There's a new book on data storytelling, and it's incredibly good.
Welcome to my content recommendations and reviews for Sep-21: a couple of books, a podcast, articles and a couple of videos.
Let's get started.
I've been studying the craft of data storytelling for more than 15 years now and have been deeply influenced by several books.
Allow me to take you through a brief history of my learning, in the order in which I read the books (The years in the brackets denote when I read the book, not when it was released).
"The Mckinsey Way" (2005) was handed over to me by one of my bosses at my first job, post MBA (I think it was Monika!). It blew my mind. I mean, it laid out, step by step, the entire consulting process. My biggest takeaway from the book: you need to have a set of intelligent hypotheses when you are researching and analysing data.
"Presenting to Win" by Jerry Weissman (gifted by another boss, Mukesh) is a good, though now dated, read. It taught me the importance of figuring out the right start point (Point A) for your data story.
"Made to Stick" (2009) by Chip and Dan Heath is the most influential book I've ever read. I learnt the importance of six key story elements - which I still remember because of the easy mnemonic used by the authors. As per the book, sticky ideas are:
Simple to understand
Unexpected (the most powerful concept)
Made to Stick was gifted to me by my colleague and good friend, Rohan Desai and in turn, is the book that I've gifted the most to others! It's a classic.
After Made to Stick, there was a longish break, and I picked up the next set of books once I started formally training the subject.
"Storytelling with Data" (2016) by Cole Nussbaumer Knaflic has the distinction of being the first to coin the term in its title - and subsequently, hoard all Google search results of that keyword!
The book is incredibly useful in one very specific (and narrow) aspect of data storytelling - how to create visual hierarchy in your slides, specifically charts, so that readers can 'get' your data-story quickly.
However, for a book called 'Storytelling with Data' the book is unpardonably weak on the 'Narrative' aspect - which to me is the most important element of the craft.
The immense success of this book is a testament to the dominance of data-visualisation in the data-storytelling space. It also tells you something about the concept of 'storytelling': the blessed word has several interpretations. Every person has their own definition of the term!
Net net, "Storytelling with Data" is great for learning about visual representation and hierarchy. But if you pick it up hoping to get a comprehensive primer on the subject, best of luck! It's a classic case of an evocative description of the trunk of the elephant... being presented as the whole elephant.
"The Pyramid Principle" (2016) by Barbara Minto is the first (and last) book on story frameworks and logical writing you will ever need. Essential stuff.
Having said that, for a book on writing well, it is not... written very engagingly...?
Then came "Data Story" (2019) from the storied house of Duarte (the Mckinsey of the storytelling world?). It came at a time when I had gotten fairly established in the field and was hungry for more interesting material, so I had great expectations from the book. I was like, "C'mon, wow me".
You could say that my expectations were .... met, to some extent?
Don't get me wrong. I think it's a great book for someone getting into the craft. In parts it is sublime... in parts banal and in some parts it is surprisingly mundane (at one point, it gets into a long list of exclamatory words you can use to express surprise)!
Overall, for someone steeped into the subject for several years, it is... a bit underwhelming. I did incorporate a few new practices (especially on the visual front), but it did not alter my thinking in a foundational way.
Brent Dykes' "Effective Data Storytelling" (2021) kinda did that.
Brent (who I subsequently managed to interview for my podcast - yay! - episode coming later) is a deep, fundamental thinker. He gets data and is also a keen and curious student of the craft of storytelling.
Brent's assertions in the book are based on rock-solid research. On top of that he has also come up with some clutter-breaking insights. Let's dig into a few of them.
Key insights from the book
Here are three most striking ones (all quotes from the book):
Brent has a simple three-part structure for a good data story: Data, Narrative and Visuals.
I've always had a three-part structure too: Narrative, Visuals and Delivery.
So, mine didn't have 'data' and his doesn't have 'delivery'.
Now, I still believe that 'delivery' (your verbal presentation) is a critical element, so I will retain that. But when I saw 'data' among his list, I was like - isn't that obvious? Wouldn't every data story, umm, have data?
Often, when we are presenting our data findings, we have an agenda to push, a point of view to share, an unconscious bias that drives us - essentially a narrative in our heads that influences the story.
