Data visualization is still a relatively new field for many of us, and in our effort to provide useful tips on data analysis, we want to provide some helpful tips for how to create great data visualizations. Experts like Edward Tufte or Stephen Few have incredibly useful resources on data visualization, but consider this your beginner’s guide in three main ideas.
The idea behind good data visualization is to simplify ideas and reveal insights that would have otherwise been buried in the raw data. Part of the simplification process is knowing what questions to ask and how to best represent those questions. To do this, simplify the visualization itself as much as possible and eliminate clutter. The graph below is just one example of how to NOT simplify your graph. In this graph, customer signups are broken down by state and time.
Even with DataHero automatically putting most states into the “other” category, there are far too many states to get any insight out of a stack. A question regarding states or regions would be better left to a map.
Jamming too much information onto one chart can be confusing or just plain useless. Usually it’s easiest to start with one question at a time. For example, begin with the question: how many sign ups have I had in the past year broken down by quarter?
Then the next logical question is: how many account activations have I had over the same period of time? In this case, that would mean password creations after a user has verified an email address.
Once you have the answers to these what/how questions, you can get to the more interesting why questions.
Whether you’re comparing different pieces of data, or several layers of data, explore your data for interesting patterns and relationships. As mentioned in the point above, start with a question or two and see where it leads. The “why” questions are the most interesting. A line graph of sales for the past year by quarter will tell you what is happening; sales are increasing or decreasing.
A column graph of the same sales data broken down by region however, will tell you why sales are increasing or decreasing.
View the data in different ways as well. Are you not seeing a trend in yearly sales if you view them by week? Change the time format to months and see what that reveals. Data visualization provides the opportunity to make a massive amount of data easily digestible in seconds. Leverage that by slicing and dicing until you uncover the patterns. Try a comparison chart for example. The chart below shows how revenue has changed from quarter to quarter.
Communicating discoveries in your data is a different skill altogether than the analysis that goes into the first two steps of data visualization. This may be the most important step though, as sharing your findings will have the largest impact on an organization or company. In communicating your data findings, consider the audience, and their familiarity with the topic. You also want to walk a fine line between creating a story and presenting the numbers to support a claim. For more tips on how to communicate your data effectively, check here.
Data analysis is really a natural extension of your thought process. The idea is to make the technology so seamless and intuitive that you pursue your natural train of thought without being interrupted by formulas or formatting, which is exactly what DataHero tries to do.
DataHero helps you unmask the answers in your data. There’s nothing to download or install. Simply create an account and connect to the data services you use everyday (like Salesforce, Stripe, MailChimp, Dropbox and Box). DataHero automatically decodes your data and shows you the answers you need through dynamic visualizations.
Get the fastest, easiest way to understand your data today.Sign up for free