Why is storytelling essential for effective data visualization?

The value of data resides in the fact that key information is encoded but not fully utilized by the business yet. As a data scientist, one key objective is to unveil that information as insights to enable stakeholders to make correct decisions. To communicate the message and translate those insights into action, storytelling—where solid data backs the narrative—is one critical step to realizing the business impact.

Time-out: data storytelling isn’t necessary for effective data visualization in all cases. Sometimes you and your audience just need to pull a fact or single piece of information from the data, and a well-designed chart would accomplish that. Other times, such as in a crisis situation, presenting an extended narrative would waste valuable time and potentially aggravate your audience. What we teach at DataLiteracy.com is to understand the virtues and benefits of a data story, as defined as a narrative or a sequence of individual findings in data, often shown visually - and this article covers some of those. Equally important, though, is to appreciate what conditions are needed in order for you to tell the data story well. Choose wisely!

As experts in the field, we know that data storytelling and data visualisation work together to communicate insights and translating data analyses into understandable terms in order to influence a business decision or action. For instance, data Stories performs automated analytics to determine relevant and accurate facts about the underlying data–from basic calculations to more advanced statistics. It also provides context, highlights trends, and makes data-driven insights more meaningful.

Storytelling is required to help identify patterns in a data set, or to explain those patterns to a wider audience. These are the rationales behind creating what are known as, respectively, exploratory and explanatory graphics. Exploratory graphics are often very simple pictures of your data, built to identify patterns in your data that you might not know exist yet.

In the realm of data visualization, storytelling emerges as an indispensable tool for communicating complex insights. A data story transcends the mere presentation of charts and graphs, weaving a narrative that contextualizes the data, making it more accessible and compelling. It involves crafting a cohesive and engaging storyline that guides the audience through the data, emphasizing key findings and fostering a deeper understanding. By integrating elements of narrative structure, such as a clear exposition, rising action, and resolution, a data story transforms raw numbers into a meaningful and memorable experience

Storytelling is not essential for effective data visualization. There are explanatory and exploratory data visualizations. One tells a story and therefore needs a story, while the other helps to find a story or the lack thereof (which is not a negative outcome). Some dataviz practices enhance both explanatory and exploratory data visualizations, making it easier to discern a story. However, exploratory data visualization (a dashboard) does not necessarily contain a story. And it's not designed to tell a story. The term "data storytelling" is often misused. For instance, when a developer creates a dashboard with numerous slicers (for data exploration?) and claims, "Look, my dashboard tells a story; we sold products worth $50M in 2023".

storytelling serves as a linchpin for effective communication and comprehension. Beyond raw data points and charts, a compelling data story weaves a narrative that contextualizes information, making it more relatable and accessible for diverse audiences. The strategic integration of storytelling principles, such as plot structure and emotional resonance, elevates data visualization from a mere display of numbers to a powerful tool for decision-making. By crafting narratives around data, professionals can bridge the gap between complex insights and the understanding of stakeholders, fostering a deeper engagement with the information presented.

In my experience, if you just build dashboards and give it back to the client then you are just not exploring the full potential of what you have. Instead of only giving back the dashboards, if you build a story around it like why the kpi trend w.r.t previous period, what is contributing to its trend, where can we focus more, or where are we losing traction... This would even help the end user understand and gain insights to call for an action.

The data collected these days is complex, high-dimensional and multi modal. Unstructured analyses may provide answers, but generally not to the desired questions. There is a cohesive structure I like to use, namely, QUESTION > MEASURE > DESIGN > ANALYSIS > INFERENCE. This is the outline to the data story. With each question, one must decide what data (MEASURE) they will use to address that question. The DESIGN is how those data were collected, such as timepoints, treatment groups, dose levels, etc. Based on the question, you then need to decide the appropriate ANALYSIS, the results of which form the INFERENCES that answer the questions.