Make no mistake - it is critical to have a narrative. In fact data without narrative is the great shortcoming that plagues almost ALL data-heavy presentations in the corporate world.
But there is a problem on the other end of the spectrum too. We often have a narrative in our heads and don't bother to back it with data.
I remember a real-life instance. In 2017, I was working with a client in the air travel space and one of the division managers (let's call him Ankit) was in-charge of the Air-Cargo business to East Asia. In the previous quarter, his division's revenue had seen a smart increase. The reason? Well, Ankit was convinced it was because of the new dynamic pricing policy his team had implemented that quarter.
Turns out, he was wrong.
A closer examination of the data revealed that much more than price, it was a change in the mix of cargo that had had the maximum impact on revenue. For Ankit though, his mind had been made - his narrative overrode the data.
In the book, Brent devotes an entire chapter to the importance of having a solid data foundation. He says that your data sources must be:
He also lists down how to avoid some common fallacies in the data gathering and analysis process.
2. Messages < Insights < Valuable insights
For the longest time I have been teaching the importance of identifying and crafting clear messages from the data. While that remains important, I learnt a nuance about the different types of messages from this book.
Often we use the word 'insight' loosely - referring to it as a critical finding from the data.
Brent uses a definition by psychologist Gary Klein, to describe what is an insight:
An insight is an "unexpected shift in the way we understand things" - Gary Klein
Another evocative description by the same thinker:
"Intuition is the use of patterns we've already learned, whereas insight is the discovery of new patterns"
As I was reading these parts, I realised - hang on, he is essentially saying that a message, to be called an 'insight', has to convey something new i.e. it must be surprising or unexpected!
I have been teaching about the importance of surprise and norm-variance (as a part of messaging) for years now. But now, I could connect the two and clear say: Message+Surprise = Insight.
But that's not all - there may be surprising findings which the audience doesn't care about - because they don't move the needle for them. Which is where the next factor comes in: value.
Some findings may be insights (i.e. new), but not really valuable for the audience. What makes an insight valuable is its ability to move the needle on a key business metric that the audience cares about.
Based on this clarity, I created a 2x2 chart (what else) to explain the different types of messages:
A few months back, someone asked me in an interaction: When should we not use a Story?
My response to the question was not very convincing. I said something on the lines of: "If you refer to a 'human story', then you would use it only if the context, audience and objective warrants it. But if you talk about storytelling as a skill, you need to be using it all the time."
After reading Brent's book, I have a better answer now (although it's for the related question: When should we not use a 'Data Story'?)
Let's face it - putting in the time and effort to craft a clear, comprehensive, visual data story is a lot of work. You should do it only if the stakes are worth it.
In the book, Brent defines those stakes using a (here we go again) 2x2 chart. And he calls it 'The Story Zone'
The Story Zone by Brent Dykes (Effective Data Storytelling: How To Drive Change With Data, Narrative, And Visuals, Wiley 2020)
Not every insight is worth creating a data story for. You don't need to create an elaborate story, if your insight is:
Data stories are most essential for insights which are both high-value and hard to understand/believe for the audience - which make up the 'Story Zone'!
This insight itself was worth the price of the book for me.
Other striking points about the book
1. Great collection of stories about data stories
Apart from several such foundational insights, Brent also includes several real-life data-stories in his book (many of them which are famous ones from history). For instance, one chapter narrates contrasting stories that illustrate the power of data storytelling:
In addition, Brent also includes his own personal stories of successes and failures when presenting data stories.
2. Superb compilation of quotes on storytelling
Despite being steeped in the subject, I came across several quotes on storytelling which I hadn't encountered before.
"Data! Data! Data!", he cried impatiently, "I can't make bricks without clay" - Sherlock Holmes by Sir Arthur Conan Doyle
This quote especially resonated given this.
Or this one:
"A wealth of information creates a poverty of attention" - Herbert A Simon
"Address the eye without fatiguing the mind" - Alexander von Humboldt (German cartographer)
In short: Brent Dykes has not written a book on data storytelling.
He has written THE book on data storytelling.
What could be better? It is a dense read at times. And the multiplicity of frameworks does make it difficult for readers to place the concept they are reading in the overall context.
But if you are interested in learning this craft, this is one book that you must read and re-read.
