8 Ways to Improve Your Data Visualizations
8 Ways to Improve Your Data Visualizations
A high-level discussion of data visualization strategies that big data professionals can use to effectively present their findings.
Join the DZone community and get the full member experience.Join For Free
Hortonworks Sandbox for HDP and HDF is your chance to get started on learning, developing, testing and trying out new features. Each download comes preconfigured with interactive tutorials, sample data and developments from the Apache community.
When working with data sets, it's important to make sure that your data is being properly and efficiently presented to your audience while following best practices for data visualizations. There are many different features and tricks you can use on your visualizations to make sure it's understandable, succinct, and informative. Here are some quick formatting choices and additions that will improve overall readability and meaning behind your graphs and charts.
Conditional formatting gives you quick indicators for the skew of your data based on the given bounds that you have previously assigned to it. For example, if having a quota over 65% is proficient, between 65% and 55% is fair, and below 55% is poor, then you can quickly see with conditional formatting who is meeting the expected quota and who is not.
You can show data trends or moving averages in your charts by adding a trendline. Trendlines are a very simple, yet powerful tools to help you determine trends in data sets and define thresholds for taking action. There are a number of trendlines out there that you can use on your visualizations depending on your data.
Filter by Rule
Filtering by rule allows you to either add a quick filter option to your axes or your data. Seeing sales by day may be an important aspect to a lot of managers and executives in companies, but, when you pull in every single day like you see below, it is hard to make sense of the data. By filtering this down to the last 30 or even 7 days you can more clearly see the days of low and high sales.
You can also do this with your data points as well. For example, you may be in a company with a long list of sales representatives. Trying to create a visualization that shows you how each on is performing may not be effective. Instead looking at the top 10 or 20 sales representative, filtering by rule will allow you to gauge who is always performing well (or vice versa for bottom 10 or 20).
Add a Hierarchy
Adding hierarchy to your visualization will allow you to view data from a high level and drill down into specifics as you begin to ask questions. By adding a hierarchy to the data below, you are able to show both the total sales of each sales rep and then break that down even further by product. This eliminates the need to create extra visualizations and extra work!
You can also use this feature in a chart just as a bar chart that will allow you to drill down into a sales reps by product sales by simply clicking on their corresponding bar.
Sort Your Data
When your data isn't based on date-oriented viewing, sorting the data either in descending or ascending order will visually display what story you are trying to tell. The example below shows your how you can quickly tell that "Complete" is the bestselling product and "Enterprise" is the worst performing in sales.
Format Your Data
Formatting your data can be a quick, simple way to make the numbers more visually appealing and easier to read for an end user. For either gauges or charts such as bar charts and column charts, you can adjust your data formatting to show a certain number of decimals, comma separators, displays as numbers, currency, percentage, or large number formatting.
Include a Comparison
You can both improve and add more insights to your visualizations by including comparisons to your charts. You can display your data based on date, like year over year, or you can use a comparison chart to compare two data points such as budget vs actual.
Keep your chart title simple and to the point since your data and visualization should tell the story. Primarily, your title should directly relate and support the chart underneath it.
Following a few or all of these best practices for data visualizations will begin to help you better present your important data. In turn, remember that half the battle is having a tool at your fingertips that enables you to make these edits and choices.
Published at DZone with permission of Casey McGuigan , DZone MVB. See the original article here.
Opinions expressed by DZone contributors are their own.