Data Storytelling is the art of telling a story through data. This is the evolution of Data Visualization, the last step of that long analytical process which has the purpose of transforming data from various sources of corporate interest into integrated and coherent information, and finally selecting interesting, unexpected and valuable information and bringing them to end users (business users, marketing, and in general managers and managers of the various sectors and departments) in a concise, clear and intuitive way. All this is functional to the setting of Data-Driven strategies aimed at making (or keeping) companies competitive even in increasingly dynamic markets full of competitors by relaunching them with a practical, valuable and innovative offer.
Data Storytelling: Definition, Strengths And Weaknesses
The definition of Data Storytelling is now clear, at least in theory. Despite being a well-known and widespread concept, there is often confusion about “how to do” Data Storytelling and about which and how many advantages a well-told “story” can bring. This is because there is so much wrong or incomplete information circulating about it that undermines the very concept of Data Storytelling: in this condition, it remains a vague and smoky concept, even if, probably, it is one of the most concrete activities for the business.
Trying to decline the definition in practical terms, we can say that Data Storytelling consists in reading through thousands of pieces of data, observing the relationships that bind them, looking for the significant patterns that unite them and finally exploiting Data Visualization techniques to “tell” them to the end user through, in fact, a story designed to highlight and give emphasis to the information that is significant for him in the specific context, those that can guide towards decisions, choices and conscious actions in a Data-Driven perspective.
The strength of good Data Storytelling lies, in fact, in the creation of added value, which is achieved at the end of the analytical process when the end user, looking at the result, acquires in a short time the knowledge necessary to understand what it is happening transparently, placing the information in the corporate context, without raising doubts or misunderstandings about the meaning of the data it reads and without “wasting time” having to work hard to interpret it.
On the other hand, and for the same reasons, all data transformation and analysis operations that lead to obtaining the “story” to tell are potentially at risk due to not-so-rare problems of understanding data, interpreting information in the corporate context, of inability to find the suitable representation, which is at the same time speaking, complete and precise. For this reason, it is essential to dedicate adequate time and resources to Data Storytelling in terms of both professional skills and Data Visualization technologies and tools to avoid obtaining untrue or incomplete storytelling.
Also Read: Storytelling On Social Media: 10 Creative Tips To Attract Audience
How To Do Data Storytelling: 6 Commandments
We can summarize in a few guidelines the steps and concepts that lie behind correct and useful Data Storytelling: here is a list of 6 “commandments” and 6 points of attention to keep in mind to develop a success story.
Start With The Right Questions
To organize a successful, valuable story for the end user, it is essential to start by defining the questions you are looking for an answer to. It’s about more than drawing the final output’s definitive boundaries. It’s still too early for that, but it’s crucial, if nothing else, to organize the story structure we will create. It can be helpful to answer the question: “What will the audience learn from my story?”
Conclude With Insights Appropriate To The Context
If the necessary tools are not obtained at the end of the process to know the data and learn something useful. All the activities above will be in vain. It will not be worth investing resources to tell this story.
Find A Compelling Story And A Suitable Representation
It is undoubtedly more accessible for business users to remember a story than data and numbers. This is why an excellent initial selection and finding the common thread that connects the data and creates the story is essential. For the same reason, it is also necessary to find the most suitable graphic representation: generally, users remember visual elements better than tables full of numbers. On the other hand, not all images speak volumes by themselves, so it is good practice to find the right compromise and exploit the intuitiveness and expressive power of the pictures, possibly accompanied by texts and numbers that detail them and, if necessary, contextualize them.
Clarity And Brevity
Anything that is not part of the story, that does not add valuable details that do not help contextualize, must be left out. Few exciting pieces of information. Here’s what the end user needs. All the “additional sugar” only risks being misleading and diverting attention from the analysis’s central concepts. In some cases, to increase the comprehensibility of the data, it may be helpful to place them even more in the corporate context, for example, by exploiting KPIs or graphical representations that allow the data to be compared over time to compare them with those of previous periods or with market trends: the figures, if contextualized correctly, can quickly transform from “simple numbers” to information with very high added value.
Credibility And Trust
For the end user to trust the output and the story without hesitation, it must be transparent and honest. Sugaring negative results or hiding them behind unclear representations is not only very immediately harmful to the business, but it nullifies all the work done (and to be done) because it destroys the reader’s trust.
Ultimate Goals? Providing The Correct Information To End Users
This is because, regardless of the complex and soft skills of Data Engineers and Data Scientists, the ultimate goal remains that of satisfying a need, responding to the needs of end users, even when it comes to accompanying the story towards a less obvious destination. For example, in the retail sector, we expect all possible levels to concern purchases. Still, in reality, the focus must be different according to the recipients of the story. Suppose the end users are the IT technicians who manage e-commerce. In that case, a behavioral history of the actions taken on the site could be more helpful. If the story is aimed at marketing, a user profiling history could be more suitable for personalizing activities and campaigns.
Also Read: 5 Best Data Storytelling Resources For Bringing Data To Life