Before we dive into what makes a good data story, let's talk about the real goals and objectives of the story. Stories help drive conversations and make them relatable to all parties interested and needed. Within a professional setting data stories should inspire conversation, which means stakeholders must leave a data story with takeaways. Takeaways - or data nuggets - are two to three main insights that help drive future conversations that lead to decisions and actionable solutions. An inspired takeaway can be a visual, a talking point, a statistic, or an answer to a question. It comes in many forms and as long as it leads to a meaningful and informed conversation - the story has accomplished its objective.

Don’t start with visualisations. Here’s the process I teach in my workshops: 1. Explore: profile your audience and find insight in your data 2. Structure: create a narrative arc from your insights and add visuals where needed 3. Design: adjust the visuals so that they reinforce the story 4. Present: add more rhythm, tension, and sequencing to the way you’ll present the story 5. Move: add emotion for lasting impact :)

Storytelling is the linchpin of effective data visualization, transcending raw data into a compelling narrative that resonates with audiences. Crafting a captivating data story involves weaving insights seamlessly into a coherent and engaging plot, guiding stakeholders through the analytical journey. By employing data-driven narratives, we transform complex datasets into meaningful, actionable insights, fostering a deeper understanding of the information at hand. Leveraging techniques like contextualization, visualization sequencing, and strategic emphasis on key data points, a well-crafted data story not only informs but also inspires informed decision-making

The creation of a data story starts with understanding the audience, data integration, and defining the scope. In my experience of creating an initial dashboard for storytelling, the approach I like to use is a top-down and left-to-right viewing pattern for organization and usability. The most important metrics (defined by the audience) and visualizations are strategically positioned at the top-left of the dashboard. The supporting metrics are placed adjacent to the key metrics, which creates a seamless flow for users to gain more detailed insights as they explore the dashboard. When building dashboards and projects, I have found it helpful to understand that this is an iterative process, so it's helpful to make the dashboard dynamic.

To craft a great data story, we should start by knowing why we are doing it and who our audience is. Then by selecting relevant data and appropriate visualization methods, supplemented with contextual narrative, we can transform raw data into meaningful insights. Additionally, seeking feedback and revising the story can ensure the narrative effectively resonates with the audience.

Define Your Message: Start with a clear message or insight you want to convey. Know Your Audience: Tailor the story to resonate with your audience's level of understanding. Choose the Right Visuals: Use charts, graphs, and visuals that enhance your narrative. Craft a Narrative Arc: Structure your story with a beginning, middle, and end. Build tension and resolution. Highlight Impact: Emphasize the implications of the data on decision-making or actions.

Examples of data stories include: Impact of Policy Changes: Illustrating the effects of policy changes on relevant metrics, such as economic indicators or public health outcomes. Market Trends: Unveiling market trends and consumer behavior through visualizations, helping businesses make informed decisions. Performance Analytics: Analyzing organizational performance metrics over time to identify areas for improvement and success stories. Environmental Impact: Visualizing environmental data to tell a story about climate change, pollution, or conservation efforts.

One of the latest visualizations that will have changed the view on the world are the the Climate Stripes, created by climate scientist Ed Hawkins. They visually narrate the planet’s rising temperatures. Utilizing my ART model, it grabs Attention with vivid colour shifts from cool to warm, helps Recognize the trend of global warming, and leaves a lasting Takeaway on the urgency for climate action. This striking visual transforms complex climate data into a clear, compelling story, emphasizing the immediate need for environmental stewardship.

Des spécialistes du data storytelling commencent à apparaître dans les entreprises. Ainsi, Pamela Peele, directrice de l'analytique pour la division des services d'assurance d'UPMC à Pittsburgh, a embauché un journaliste chargé de la mise en récit des données.

Data stories serve as powerful vehicles for presenting information in captivating ways. Among the most renowned are the Rosling Gapminder, the Minard Map, and the Snow Map. Hans Rosling's Gapminder employs animated bubble charts and interactive features to debunk global myths. The Minard Map, crafted by Charles Joseph Minard, integrates line charts, maps, and tables to illustrate Napoleon's ill-fated Russian campaign. Additionally, John Snow's Snow Map demonstrates how contaminated water sparked a cholera epidemic in 1854 London, pioneering epidemiology. These data visualizations not only inform but also engage audiences, leaving a lasting impact on how information is understood and acted upon.

