Data Storytelling: A Critical Skill for a Data Scientist

Numbers have an important story to tell. They rely on you to give them a clear and convincing voice.” –Stephen Few

If you thought critical business decisions are based on data and other quantitative factors, you are far away from the truth! The best decisions are taken when numbers meet logic, reasoning and emotions. But the question arises, can numbers convey the message alone? The answer to this is No. There has to a story woven around the Data to persuade audiences faster.

What is Data Storytelling?

Data Storytelling is a very critical skill for Data Scientists that find little mention. It is nothing but the art of weaving a coherent story around the numbers that convey the logic, strike the right chord with the stakeholders and convince them with reasons appropriate enough to drive a decision. A data scientist’s motive is to ensure that the actions are taken on the insights that he/she has drawn from the data. This depends a lot on how well he has presented the data to the end-users or decision makers. Data Storytelling is all about presenting the right information in an appropriate format at the most opportune time.

What are the various types of Data Stories?

1. Reporting

This is the simplest kind of stories to start with. Such stories are data rich and only state facts relating to the data. It is more of data commentary. However, the facts need to be presented in a coherent manner that puts across a point and decision makers can easily draw inferences from it.

A typical example of this is a sales report story which states the number of sales or amount of revenue that has happened in each quarter, in each geography and across each of the product type. The details can be even more. In such cases, the story presented needs to be woven with charts, graphs and comments supporting them.

2. Explanatory

Many times a business problem is so complex that we need loads of data to understand it. Explanatory data stories explore the business problems, scrutinize the nitty-gritty, helps the decision makers to identify the real issue at hand. The story needs to be presented in such a manner that it can help the audience to understand the reason behind the problem through data.

For Eg: A telecom company faces sudden large-scale network issues, call drops and many subscribers opt out of their services. In this case, an explanatory story will focus on the details of the problem such as frequency of call drops/network issues, the areas will be precisely identified, the technical details towers, data cable will be highlighted, the details of the handsets used by customers will also be scrutinized. All this explanation when represented in a sequence through graphs and charts will help organisations to get to the root of the problem and take corrective actions.

3. Predictive

This is one area where data scientists are most needed. They analyse the problem, find trends, draw insights and also do predictive modelling to forecast future happenings. A predictive data story should connect the dots for the current trend being depicted, find the underlying pattern and reason for the same and predict the future course of events. The story should have a natural flow of events.

Essentials of a Data Story

Why do organisations need Data Storytellers?

  • As data gets deeper and more complex, it is becoming more important to bring in simplicity. For this reason, good storytelling becomes a very crucial component of business intelligence (BI) tools.
  • While data and analytics may reveal great insights, an absence of narrative makes it difficult to relate to the facts.  Data storytelling goes beyond Data Visualization. A story with a real-life example or personal experience will help the end-users to find a context to the numbers and lend more meaning to the analysis. The entire business problem at hand will be simplified, made more relevant and interesting.
  • No matter how high-quality data you use or how in-depth your analysis is, it will not steer a decision maker into action unless they understand the logic behind it. With the use of attractive visuals, right format and a strong narrative, decision makers will gain required clarity of the issue and understand the most appropriate action to be taken.

What are the skills required to be great Data storytellers?

1. Knowing the audience and adapting the story to their needs

One of the biggest mistakes in Data storytelling is making an assumption of one size fits all. The story has to be tailored to the sensibilities of the audience and their level of understanding. It is still acceptable if the story doesn’t engage audience, but if they are unable to understand it, then there is a problem with storytelling. So knowing the target audience is of utmost importance.

2. Understanding the business problem

Before weaving a story to address a business problem, it is important for a Data Scientist to understand the problem very well. Without proper understanding of the issue, the scientist will not be able to present the story in the correct context. Storytelling is done with the sole purpose of expediting the decision making process, so it imperative that the business problem is broken down to the most granular level and to facilitate making a clear and concise story.

3. Identifying the probable questions and preparing answers

If the story presented is interesting enough, it is sure to attract loads of questions from the audience as well. As a good storyteller you need to be always prepared with the answers to the possible questions. A good story spurs inquisitiveness and the Data Scientist should have answers backed by relevant data to drive decision making.

4. Getting the right data at hand

You need to always ensure that you have the right data to tell your story. While a good and compelling story may be great for persuasion and striking an emotional chord, incorrect data will spoil its credibility. As a good storyteller, it is of utmost importance for you to balance beauty of a storyline alongside factual accuracy.

5. Good presentation skills

A good storyteller knows how to present the story with appropriate visualisations such as charts, graphs, maps, infographics etc. This makes it easier for the audience to spot trends, patterns and the get underlying message. Along with enhancing the visual appeal of the story, it makes persuasion easy.

The Final Word

Storytelling is not as easy as it appears. It takes a great deal of visualization, creativity and good presentation skills to ace the art of Data Storytelling. Most Data Scientists are trained to just stick to numbers and analytical skills and are not quite aware of the requirement to weave a good narrative around their analysis. However, it does pay to develop this critical skill because it adds a great deal of value to your analysis. Good Storytelling only helps decision makers to make sense of your analysis faster, be fully convinced with your findings and forecasts, and take the right course of action at the right time.