometimes the highlight feature doesn’t appear to work properly. Let’s look at an example.
Take a graph like the following:
Let’s say you want to visually distinguish the 5 days with the lowest profit ratio. I’ll show you how I prefer to approach situations like this.
Designing dashboards in Tableau can be tricky. One problem my students run into is how to space worksheets evenly. It is time consuming to drag the edges of worksheets, making minor tweaks, until they are all the exact same pixel width.
It’s easy to want to cram too much information on a single Tableau dashboard. Whenever I have a scenario where I want to provide additional information to my end user without overcrowding my dashboards, I include an info button.
There will be times when building dashboards when you’ll have multiple worksheets that use the same measure on color. It’s best practice to use color consistently throughout your dashboard, but it can also lead to unforeseen complications. Take the below dashboard as an example.
Every dashboard you create is designed to do more than just communicate numbers. It’s a tool with a purpose. That purpose might be to increase revenue, decrease costs, or add value to your product. When creating dashboards, I like to keep that purpose forefront in my mind.
A powerful addition you can apply to your dashboards right now is to lead with questions. When you change the title of your dashboard into a question, you turn your audience from passive viewers to active viewers. They aren’t just reviewing numbers, they are now trying to answer a question.
It’s easy to be more complex than necessary. It’s hard to be simple and succinct. As Mark Twain said, “I didn’t have time to write a short letter, so I wrote a long one instead.” The same is true in the world of data analysis.
Most people just want to know: “based on what’s happened (data), what should I do next?”