PS: Incidentally, this book was also a gift! My colleague, Sanket, wanted to give me something as a gift and I suggested this book, given that it'd been on my to-read list for a long time.
It's amazing how much book-gifts can influence your life :)
I thought I'll mix up my reading diet a bit with some easy-reading non-fiction and picked up this book on a whim. I'm so glad I did.
Tamarind City is an ode to the city of Chennai. Now, despite being a Tamilian, my connection with the city is tenuous.
My wife has a lot of relatives there and I have great memories of visiting them over the years. Another fond memory is attending one of the famous December Music Season 'kutcheris' with a music-loving relative...and once taking a charming walking tour by Storytrails.
But despite all that, I hardly knew the city (as much as Pune, Mumbai or Delhi).
This book changed that - and made me want to visit Chennai. It made me want to stroll through the beach at Marina, walk the lanes in Triplicane and Mylapore, explore the historic Fort St. George... and most importantly, indulge in some authentic sambhar with kara dosai and filter coffee at Saravana Bhavan. (The Udupi restaurants - God bless them - have absolutely ruined sambhar).
'Tamarind City' made me homesick for Chennai.
The book effortlessly straddles and celebrates Chennai's rich colonial past (of which, shockingly little is known) and its tech-meets-tradition present.
It also does a great job of giving Chennai it's due position as the cradle of modern India's institutions (which the present-day city is surprisingly shy to celebrate). Here's the author:
... almost every modern institution in the country—be it education, engineering, medicine, the army, or judiciary—has its roots in Fort St George. Modern India originated in Fort St George.
Many clerks and soldiers and administrators who came to serve in Madras as non-entities were catapulted to unbelievably high positions—high enough not only to decide the destiny of India but also of Britain. During the eighteenth and nineteenth centuries, a number of illustrious Britons, including prime ministers, commanders-in-chief, governors-general, members of Parliament and bureaucrats had one thing in common—the Madras connection.
Present-day Madras, that is Chennai, is somehow shy of celebrating this connection. It would make far more sense to have the entire Fort vacated so that it could be restored and turned into a museum that would welcome visitors with the signboard, ‘Modern India began here.’
Apart from the history, the book has some lovely interviews with leading present day Chennai celebrities - including the poet Meena Kandasamy, the Carnatic vocalist TM Krishna and the IVF specialist, Dr. Kamala Selvaraj - as well as ordinary citizens.
Biswanath has a languid, easy writing style. He often does nothing else but get out of the way between the subject and the reader... and lets the story gently unfold in front of you.
A highly readable, engaging book on one of India's most fascinating major cities.
This is one of the podcasts I stumbled onto by happy accident.
In fact I was doubtful of giving it a listen, because I had never heard of the host, Jason Feifer, who, to be honest, doesn't have a stellar resume.
But oh my gosh, is he a fabulously engaging storyteller...
His podcast is structured as a 'narrative' show - the apex format of non-fiction podcasting (think 'Revisionist History', 'Freakonomics', 'Against the Rules'). What makes it interesting is the kind of topics he picks up for his deep-dives.
For instance, consider this utterly fascinating episode titled "When exactly were the good old days?" - in which he digs deep into history to find out - was there ever a period of time when people did not crib about some period in the past being the 'good old days'?
Turns out, no! Present day Americans looked up to the ... 1970s/50s maybe? But those dudes looked back fondly to the 1920s... who looked back further still, who... you get the picture.
The episode culminates its history joy ride with a fascinating revelation from Sumer, the cradle of written civilisation.
When asked on Twitter for newsletters that he subscribed to, my friend Anustup Nayak mentioned this one (thank you!) and another called Womaning in India by Mahima Vashisht.
I looked it up and came across this write-up in Mahima's newsletter intro-page on why she chose to start it:
"...half the world's population, men, have virtually no idea what the everyday life of the other half is like. Even the most woke of men out there would be shocked if they had to walk a mile in women's shoes.
People - men and women - are mostly decent and good. And, stories are power. Together, these two things lead me to believe that sharing stories about lives that are different from ours can make the world a better place for everyone.
This thought inspired me to start talking to women about their stories."
I was intrigued and signed up.
In her newsletter, Mahima picks up a theme relating to gender every week (e.g. how doctors treat women or, how office parties can become smoke and booze-filled bro-fests) and then - here's the important part - shares stories.
