Storytelling with data — Part 1
People are tired of me advocating about Learning How to Learn and the incredible Dr. Barbara Oakley, I get it. However, whenever asked about which skill was the most important to develop my career, my answer without a second thought is: I got good at learning.
What do I mean by good at learning?
I mastered the intricacies of how to learn with efficiency (in a short time) and efficacy (retaining the most), and it’s no rocket science. A series of best practices and healthy habits that I incorporated, most of them approached on the best MOOC available, you know the name.
But why the title is storytelling with data and we’re talking about learning strategies? So, I’m working to add another level of complexity to my learning process, recalling and teaching. Dr. Oakley in a talk with Jonathon Kendall, COO of Mentor Box, explores this idea that if you can practice the recalling technique to teach someone this is a highly effective way to chunk the material in a way you become capable of quoting the material. Furthermore, Jonathon adds that during his practices, whenever he doesn’t find the time or people to teach he finds himself writing down the textbook in a piece of paper simulating a class, getting comfortable with the material, being able to recall and reproduce.
Therefore, the idea is the following: I don’t find the time to keep teaching what I’ve been reading, but at the same time I feel the urge to fully comprehend my studies, bringing theory to practice I’m sharing my understandings and insights from some of the chapters of the book Storytelling with Data: A Data Visualization Guide for Business Professionals, from Cole Nussbaumer Knaflic an ex-Google employee and data visualization expert. Enjoy the part one of a possible series.
One big idea in context.
The two key concepts of the first chapter from Storytelling with Data revolve around fully understanding the context of the communication intended aided by data and being able to encapsulate the speech into a single “big idea”.
Context is everything
I had bad working experiences in the past where my presentation tone was not aligned to the public I was targeting, it wasn’t lack of understanding the material, but I didn’t know my public in depth to provide the right context. There’s no point of getting extra technical with stakeholders concerned with the final output. Context is everything when conveying information.
Cole suggests a quick framework to get comfortable with context parameters, ask yourself the following:
Who: Who is the public receiving the intended information and how they usually interpret this kind of information. Do you have a trust established with this public or you’re building it up?
What: What do you want your public to know or do, in other words, after this communication what is the expected reaction and the following actions? As the analyst, could you suggest the next step to be taken?
Pro tip: As a best practice when action can’t be suggested, guide your public towards an action plan discussion.
How: Two scopes for the how question, the medium, and the tool. In terms of medium analysts usually have live presentations or team meetings to showcase their findings or sending them over emails.
Pro tip: Whenever possible to chose, meetings are the best choice for the diligent analyst. In a confusion matrix comparing meetings and emails regarding the information control and necessary detail level meetings have high information control, you guide your public towards your conclusion, and require a low level of detail since questions raised are answered on the spot.
The big idea
My favorite analogy for the big idea concept is that it is the elevator pitch of a full analysis, where the entire storytelling is summarized by a unique point of view, conveying the conclusion and suggestions in a complete sentence. To clarify the concept let's recap one of my last visualization design done for one of my clients.
We will come back to this example in the near future as a reference to the other teachings from storytelling with data.
A practical example of data storytelling studies
Period of analysis: August 2018 to January 2019.
Purpose of analysis: Understand users distributions on competitors websites.
Focusing on data storytelling instead of the technicalities of the inference process we have a working dataset extracted from Google Analytics and the suggested by the Explore feature of Google Sheets when data was copied on the spreadsheet.
Many people would say that the stratified bar graph it's more than enough to inform our point, and I agree to disagree.
After reading through the chapters and understanding concepts such as visual perception principles, the cognitive load generated by visualizations, saturation, and pre-attentive attributes I ought to say that I'm now a designer when planning a visualization, not only an engineer.
On the table above, the competitors' websites were removed from to preserve data security.
Now you might be asking yourself if names are labeling the x-axis does the legend mean the months? Exactly, and this is just one of a series of problems caused by this poor chart design.
Note how difficult it is to compare month performance across competitors, feel the number of unnecessary eye movements!
I will cover all improvement points and the reasons why in the next post while covering new topics from the book. For now, have fun comparing the before and after Storytelling with Data from Cole Knaflic.
Last but not least, the big idea:
Our analysis showed that in terms of traffic our direct competitor is currently XXXXXX and. We did not expect that the other competitors would be in such an advantage, therefore we suggest continuing the analysis with an SEO profile comparison between the top 4 players.
I'm constantly publishing on LinkedIn if you're interested in Digital Marketing Data Science topics in general.
For more productivity, business tips, and bad nerdy jokes check more of our stories at ProductiveTalk!
This story is published in The Startup, Medium’s largest entrepreneurship publication followed by +423,678 people.
Subscribe to receive our top stories here.
