By Eva Murray
The two terms data analysis and data visualization seem to have become synonymous in everyday language in the wider data community.
Numerous job adverts focus on data visualization skills while not necessarily specifying the importance of analytical skills. Job titles reflect this trend with the emergence of new roles such as ‘data artist’, ‘data visualization expert’ and ‘data storyteller’, but organizations are still looking for people who can extract value from their data, so these roles must include analytical skills.
Data Analysis versus Data Visualization
Data analysis is an exploratory process that often starts with specific questions. It requires curiosity, the desire to find answers and a good level of tenacity, because those answers aren’t always easy to come by.
Data visualization involves the visual representation of data, ranging from single charts to comprehensive dashboards. Effective visualizations significantly reduce the amount of time it takes for your audience to process information and access valuable insights.
Visual analytics in the process of analysis
However, that’s not to say that the two never work in harmony – far from it. In working with data, analysis should come before the visual output, but visual analytics can be an excellent method for running more effective analyses.
Visual analytics involves the process of building different charts with your data to give you various perspectives. This helps you identify outliers, gaps, trends and interesting data points that warrant further investigation.
The process of analysis is similar to the design process depicted through ‘the design squiggle’ by Damian Newman.
The Design Squiggle
Damian Newman
On the left you have the process of analysis, research and visual exploration which turns into more clarity, an understanding of the data and the finding of insights as you move to the right. Only at the conclusion of the process comes the dashboard, the output that brings everything together in a neatly packaged output.
What is the role of a dashboard?
In your job as a data analyst or visualization expert you will likely be creating dashboards for your stakeholders. What is the role of the dashboard in the entire analytical process, though?
Many people see it as the ultimate deliverable that will answer all their questions. I suggest, however, that the dashboard is just a starting point for further discussion and analysis.
A dashboard, infographic or data story can be an excellent and very effective method for communicating insights. It shouldn’t stop there, though. At the point when your stakeholders work with their interactive dashboard, printed PDF report or the screenshot they have received by email, that is when further discussions should come about. It shouldn’t be the end point.
Look at the below Sales & Profitability dashboard created by Ann Jackson. It is a visually compelling, cleanly designed summary of the data that shows the changes over time, the geographical differences, losses for certain product categories and summarizes key performance indicators as numbers.
Sales and profitability by US state Ann Jackson
Would Ann hand this dashboard over to her stakeholders and call it done? No, because now the real discussions start. Ann can sit with her audience and drill into further details to explore why certain results have come about and identify opportunities for improving business performance. Further investigation of the data, exploring it with your audience, that is the value you add as an analyst, beyond producing dashboards. They give you an excellent basis for these discussions but shouldn’t be the end point.
Don’t stop at the visual
One piece of advice I have shared with many analysts in the wider community is to not simply visualize data, but to show insights and to demonstrate your analytical skills.
Many tools make it very easy to build visualizations quickly. Your role as an analyst is to ensure that the information they present is accessible, easy to understand and clear. I strongly encourage you to put your findings into actual sentences, annotations, titles and subtitles to guide your audience through your report or dashboard and make information accessible, regardless of their level of data literacy.
Ask yourself the question: are my conclusions easy to understand for someone who didn’t go through the same analysis and research process with this dataset?
Look at this visualization by Justin Davis. Justin created two charts and supported them by stating his findings in text. He could have stopped at using titles for each chart, but the inclusion of three sentences ensure that his audience doesn’t have to do their own analysis first and can instead understand quickly what the data shows. And then, they can ask further questions.
The number of women in the House has increased over the last 50 years Justin Davis
Another great example comes from Jenna DeVries who created a simple visualization of Major League Baseball beer prices.
Adding annotations throughout the visual to help her audience understand key findings makes the information accessible and also shows that Jenna didn’t just create a chart but rather went through a process of analysis which resulted in a visualization that best presents her conclusions.
Beer prices in the MLB Jenna DeVries
Show your analysis
Data analysis and data visualization may be different activities, but they’re intrinsically linked, and one can usually support the other.
When you work with data and build your next dashboard, I encourage you to give additional thought to how you can more effectively incorporate your findings. How can you show the key insights to your audience in a way that is most accessible to them?
An effective, well-designed visualization is great but you risk losing your audience if the information is hidden in data art and cannot be acted upon by your stakeholders.
Feature Image Credit: Pixabay
By Eva Murray
I lead the Business Intelligence department at Exasol, provider of the world’s fastest in-memory database. I‘m responsible for executing the company’s data-driven strategy. I am one of Tableau’s Zen Masters, sitting with 29 other world-leading data experts. My passion is bringing data to more people, using it to help change lives, creating educational content and collaboration opportunities for data professionals across the world. I have co-authored a book about Data Analysis and Visualization best practices and co-host the weekly social data project #MakeoverMonday, which connects data professionals around the world. I also run weekly webinars and collaborate with industry experts to create content for data analysts, data scientists, and data storytellers. Initiatives such as #Data+Women and #WomenInTech are important to me, so I work to connect people with the right opportunities for their career progression and development. Outside of work, I’m a passionate triathlete and I love to travel.