As an analyst, if you want action to be taken on your analyses, you will often need to present data and associated insights to stakeholders. It is therefore essential to put insights and actions into the right narrative or presentation form. This will make the data easily understandable to everyone.

In the presentation Stories beat Statistics from Brent Dykes during AB Testing Summit he explains why stories work better than just the numbers and how best to do so. Brent opens his presentation with two quotes:

“The ability to take data - to be able to understand it, to process it, to extract value from it, to visualize it, to communicate it - that's going to be hugely important skill in the next decades.” - Hal Varian

The quote is already 12 years old and the future that Hal Varian, Chief Economist at Google, is talking about is happening now.

“Storytelling is the most powerful way to put ideas into the world today.” - Robert Mckee

Why tell a story?

People love good stories. They are a big part of what makes us human. We are social creatures and have always used stories as a fundamental way to connect with each other. According to Brent, there are two reasons why it's better to use a story rather than just the data and insights:

Stories are better remembered

In a communication class taught by Chip Heath, Communication professor at Stanford, he instructed his students to use certain data points to make a pitch and pitch it to a fellow student. Ten minutes into the pitch, Chip asked which of the students could remember the data points that were pitched. What was found was that students who had incorporated a story into their pitch were more likely to remember the information (63%) than students who had not (5%).

Stories can seduce better

In a study at Carnegie Mellon University, students were given five one-dollar bills and a brochure about a charity, Save the Children. The brochure highlighted problems the children faced in their daily lives such as famine, war and various diseases. What the students did not know was that two variations of the brochure had been handed out. The first variation contained different statistics that explained the problems, the second variation told about Rokia, a seven-year-old Malian girl and the problems she and her family faced.

Finally, students were asked how many dollar bills the students wanted to donate to charity. The students who were presented with the story donated many more dollar bills ($2.38) compared to the other variation ($1.14) which contained many statistics.

How do you tell a good story?

Narrative

To determine the structure of a story, Brent uses a modified version of the Storytelling Arc from Gustav Freytag. The story gives structure to the message you want to convey, first of all you need to know who you are presenting to because a story must be tailored to the listeners.

The Storytelling Arc begins by introducing a story, what was the analysis question and what did you analyze? This is also the moment you place a hook, something interesting or striking that you build on later. Then you reveal one or several insights that lead to the main insight, the ‘Aha moment,’ that concludes your story. After that revelation, it's time to present the solution and next steps.

Visuals

There are five steps for creating a good data visualization:

• Identify the right data: in this step, you take another critical look at the data you have and ask yourself if the metrics you have best visualize the message.
• Choose the right visualization: Use a chart that best visualizes the data and your message.
• Match the visuals to the message: tailor visualizations to the audience's need for comparison.
• Remove unnecessary noise: Not all data is relevant. Remove unnecessary data points that do not add value to your equation.
• Focus attention on the important points: Highlight key insights with color, additional symbols or additional explanations. Where an insight is also explained in multiple steps and graphs.

Data

Data and the insights that flow from it are the fundamentals of any story. There are three points that ensure you are using the right data:

• Accurate: always check data quality and use unsampled data.
• Valuable: know what the important KPIs are before you start analyzing.
• Actionable: the advice and next steps must be feasible.

At the intersections of the narrative, visuals and data bubbles are 3 action points that lead to the change you want to bring about:

• Explain: add a story to the data to provide additional explanation.
• Enlighten: highlights interesting data insights with the visualizations.
• Engage: ensures that the viewer is captivated by what is being told.

Brent ends his presentation with a Hopi India proverb:

“Those who tell the stories rule the world”

We, as analysts and those responsible for digital data, can tell stories about data that can change the world (or your business) if you use the right narrative, visuals and data.

Want to see the complete presentation and related stories?
View Brent Dykes' presentation here, all other presentations from the AB Testing Summit you can here Finding.