Telling Stories

Did you know our brains process visuals up to 60 000 times faster than text alone?

Cave paintings are the earliest known human data visualisations with some dating back 40 000 years. These paintings are widely thought to depict maps of the stars, historic events, directions to food and water resources and many other topics of importance and interest to their creators (and to us). 

Today, we live in a world where access to increased computational power and storage means that the volumes of data we can maintain and interrogate have literally gone through the roof. It is staggering to think that 6% of all information comes from books and printed material while 94% is digital. The amount of data generated increases by about 2.5 quintillion bytes - and that’s a very big number - every day. 

Telling stories In addition to statistical and analytical skills and the curiosity to interrogate data through novel means, data scientists need to communicate their findings to the outside world by telling complex stories in intuitive and accessible ways. A critical part of this challenge is to familiarise ourselves with our clients’ specific business issues, a recurring theme in this post. This allows us to contextualise our findings and present relevant recommendations in a frank and compelling manner. Our stories need to be easy to understand, impactful and scalable.

Now, more than ever before, clever data visualisation techniques have become critical to condensing the multifaceted patterns within data into pragmatic, intuitive stories. Data visualisation enables decision makers to grasp difficult concepts, identify new patterns and most importantly, take decisions.

Some of my favourite recent examples include the Washington Post’ssynopsis of the 2016 US election exit poll, their beautiful insight into thethe persistence of wartime inequality in modern Germany, and the New York Times’ historical exploration of Usain Bolt’s Olympic performances.

Running at Bolt speed Let’s talk about the Usain Bolt example for a moment. How does Usain Bolt’s 100m dash relate to decisions in business? At a high level, perhaps the most important outcome of effective data visualisation is the enhanced speed of decision making. In order to stay competitive, executives in pressured and ever-changing environments need to make the right decisions, fast. To do so, they require access to insights that are tailored to their particular business challenges, and often in real time. What this amounts to is replacing impenetrable tables of rows and columns with a much more useful distillation of just the right amount of data that is crafted into dynamic, interactive visualisations and tells only the story of interest.

It is also important, particularly for data analysts, to be able to drill into and expand upon high-level insights. This usually requires a higher degree of interfacing with a robust back-end platform where the data is wrangled (i.e. coaxed into order), models are fitted and forecasts are projected. The rewards for the higher interfacing complexities are answers to the whys, hows and what ifs. 

Data visualisations don’t always need to be interactive though. Static infographics can also provide powerful insights. Fundamental questions such as: “How do your customers interact with your business?” can be answered with approaches like customer journey maps which provide insight into your customers’ behaviours, goals, obstacles and even emotions. In a retail environment this can translate to better product placement while in banking this insight may inform the reasons behind an increase in loan defaults.

The right amount of simplicity Clever data visualisation decisions can also play important roles in communicating insights to people from diverse backgrounds where one size does not fit all. Instead of slideshows full of spreadsheets and small print, a well thought out visualisation can quickly unclutter your slide while conveying your points clearly.

Simplifying insights for the sake of it may not always be the smartest approach however. Barack Obama once said “Issues are never simple. One thing I'm proud of is that very rarely will you hear me simplify the issues.” While the former President wasn’t talking about data visualisations, his caution rings true. Careless data visualisation can smooth out fine scale fluctuations and interpolate data which don’t exist, leading to us missing important subtleties.

In order to provide meaningful insights into key drivers, data visualisations must always be considered very carefully and in the context of your unique business challenge.