Recognizing the Three Traits of an Effective Visual - dummies

Recognizing the Three Traits of an Effective Visual

By Mico Yuk, Stephanie Diamond

When your black-and-white mock-up is complete, you’re ready to add the oh-so-powerful visuals that will make it pop. That’s why you started this journey to begin with, right? When adding visuals to your mock-up, it’s important to focus on adding effective visuals. Unfortunately, due to a lack of thought leadership and training in the business intelligence (BI) industry, tons of visually attractive but ineffective data visualizations provide zero value. Just do an Internet search for data visualizations to see a few examples.

The table below lists the three main traits of an effective visual.

Three Traits of an Effective Visual
Trait Details
Data is clear. Make sure that the data is clear, both in purpose and
Visual fits the data. Whether you choose a chart or text, be sure that you’re using
the right visual for the job.
Exceptions are easy to spot. Whether you’re highlighting a comparison or outliers in the
data, you should make it easy for your users to identify exceptions
in the data.

The preceding table was influenced by Edward Tufte, who is considered to be the godfather of data visualization. His book The Visual Display of Quantitative Information, 2nd Edition (Graphics Press), is one of the best-regarded books in the data visualization field. Though it takes a scientific approach, it’s a must-read for data viz beginners and experts alike.

These three traits aren’t all-inclusive, so you shouldn’t expect to have all of them to decide whether a visual should make it into your mock-up. Instead, use them as guidelines as you choose your visuals. The more traits you have in each visual, the more effective your overall data viz will be!

Data is clear

Effective visuals display data that’s clear in both presentation and purpose, not distorted in any way. A common mistake is to push too much data into a single visual, causing the important point of the data to be hidden, overshadowed, or distorted by all the noise. The following figure shows a good example of a data visualization that uses a donut chart to show what types of mobile devices people are using. See how the 3-D effect makes it very difficult to understand the data.


It’s also important to ensure that the purpose of the visual is super-clear so that the user has no room for misinterpretation. Good data visualizations tell a story at a glance, leaving the reader wanting more. If the data visualization is confusing or misinterpreted, most users get turned off and abandon it. The figure below shows an example of a confusing visualization that depicts the usage of social network. Unfortunately the colors and percentages seem to have no correlation and are therefore very confusing. Can you tell what the visualization is portraying?


Visual fits the data

The visual has to fit the data. Visuals are more than just charts, however, and certain visuals just don’t fit certain data. Usually, you can present data in multiple ways. Your job is to find the most effective way to do so.

You should never use a pie chart, for example, to show data with more than five data points or to display any data set with little to no variation in magnitude. Similarly, you should never use a table or scorecard to show a trend over time.

The figure below shows two visualizations that chart the same data. The line chart at the top is the best option for showing the Sales Margin trend in 2014, because it makes it blatantly clear that the company’s expenses are soaring way above its profits. The column chart at the bottom doesn’t convey that trend as clearly. Column charts are best used to compare items.


Exceptions are easy to spot

Whether they’re in the form of alerts, comparisons, or outliers, exceptions in the data should be easy to spot in an effective visual. If an exception requires additional deep analysis to understand, chances are that your visual isn’t effective.

Exceptions in data visualizations are extremely powerful and can add great value. When users can spot exceptions and decipher them quickly, they know whether immediate, moderate, or light attention is needed. Highlighting exceptions also provides insight into potential trends that may require attention.

The following figure shows a chart that uses an alert to highlight some of the exceptions in the sales data trend.