Forbes

Brent Dykes

Adventures In Data Storytelling: Three Key Traps To Avoid

As your organization seeks to become more data-driven, data will become an integral part of all communications. It should play a key role in your business meetings, internal conversations, and other daily interactions. As more of your people are exposed to data that is both relevant and crucial to their roles, communicating effectively with data will be an essential skill everyone will need to master – from the CEO to frontline employees.

 

A data story is an adept combination of data, narrative, and visuals that seeks to explain, enlighten and engage.

 

A data story is an effective combination of data, narrative, and visuals that seeks to explain, enlighten and engage.

Whenever you combine or arrange data and other elements for a specific purpose – analysis or communication – you’re creating a data composition. Today, data specialists are no longer the only ones being tasked with creating various data compositions such as data presentations, reports, dashboards, infographics, data visualizations, and so on. Data stories are a unique set of data compositions that blend data, narrative and visuals in an effective manner to help explain, enlighten and engage. Data stories are powerful vehicles for sharing data insights to influence and drive change within an organization.

However, there’s more to a data story than just a flashy set of data visualizations or a compelling narrative. To a casual observer, many data compositions may look and smell like data stories because they include numbers and charts. While all data stories are data compositions, not all data compositions are data stories. On closer inspection, you’ll notice many supposed data stories are missing some of the essential attributes that constitute an effective data narrative. Flawed data stories can end up impeding business decision-making as their weak narratives, biased data and confusing charts fuel bad decisions or indecision. To avoid falling into potential data storytelling traps, it is important to first understand the two-step process in which data stories are formed.

Indiana Jones and how data stories are created

To help explain how data stories are developed in two stages, I’ll compare the process to my childhood hero, Indiana Jones. As a field archeologist, Indy would travel around the world pursuing rare antiquities and battling evil Nazi henchmen. On the other hand, he was also a college professor who would share the tales and spoils of his archeological adventures with his students, fellow faculty and university museum.

 

The formation of a data story begins with using exploratory data visualizations to discover insights. Once a meaningful insight is uncovered, explanatory data visualizations are used to tell the story. A data story checks all the boxes on data (D), narrative (N), and visuals (V).

 

The formation of a data story begins with using exploratory data visualizations to discover insights. Once a meaningful insight is uncovered, explanatory data visualizations are used to tell the story. A data story checks all the boxes on data (D), narrative (N), and visuals (V).

Similarly, all data stories begin with an exploratory phase where someone examines a dataset for meaningful insights. At this stage, you’re the daring, inquisitive adventurer, Indiana Jones. This initial phase can be as straightforward as studying a simple data table or as complex as performing a series of advanced statistical analyses. Data visualizations will often serve an integral role in helping you to uncover key patterns, trends and anomalies in your data. Speed and flexibility drive these initial data visualizations as you navigate through the data, exploring different paths and perspectives.

At this stage, you are the sole audience of these exploratory data visualizations so they only need to speak to you and no one else. As you assemble the data for these charts, you gain a more intimate knowledge of the underlying data. While you may seek to confirm a hypothesis or hunch, you won’t yet know what narrative will emerge from the data. Your initial ideas may be proven right, completely contradicted or even pulled in an unexpected direction.

However, once you discover a meaningful insight you now have a story to tell and must transition to the explanatory phase. You now become Dr. Jones – professor of archeology and popular classroom lecturer. Now, you are no longer the audience as you seek to share your key findings with others. There’s also a very good chance your intended audience won’t know the data to the same degree as you do, so simplicity and clarity become essential elements in your usage of explanatory data visualizations. Notably, the data visualizations that helped you to discover an insight will often need to be refined or even replaced to more effectively communicate what you discovered to a less-informed audience.

With this understanding of how data stories are formed, you’re ready to examine the three key traps that people can fall into when attempting to tell effective data stories.