Examples range from sales trends impacting revenue to the effects of marketing strategies on customer engagement. Each story focuses on a specific aspect, using data to unfold a compelling narrative.

Storytelling is essential for effective data visualization because it provides context, engages emotions, simplifies complex information, creates memorable experiences, enhances decision-making, and appeals to a broader audience. By weaving data into a narrative structure, it becomes more relatable, understandable, and impactful, enabling better understanding, retention, and action based on the insights derived from the visualization.

I find that a powerful way of understanding is that if you could essentially find data for everything, that you shouldn't start with a data-first approach. Rather, make sure that you have a powerful story to tell that will capture the entire audience. You will then have data at your disposal to make sure that it backs up the story. This is not to say that you should cherry pick data to form a false narrative, however.

Know your audience: Always tailor the story to their needs and interests, language they understand and focus on information relevant to them. Engagement- Find the hook: The compelling question that data is going to answers and frame story around this central theme to keep audience engaged. Visualisation: Charts and graphs can enhance story, but sometimes can be overloading audience. Choose visuals that clearly illustrate key points. 10-Second Rule Don’t create fancy charts and graphs that are difficult to understand for audiences. Simple is best.

I think data storytelling skill entails a little bit of artistic creativity in heart of the data scientist. That is the secret ingredient which makes the visualizations of data more eye catching and interesting to the audience. A good piece of art should be able to speak for itself, same as the visualization of the data. Fortunately we have access to multiple tools such as Python and Tableau that can enable us to create fascinating data stories.

Además de todas las herramientas tecnológicas que se pueden utilizar para hacer story telling, debemos de tener en cuenta los siguientes factores: 1. Usar los gráficos o diagramas que más se adecuen a nuestros datos. 2. Optar por lo simple y minimalista. 3. Mantener el enfoque a través de la narrativa. 4. Usar ejemplos claros y sencillos para ilustrar los puntos más relevantes. 5. Prestar atención a la carga cognitiva que refleja nuestra presentación.

In my experience, data storytelling has helped simplified trends and narratives from data analysis. Data storytelling helps the audience (especially those who aren’t tech savvy) to better comprehend data visualizations with ease.

Sempre considere o seu principal público-alvo e como ele consome/analisa os dados e os transforma em decisões. Isso é importante para construir a melhor história possível. Por fim, algumas sugestões de excelentes livros para aprender mais sobre o tema: Effective Data Storytelling (Brent Dykes), Data Story (Nancy Duarte) e The Data Storyteller's Handbook (Kat Greenbrook). -- Always consider your main target audience and how they consume/analyze data and transform it into decisions. This is important to build the best story possible. Finally, some suggestions for excellent books to learn more about the topic: Effective Data Storytelling (Brent Dykes), Data Story (Nancy Duarte) and The Data Storyteller's Handbook (Kat Greenbrook).

One important thing I would like to point out is that to truly excel in data storytelling, do ensure you get out of the data bubble! Get out there and talk to non-data people and get their feedback. All the best in your data visualization journey!

At its core, using stories when showing data is crucial because that's how humans are wired. Since ancient times sharing tales around the fire, we've relayed information through narrative. Data alone can seem cold and confusing. Stories make data relatable! By combining statistics with characters and narrative, we leverage how people process information. We connect with the challenges the characters face, drawing meaning when numbers demonstrate tangible impacts. Structuring the data in a logical flow helps audiences grasp complex insights more deeply. The key is to balance storytelling to engage audience without distorting truth. Used right, stories help numbers inspire and empower. We become invested in the data, enabling lasting change.

Data storytelling is also like a movie story. In my experience I would say start the data storytelling… 1. By asking a question to your audience, this creates a visionary clue and what can they read from the data. 2. Introduce the hero of the data and show how the hero is driving the product performance. 3. Dive into the story with the vilan example and show how it is hitting your product. 4. Finally close the story by highlighting the small data points that helped to mitigate your product performance and show the charts that visualises the same