She shares stories after stories after stories (as she says in her intro, "Where do I even begin?"). And then she rounds it up with some credible research on the topic.
A lot of it is utterly eye-opening. For instance this one:
Once upon a time, I remembered the birthday of a kind colleague and reminded the boss. The boss thanked me and said, “Why don’t we order a cake for him?” Of course, when he said that, he really meant “Why don’t you order a cake for him?”
It seemed an innocuous enough request (aka order) and I did it. After the party, Boss turned to me and said, “Great cake, Mahima. Good job! From now on, you are in-charge of ordering the cakes for all team birthdays.”
It made me sit back and realise - even when I was at work, 'cutting the cake' and putting it on plates for distribution would usually be done by women.
Prakash Iyer talks about 'Vuja de' - spotting the unusual in things that we take for granted. Mahima is brilliant at spotting such vuja-de moments and writing about them.
We need more voices like hers.
Plus, extra points for the kooky Bollywood memes she uses. Hilarious.
Ruchir Sharma, Global Head of Emerging Markets at Morgan Stanley is a superb writer (apart from being a sought-after investor). He makes an interesting case in this piece - of how the 'dirty old economy' might still be needed, even for the new cleaner one to be built.
This small extract summarises the issue well:
Building green economies will consume more oil in the transition period, but producers aren’t responding the same way because political and regulatory resistance has darkened the future of fossil fuels.
Even as oil prices rise, investment by the big hydrocarbon companies and countries continues to fall, as it has for nearly a decade. Instead, oil powers are reinventing themselves as clean energy powers. One broker recently wrote that of his firm’s 400 institutional clients, only one is still willing to invest in oil and gas.
...distinction between 'science as a philosophy' and 'science as an institution.' The former grows out of the Enlightenment... the primacy of rational and objective reasoning. The latter, like all human institutions, is erratic, prone to falling well short of its stated principles. The Covid pandemic has thrown into sharp relief the disconnect between science as a philosophy and science as an institution.
The article makes a provocative claim:
Mr. Ridley notes that the question of Covid’s origin has “mostly been tackled by people outside the mainstream scientific establishment.” People inside not only have been “disappointingly incurious” but have tried to shut down the inquiry “to protect the reputation of science as an institution.” The most obvious reason for this resistance: If Covid leaked from a lab, and especially if it developed there, “science finds itself in the dock.”
(h/t - Sajith Pai on Twitter)
I loved this collection of terms from the world of tech development. For instance, take "Bikeshedding":
The term comes from a story in which a group of engineers, architects, and scientists are hired to build a nuclear power plant, but get stuck deciding where and how to build the staff bike shed.
Where will the bikes go? How many bikes should it be able to hold? What color should the bike shed be painted? All of this attention on the employee bike shed results in a loss of funding and neither the bike shed nor the power plant being built. This is also known as “The Law of Triviality”, that is, that people will give disproportionate weight to trivial matters. Or…bikeshedding for short!
Other cool terms: rubber ducking, dog fooding, bus factors, and my favourite, yak shaving!
A lot of us indulge in these practices in our workplace - now we have a name for them.
A close friend, Sean Mathew, introduced me to this mad wonder called Shraddha Jain. Also known as RJ Shraddha (and famous for her role as the hostel warden Vasu, in the Amazon Prime series, Pushpavalli), this one is a force of nature.
In this hilarious video, Shraddha invokes your sympathy for the poor online school teacher. She parodies how kids behave in class and makes you marvel at the unending patience which a teacher needs to have to handle these little tykes!
(h/t Prahlad Viswanathan)
I can't get enough of Brit comedy! 'Would I Lie to you' is a crazy show pitting two teams captained by David Mitchell (famous for his rants) and Lee Mack (known for his razor-sharp quick wit) against each other.
The teams (also featuring special guests for each episode) try to tell lies in the form of stories, and get points for correctly guessing whether the opposing team member is telling a truth or a lie. The whole process of storytelling and cross-questioning is hilarious and super fun.
In this video, they play a popular segment of the show, where each of the three members of one team gives a story about a stranger - and the other team has to identify who among them is telling the truth. Laugh-out-loud stuff.
That's it folks: my recommended reads, listens and views for the month.