Trap #1 – the Data Cut

Everything starts off the right way as you begin by slicing and dicing the data to discover a meaningful insight. However, once a particular cut of the data yields an interesting insight, nothing is done to then package it up for others. In this scenario, you mistakenly assume because the raw information speaks to you, it will speak equally well to your audience. Unfortunately, like an unedited director’s cut, the data cut leans too heavily on the impact or persuasiveness of the raw facts. It ignores the importance of having a well-crafted narrative and explanatory visuals to help others better understand the insight’s significance.

 

A data cut starts off right by exploring the data for insights; however, it fails to communicate the insight using an effective narrative and explanatory visuals.

 

A data cut starts off right by exploring the data for insights; however, it fails to communicate the insight using an effective narrative and explanatory visuals.

Trap #2 – the Data Cameo

Interestingly, this next trap is rich in narrative and showcases some key data points. However, it begins with a predetermined story – or perhaps more accurately an agenda – and then looks for supporting data. Only data points that uphold the desired narrative are selected while conflicting ones are ignored – either intentionally (selectivity, omission) or unintentionally (confirmation bias). This approach is common whenever someone feels they must justify a decision or show why a particular initiative was successful. Unfortunately, data only makes a cameo appearance in this narrative – more for show than as the foundation of the story that it should be. When data isn’t central to the overall story, a data cameo can quickly unravel under closer scrutiny.

 

The data cameo starts with a pre-defined story – not data. Various data points are selected to bolster or substantiate the desired narrative. Without a solid data foundation, the data cameo can quickly unravel under closer scrutiny.

 

The data cameo starts with a predetermined story – not data. Various data points are selected to bolster or substantiate the desired narrative. Without a solid data foundation, the data cameo can quickly unravel under closer scrutiny.

Trap #3 – the Data Decoration

The last trap has emerged as more individuals gain access to data visualization tools. People now have access to more data than they know what to do with, and they can display this data in lots of different ways. This addictive combination of limitless data and graphical eye-candy has led to the emergence of data decorations. This scenario occurs when individuals stumble through the exploratory phase without identifying a clear insight, and then jump ahead to visualizing data for others without crafting a cogent narrative. By simply sharing the data charts and dodging the real analysis work, they hope someone consuming the data will somehow find something meaningful. However, rather than adding value, data decorations can often just end up adding unwanted noise.

 

With a data decoration, insufficient time is spent on actually analyzing the data for takeaways or insights before visualizing the data for consumption.

 

With a data decoration, insufficient time is spent on actually analyzing the data for takeaways or insights before visualizing the data for consumption.

Each of these data storytelling traps excels in one key aspect of what is needed to form an effective data story but has flaws in the other two essential areas. For example, the data cut is strong on data but is weak in both narrative and visuals. The data cameo is rich in narrative but suspect on data. The data decoration offers appealing visuals but lacks a focused narrative. Only a true data story combines all three key aspects – data, narrative and visuals – effectively.

 

As a subset of data compositions, only data stories combine all three aspects of data, narrative and visuals effectively. The other data storytelling traps are only strong in one of the areas needed to be truly effective.

 

As a subset of data compositions, only data stories combine all three aspects of data, narrative and visuals effectively. The other data storytelling traps are only strong in one of the areas needed to be truly effective.

As more and more of your people leverage data in their roles, it is imperative that both managers and employees learn to distinguish between real data stories and other less-effective approaches. While data offers tremendous potential for your organization, it won’t lead to the right business decisions if it’s not properly analyzed and communicated. Without effective data storytelling, you may end up “digging in the wrong place” as Indiana Jones once noted – lots of activity and noise with little success. However, when you base the right narrative and visuals on the right data, you have the power to inspire positive change and improved performance. Your data has all kinds of great stories to share – it just needs someone to find and tell them. And so the adventure begins.

 

This article was written by Brent Dykes from Forbes and was legally licensed through the NewsCred publisher network. Please direct all licensing questions to legal@newscred.com.