Storytelling (not Showing Off) With Data
This is the Three Data Point Thursday, making your business smarter with data & AI.
Actionable Insights
If you only have a few minutes, here’s what’s going to make your business smarter with a book on data.
Everyone should have skimmed this book: “Storytelling with Data: A Data Visualization Guide for Business Professionals” by Cole Nussbaumer Knaflic.
Analysts need to study this book, work through it, and take a week to redo their graphics.
Neither our world nor data is 2-dimensional. Yet we make it so by using dashboards and tables, plain old, mostly 2-dimensional representations. Don’t do that. (1) Realize our world and our decisions are multi-variate. (2) Use data to tell a story, to make a point about a large dimensional space, not to impress.
Feel Good button, click once, feel good twice! (One reader shared the last edition 29x!)
Chartjunk! That word is a good description of most of the trillions of statistical graphics published in reports, companies, and on the Internet each year. It is probably not the best marketing slogan for Tableau, Looker, and all the other business intelligence pioneers, but it is true nonetheless.
Our world is inevitably multivariate, and that’s exciting! Yet we tend to try to squeeze it into little two-dimensional paper format, derive decisions, and then wonder why they don’t work out. Or why no one is listening to our reason in the first place! After all, we have all those graphs and tables to convince people - right?
I think Edward Tufte would be with me on this point when I say if you try to reason with data today, you fall into either one of these two camps:
The set of people who try to convince, using data, try to push for a certain decision, using tables, reports, and dashboards
Or the ones that decide based on data using tables, reports, and dashboards.
The problem? That’s not how data works! We’re missing the crucial third category: The people who understand data isn’t two-dimensional; our world isn’t two-dimensional; we make it so and hope for the best. Well, think again: flattening everything with a hammer isn’t going to work forever.
Enter data storytelling
Luckily, there’s a great book trying to move the two aforementioned categories of people into a more plastic world - one that sees storytelling and multivariate statistical graphics as the gold standard. It’s called “Storytelling with Data: A Data Visualization Guide for Business Professionals” by Cole Nussbaumer Knaflic.
“Having all the information in the world at our fingertips doesn’t make it easier to communicate: it makes it harder.” - Cole Nussbaumer Knaflic
We all grew up in a world of data, with an abundance of information, which can make it hard to appreciate and make the most of it. It’s like an all-you-can-eat buffet. If you’re at that buffet, do you arrange your food nicely on your plate? Most people do not, they fill their plates all the way, never being able to appreciate, let alone communicate the quality of the food.
This book offers a stab at a solution: Five practical tips to pimp up your reports and dashboards, usually turning them into something larger than a report/dashboard, a story.
These five practical things you and I can apply to every single statistical graphic we have ever produced to make it better. Make it tell its story in a fuller way.
Let me quickly recap the tips. Don’t worry. The content is great, and I’m not stealing anything here; if you want to implement any of this, you’ll need to read the book. Context & practice are everything:
Add context. Growth is nice. But what is growth compared to last year? Better.
Choose a visual that conveys your message. If you want to show a decline, make sure your visual is all red and turning down, not up!
Declutter. Throw away Q1 and Q2 if there’s nothing new. Keep the data you need to make your point, but make that part high-quality and clear.
Focus your audience on your point. Use highlights! At best, only one.
Make your graphic easy to access; don’t let someone look into the footnotes to find a definition.
Make a difference
Sadly, Cole waits until chapter 7, “Storytelling with Data,” to reveal the most important point of the book: Storytelling matters! In fact, it’s all there is. The whole point of having graphics that tell good stories is to tell a grand story itself, one that makes a decision simple.
I think the easiest way to explain the difference is the very first graphic Cole shares.
While the book is a treasure trove of great storytelling, I don’t think it’s a book everyone needs to study all the way.
Who should read this book?
This is a book that will give you two things: A hope that statistical graphics can be made much better and the practical lessons you need to make it happen yourself.
Everyone who works in data should have skimmed through this book to fine-tune his bullshit detector. Keep it at your library to pick it up whenever you look at a graph.
For every analyst, I recommend taking a week to truly digest it. Take a look at your old work, and redo it! Declutter, highlight and tell a story with your data. I promise you that just a week of practice is going to elevate your skill as an analyst.