Dashboards and Data Story-telling Explained!
These two concepts are probably at the highest level of Data Visualizations and the most sought after. We often think of Data Visualization to be just about graphs and different types of beautiful plots. But that cannot just be it. There is something beyond just Data Visualizations and that can help your reports or analytics stand out from the crowd. While Dashboards have been around for a while and there is a probability that we must have used Dashboards without even realizing that we were doing so, Story-telling is a new concept that has emerged recently.
While Storytelling relates to a literary term, it’s use in Data Visualizations is similar to literature. We weave stories and tell them in ways that can make the story very interesting and can show you the various characters and the plot of the story and the importance of each and every character and element in the story. That helps us understand the story in a way that’s better than usual and can help it stick with us to recite to anyone else as and how it was understood by us. All in all, it is done to enhance the effect a usual story has on the reader.
Think of the same thing but in a Data Visualization scenario and place the characters in your mindset to be the multiple variables contributing to the story, the plot of the story to be the baseline we’re trying to explain or the situation a company is facing or just another scenario. The story is told about the various characters and how they impact the plot (which is shown specifically using annotations as we’ll understand ahead).
Curious about how it works yet? Let’s go over an example of Dashboards and Storytelling and understand how they work (Note: The following visuals have been created using Tableau Public)
We’ll use the same example as we did earlier in my previous article titled “101 Reasons to Visualize Data” (the link for which is in the end of this article) which is as follows:
You are analyzing a dataset of customers of a relatively new Bank situated in the United Kingdom and studying the customer data to find out more about the customer base and help in targeting them accordingly in the future.
You have the Name, Surname, Gender, Age, Region, Job Classification, Date of Joining and Bank Balance of 4014 customers in your database. Since understanding these on the basis of just the entries and using some statistics would provide you with information, it will be limited. An easier and more efficient method can be used by visualizing the data and making use of Dashboards.
First, Let us look at some basic visuals:
Whilst these visuals look beautiful and do communicate how the customers are distributed in terms of Balance, Job and Age; they are still singular visuals that do not really communicate information clearly. You can understand how the no. of customers decrease as the balance increases in the customer’s account or how White Collared jobs are the majority of our customers and that they come mostly in age groups of 30–40.
DASHBOARDS:
Let us look at a better way to do this which uses some parameters and filters using Dashboards:
As shown in Dashboard 1, Dashboards are a compilation of all your visuals and can communicate way more information than a singular plot. This Dashboard shows us how many customers are there in our entire database and how they are distributed across variables like Age, Balance etc.
We can further manipulate this down by filtering data. Here are 3 examples for filters on this Dashboard:
Filter 1 : Data of customers in the age group of 30 to 40 years. *Dashboard 2
Filter 2 : Data of customers situated only in Scotland in the UK. *Dashboard 3
Filter 3: Data of customers whose balance exceeds 100K. *Dashboard 4
As you can see, the Dashboards when used with filters tells us information that would’ve taken a lot of time to extract using just simple visuals. From the first filter it’s easy to understand that a majority of our customers in the age group of 30 to 40 years are in England and are equally Male and Female. The second one shows how there’s relatively more senior citizens in our customers coming from Scotland. The third one surprisingly shows us how a majority of customers with a balance greater than 100K came from the age group of 25 to 35 which is a relatively younger population.
Now, you must have realized something. To communicate information about Data of this volume, you’ll need multiple Dashboards and that will take a lot of time and effort to prepare. This drawback of Dashboards which is pretty much negligible compared to the advantages led to the development of what is known as Data Storytelling.
As discussed earlier in the article, a story is woven around the required variables and annotated to be more understandable in order to communicate your findings with your audience. There are minute details that matter in terms of presenting the data in a story as well. The way you enter it, the visuals you use, the colors you use, the size of texts etc. everything makes a difference and upon which we’ll try to touch in another article as it is beyond the scope of this one.
DATA STORY-TELLING:
So let’s go over a example of a story and understand what it is all about practically.
Here we first begin with a simple map of UK to start the story and show how our customers are scattered in different regions of the United Kingdom. Text boxes here denote what the analyst wants the audience to observe and to show numerical figures which are not clearly visible. It is essential to start a story from the beginning and then diving deeper into specific concepts.
In the second part we dive deeper into Wales and understand our customer base there. The part where the analyst wants to show emphasis is shown using an Annotation.
We then dive deeper to understand them based on their age group.
Finally, we see what are the distributions like when it comes to female customers in Wales.
As you can see here, the story with annotations is self-explanatory and can give out enough information to know where the company has to invest in order to efficiently gain profits over time and expand. Data Storytelling is just an upgrade to the routine Dashboards and can exquisitely show patterns and give out information as and how required by the audience. It is one of the major tricks that every Data Analyst needs to have in order to communicate information about data easily and precisely.
This is essentially everything you’ll need to start out with simple Data Storytelling. Use annotations aptly and don’t over-use them.
Tableau, as used in this article, gives out a clean and easy way to present this Story as shown and does not take more than half an hour for these 4 parts provided you have clean data. I recommend you to give Data Visualizations a shot with this tool.
101 Reasons to Visualize Data:
Tableau Public:
For more such articles, stay tuned with us as we chart out paths on understanding data and coding and demystify other concepts related to Data Science and Coding. Please leave a review down in the comments. It was a long article, thank you very much for reading it all the way here! Great going!
