Interactive Data Visualization
Written by Tereza Seidelova |
Data visualization is the process of creating a visual representation of information. For centuries, people have been using static data visualizations — the map being among the oldest and most famous examples. Every aspect of data analysis has evolved with technological developments, including data visualization.
Interactive data visualizations have become a common part of most reports and dashboards. They allow users to engage with the data and easily find answers to their specific queries.
What Is an Interactive Data Visualization?
Interactive data visualization, such as insights and dashboards, aims to represent data graphically. Compared to non-interactive visualization, user engagement is needed, such as clicking a button or moving a slider. The core of the visualization is action and reaction, specifically human input and quick visual output.
Non-interactive vs. Interactive Data Visualization
Non-interactive data visualization is static and simple. It can include forms such as graphs, heat maps, and various types of charts (e.g., pie, bar, or line). For example, imagine a chart created in Excel. It is easy to quickly create for simple data queries. In general, non-interactive visualization is more suitable for less complex data stories in which you only perform one or a few queries. It is also the optimal format for printing and sharing reports via email, as the information is static in time and easy to view.
On the other hand, interactive visualizations are perfect for large amounts of data when you have more questions and a trend to investigate. Interactive data visualizations enable you to fluidly answer questions and travel from one visualization to another. It displays data in context, allowing you to easily find answers to your questions or hypotheses.
Interactive visualizations are often used in dashboards and business intelligence (BI) reports. It provides an easy way to understand insights and is more practical and time-saving than using long tables of numbers as in the case of static reports.
Features of Interactive Visualizations
The goal of interactive visualization is to attract users’ attention. When using interactive data visualization, the data you display is all up to you and your curiosity.
The following are typical features of interactive visualizations.
- Filtering allows you to reduce or specify the data that is displayed in the visualization.
- Drilling allows you to move from one visualization to another. It also allows you to send an action from the dashboard.
- Zooming and panning is useful if you want to see a specific detail. You can zoom in and see only a particular part of the visualization without getting distracted or needing to create a new insight.
Let's take a closer look at the features of non-interactive and interactive visualizations. The insights below examine the data from Lenstore’s “Healthy Lifestyle Cities Report 2021.” The dataset contains details about components of the lifestyles of some of the world's cities, such as level of happiness, life expectancy, sunshine hours, and annual average working hours.
We will focus on the level of happiness and try to answer the following question: What is the level of happiness in Helsinki? How does the level of happiness in this city compare to other cities?
First, we’ll check the non-interactive data visualization.
You can see that the level of happiness in Helsinki is relatively high, as it is higher than other cities and even higher than the average world's happiness level. You might wonder what causes the difference in levels of happiness and raise another question: Is there any connection between the happiness level and the annual average working hours? If we check the first graph again, we won’t find the necessary data there.
However, if we consult the interactive data visualization below, we will find the answer.
First, you can zoom in and see the data about Helsinki up-close.
After that, you can drill in and check the annual average working hours in Helsinki and compare them with the world's average. Additionally, you can check the life expectancy in the city and drill into another insight (dashboard).
Additionally, you can filter the visualization to compare Helsinki with other European cities only.
Custom Build Applications
In general, there are two possible ways to create interactive data visualizations: End users can use predefined options, including drag-and-drop, while developers can define the business semantics on top of the data so that non-technical people can work with analytics and know what they are dragging and dropping. However, interactivity isn’t limited to dashboard visualizations. It can also apply to building applications on top of data. These applications include various formats of interactivity, such as new visualizations, chatbots, natural language processing, and in-game analytics.
Data is part of many other applications that companies work with and offer to their customers. As such, working with data and its visualizations needs to go beyond the scope of traditional BI tools. The concept of headless BI enables the connection of any application, data platform, or visualization tool to the semantic layer. Compared to traditional BI tools, with headless BI, the semantic model is decoupled from the BI components and exposed as a shared service via APIs and standard interfaces. In simple terms, the analytical backend is separated from the consumption layer so that you can use APIs to access and present data from the semantic layer and visualize them in any application.
Set side-by-side with traditional analytical platforms, in the headless BI platform, everything you build is both human- and machine-readable so you can manage your analytics as a code. Headless BI enables access to data for end users as equally as IT or other technical owners. The platform is flexible and doesn't limit the user. Interactivity can be understood as a way to choose how to display the data.
For instance, chatbot is an application built on top of GoodData Python SDK and Pandas DataFrame. In chatbot, you can manage your data using code in order to create a visualization and interact with it.
Another application built on top of the GoodData platform is Natural Language Query (NLQ). It allows you to search for information in conversational language. The NLQ server communicates directly with the GoodData semantic layer and is able to create an insight based on your request.
Let's revisit the level of happiness in the world and what can affect it. Now we will focus on the amount of sunshine hours in the city per year. Using NLQ, we can request a column chart displaying the desired information.
Benefits of Interactive Data Visualizations
Simplify Complex Data
Interactive dashboards can represent a complex data story clearly. Incorporated features such as filtering and zooming can help make the data more manageable. Interacting with large datasets and using visualization aspects helps users to quickly understand the story of the data.
With interactive data visualization, you can easily identify trends and relationships between data. Additionally, with the ability to observe how data changes over time, you can identify overlooked cause-and-effect relationships. Hence, you can develop business insights that help assess KPIs' statuses and lead to data-driven decisions.
According to Seyens, half of the human brain is directly or indirectly connected to visual processing. Additionally, at least 65% of people are visual learners. Interactivity based on visualization rather than simple charts enables users to process information easier, as humans naturally interpret visual information better than numbers.
Boost Engagement and Productivity
Interactive insights and dashboards enable you to engage with data in ways that are impossible with non-interactive dashboards. Interacting with data by employing dynamic charts, incorporating shapes, or changing colors can boost users' productivity.
The users get control over what they see, having the opportunity to adjust the visualization according to their needs based on location, age, job, or any other factor. The user becomes an active participant instead of a viewer alone. Diving deeper into the data may raise further questions. However, in comparison to using static visualization, users can find the answers they need without distractions or needing to create a new chart.
A direct connection between the user and the data is a huge benefit. The ability to customize and change the perspective of a visualization depending on what the user needs is the result of personalization of BI. Interactive visualizations offer the flexibility to choose whether to create visualizations directly in the analytics platform or to build your own application on top of the platform. Everything you build is human- and machine-readable, so the data is as approachable for business users as it is for technical users.
If you strive to build a data culture that ingrains data in all processes, products, and people, you need to focus on the interaction of data, not just pure consumption. With GoodData, you can work with predefined interactive dashboards or create your own interactive application and manage your analytics as a code. To get started today, simply request a demo.
Written by Tereza Seidelova |
Subscribe to our newsletter
Get your dose of interesting facts on analytics in your inbox every month.Subscribe