Top 10 Blunders to Avoid When Creating Data Visualizations
Creating data visualizations can be exciting — and a reflection of your business skills. Avoid these common blunders so that the path to good data visualizations is less slippery for you.
Overloading your data visualization with too many charts
You want to add as much data to your visualization as possible so that no insight is missed. It can be horrifying to imagine that you have left out a key component of your data. That’s a common sentiment, but executing your data viz that way can be detrimental to your results.
Throwing in absolutely every pieces of data might obscure the data you really need to see. No one will thank you if they have to wade through a lot of information that doesn’t add any value. The best way to approach selecting data is to choose only those data sources that are critical. You can then add more to your data viz after you see what story the data tells you. It’s also helpful to get feedback from your users about what works and what doesn’t.
Opting not to use dashboards
In recent years, dashboards have become an important part of business intelligence (BI) analytics. Dashboards are screens that show critical information about how the business is succeeding or failing. It might show key performance indicators (KPI) or inventory data critical to sales performance.
Two major benefits of dashboards are that they provide both specific and general information as needed. For example:
They help all stakeholders focus on the same information displayed on the dashboard. This way everyone is on the same page and can develop solutions together.
Users can use interactive dashboards to drill down to get information specific to their corner of the business.
Whenever possible, consider using dashboards to help users see a snapshot of the business.
Failing to recognize the importance of data visualizations
Even though most business users have seen a variety of data visualizations, it is possible that they still don’t recognize how important data visualization is to the health of their business. People who haven’t tried to analyze large amounts of data by hand may not realize that it is no longer reasonable to process Big Data efficiently without a data visualization. Data visualizations can be used to gain new insights about the business that would never be recognized otherwise.
Even very small businesses have seen the value of sharing infographics with their customers. Larger businesses have been pushed into using data visualizations because of the onslaught of new types of data like tweets, reviews, and Facebook posts. They simply can’t ignore the valuable information that can be scooped up and analyzed with the right tools. Make sure that you try to use data visualization whenever possible.
Forgetting to list your data sources and copyright
There are lots of interesting infographics and data visualizations online. Unfortunately, not every one of them lists its data sources and/or copyright. Leaving this information off a data viz is fine if you are creating it for a small internal audience. But if you are creating something that will be seen outside a controlled internal group, add your copyright and list your sources.
For an infographic, copyright and data sources are a must because you collect information from a variety of sources to make an appealing graphic. If the users don’t know where the information comes from, they will be uneasy about citing the statistics to others. One goal of an infographic is to have it shared by others. If it lacks credibility and doesn’t cite specific sources, you defeat one purpose for creating it.
In addition, remember that protecting your copyright is only half the battle. You want to make sure you protect and honor the copyright of others. Don’t open yourself up to a lawsuit by using material without attribution.
Picking the same colors for different items on a dashboard
Okay, you’ve picked a color scheme for your data viz that you are happy with. That’s great! It’s not always an easy task to choose appealing colors. What you need to remember you can’t use these same colors to depict different items on all your charts.
For example, if you create a legend for one financial graphic that shows yellow as bad loans, red as foreclosures, and blue as good loans then you can’t also use yellow to depict good loans in a second chart. Consistency is the name of the game. Don’t get caught up creating data visualizations that are pretty but inaccurate or confusing.
Using a black background for all your data visualizations
If you consider all the data visualizations you’ve seen over the years, you would think that it’s a requirement to use a black background. However, black is not always the best choice for the background color. There are many data visualizations that are terrible because they have a black background that obscures what they are trying to show.
Some people equate a black background with drama and feel that a dark background lends importance to the visualization. This is not necessarily the case. Visualizations should be understandable at a glance. If a user needs to spend too much time determining what’s on the data viz because the data blends in with a dark background, you’re doing your users a disservice.
Don’t think that you’re being cautioned to never use a black background. You can use black, but make sure to use appropriate white space around the elements of your data viz so that they’re easily readable.
Failing to design for mobile screens
Using mobile devices is no longer a novelty. According to Google, more than 138 million U.S. users have smartphones. And the tablet market is also gaining steam; Gartner Inc, a tech research company, reports that more than half a billion tablets are expected to ship in 2013 and 2014.
You can’t afford to create data visualizations that won’t properly display on mobile devices. If you do build only for large screens, you are preventing your users from accessing their data when and where they need it.
The best way to tackle the problem of designing for multiple screen sizes is to create your data visualizations using responsive design. In a nutshell, responsive design allows your graphics and content to flow into the correct place on a variety of screens — tablets, smartphones, desktops, and so on. Make sure to plan for this type of design upfront. Adding it later is time consuming and costly.
Making it unattractive
The fact that people make judgments about the attractiveness of a visual is a given. Human brains are hardwired to process and understand images.
Creating an ugly data visualization is a blunder that you don’t have to make. You have resources at your fingertips both online and analog that can help you create a visually appealing data visualization or infographic. This doesn’t mean that you will produce something stunning every time, but it does mean that you can create something that is appealing and won’t drive readers away.
Using wrong or incomplete information
This is a tough one. Sometimes it’s not immediately apparent that a particular slice of data is not a good choice. It may even happen that you actually include bad data in your data visualization at the start. After all, on the surface it’s difficult to determine if chunks of data are left out or just wrong for the data viz at hand.
This is where a relationship with the IT department can come in handy. Typically they know just how accurate the data is and where it falls short. Even if they don’t know about every detail of the data, they will know more than you do. Make sure to cultivate a relationship with them so that you know the good, bad, and ugly about your data.
Forgetting to implement proven reading patterns
Because of advances in neuroscience, we know more than ever before about how our brains see and process information. For example, we know that there are ways that data can be arranged to make it easier to interpret. You should use this information to make your data visualization as understandable as possible.
Specifically, you should consider the Z reading pattern for the layout of your dashboard. Typically, people’s eyes follow a Z pattern when they scan a page. The eyes travel left to right across the top of the page, down at an angle to the bottom-left corner, and then across again from left to right — it’s as if the eyes are drawing a Z on the page. This means that if you put a menu across the top, content in the center, and copyright or other information at the bottom that you will have set up your dashboard with a layout that most people will be able to scan and process quickly.