Marc Wintjen is a Risk Analytics Architect at Bloomberg with over 20 years of professional experience. He joined us during a webinar to discuss how data literacy helps improve communication while empowering teams throughout the organization. If you’re interested in learning more about Marc, listen to the full webinar here.
The Power of Data Visualization
Nowadays, organizations generate terabytes of data every day. Once the data is available, the team needs to be able to ask the right questions in order to derive valuable insights from data. Finally, by combining data with a detailed narrative and the right visuals, you will be able to communicate your data insights efficiently and effectively. Marc emphasized how data literacy is the cornerstone of great data visualization, which is achieved when everyone in the organization is empowered to ask the right questions about their data.
Marc defines data literacy as the ability to read, analyze, or argue with data, in order to draw useful data insights for your organization. Data visualization is a key component when communicating data insights across your organization. Selecting the right data visualization tools and visuals will help you break down the complexity behind the data sources and make it easier for others on your team to consume. Data visualization tools also reduce the ramp-up time to understand the context behind the data and invite the reader to discuss the what and the why.
"Visualizing data breaks down the complexity behind the data sources to make it easier and faster to consume." - Marc Wintjen, Risk Analytics Architect at Bloomberg
Ask the Right Questions for Better Data Insights
Data visualization is both an art and a science: it is possible to construct a data visualization that is both aesthetically pleasing, like a work of art, and also provides insights from the underlying statistical evidence captured from the data. The goal is for the audience—or in this case, your team—to quickly digest and interpret the information without any guidance. To determine the right visualization tools Marc suggests that you need to be able to determine what is the core business question you are trying to answer. Hence, it is important that the team is empowered with enough information to ask the right questions. There are three main techniques to ask the right questions:
Know Your Data (KYD): Understand business requirements and what the users will do with the data, including the people, processes, and technology used.
Voice Of the Consumer (VOC): Understand specific pain points by interviewing your business users and actively listening to their needs.
Always Be Agile (ABA): The process involves a series of sprints, also called stories, which commonly last 2-3 weeks, and the goal is to understand the what and the why behind each story.
As explained by Marc, Bloomberg uses the Agile methodology, which has become common in the tech industry for application development. The main advantage of the Agile methodology is that it creates an interactive communication line between the business and engineering teams to iteratively deliver value. But all of these techniques help improve your data literacy, regardless of your role within the organization.
Select the Right Visualization Tool Based on your Data Type
Marc argues that to achieve a good level of data literacy in any organization, you need to master the techniques of asking questions and data visualization. Once you understand what the right questions are for your organization, you will be able to implement the right visualization tools. However, you need to keep in mind that good graphics are not just displays to extract information from, but devices to explore information with.
Before defining the right displays and graphs, you need to know what data types you are analyzing. In short, data types are the details of how data is stored and these are commonly well known in different programming languages. Understanding how different data types work will help you create clearer data visualizations much faster. In his book, Practical Data Analysis Using Jupyter Notebook, Marc classifies data types as follows:
Continuous: Numeric values like integer, float, or time.
Categorical: Descriptive values, like a stock ticker, first name, or last name.
Discrete: Define boundaries around the possible data values, like numbers on a roulette wheel.
"Good graphics are not just displays to extract information from, but devices to explore information with." - Marc Wintjen, Risk Analytics Architect at Bloomberg
Steps to Create Great Data Visualizations
Data types and data visualization are core elements of your data model. Marc defines the data model as the relationships between one or more tailored data sources. To achieve a complete model, Marc describes a set of steps that you should follow when implementing your data model.
Understand the question: You must know what questions you are trying to answer with the data, then you can decide which model will best fit your goal.
Identify key parameters: Identify what data fields will provide the right data insights. In many cases, you will have to conform the data so that each row has to be a consistent data type, to make your analysis faster and easier. There is no need to use every single field of your data source, just the necessary ones.
Visualize your data: Creating a good chart narrows down to defining what fields you plan on using for analysis dimensions and measures. Dimensions are values with descriptive attributes that are commonly used to identify a person, place, or thing. A measure is any field that can be aggregated using statistical calculations, like max or average.
Reuse your code: You don't have to reinvent the wheel every time to create a new chart, app, or dashboard. If you model your data correctly, adding the right trend chart could help you save time and focus on the right tasks.
Data literacy can be a great tool to improve communication across different teams in the organization. To effectively communicate your data insights, your team must acquire a certain level of data skills to create effective data visualization models, which are used to provide context and insight to users that do not have a data science or engineering background.
Data models are the secret sauce to create great graphics that answer the right business questions. If you are interested in learning more, check out the blog on Storytelling for a More Impactful Data Science.
Lastly, sharing is caring, and even more when it comes to your team. To break down those data analytics silos, you will need to empower others within the organization to read, analyze, and argue with data. If you want to learn more about Marc’s approach to data communication you can listen to the full webinar “Bridging the Communications Gap with Data Literacy”.
← Back to blog