From The Data Marks
As a beginner, or even a more experienced developer/designer, you may find your self missing some obvious things in your visualizations. It’s the small details that can detract from a great viz, so here is our attempt to provide a checklist of things to ensure you’ve completed right before you hit publish.
Data Source Identified/Credited
So you found amazing data set, and built an amazing visualization, if you fail to site where the data came from it can bring question to the credibility of the visualization. A simple text box listing the Data Source and either the organization’s name, the study, or a hyperlink to the original source. This small step is critical.
Tool Tips; Beyond Default
Every time a visualization gets publish with default Tool Tips, Adam Crahen kicks a puppy (We’re kidding). Seriously though, tool tips are awesome and you need to either A) Use them! or B) Turn them off. What additional insight can you add with the tool tip? What formatting can you apply to make it visually interesting? Be thoughtful with the use of tool tips and give them a purpose.
Avoiding Chart Junk
Take a look at your worksheets, what is the first thing you notice? Is it critical or important enough to be the first thing you process? If not, rethink what you are doing. Do you have grid lines, or row/column banding? Consider removing them to lightening them to push them to the back of the viz. Are you showing the row/column field tables? Are they needed? If not, remove them. How about your tick marks on the axis, if there a larger interval that makes sense? Remove the excess tick marks by editing the axis.
These “extra” things can detract from the visualization, and removing them helps clean up your work.
Avoiding the Redundancy Office of Redundancy
Do you have an X or Y axis and data labels? Pick one or the other; Removing redundancy from your vizzes can help strengthen the message and remove chart junk. Are you using multiple preventive attributes to identify one element type? If so, pick one and stick with it. Are the tool tips repeating the information visible in the chart or graph? Turn them off or add additional information that supports the message.
Color Consistency
Use color sparingly, and if you are using color in multiple places on different aspects of your visualization be consistent! Don’t use blue to represent sales and use it again to represent one of your categorical values. Use (or overuse) of color is one of the most common mistakes in beginners.
Design Attribution (if needed)
If you have reversed engineered a new chart type to you, or have been inspired by someone else work, be sure and credit them somewhere in the visualization. While the Tableau Community is giving, nobody likes theft. Credit where credit is due is just good practice. This credit fits nicely with the Data Source listing, so go ahead and place it in that text box if needed.
Titles, Texts and Instructions
If you design isn’t completely obvious (and even if it is) give your user some instructions, whether its “How to Read this Chart” or maybe an overview of what you see in the data. Text adds value and insight to your visualization, so by all means add it to your work.
Tableau-Safe Fonts
There are thousands of different fonts, but not all of them translate to the Web, and even fewer are supported by Tableau Public. Be sure and test your viz from the server and make sure your fonts are doing what you expect them to do. If you fall in love with a font and really want to use it in the viz, consider making an image for title and texts. You can easily create these in PowerPoint and save the text boxes as images.
What things do you always find yourself going back to “fix” after you’ve published your viz? Let us know in the comments!