Write and Design your Data Story, Step-by-Step Playbook
Why is storytelling so important?
One of the oldest ways humans transmit information is by telling stories.
Science has shown that people read information, but they feel a good story, and feeling a good story makes it memorable.
Studies into how we learn show that we retain only 10% of what we read, 15% of what we hear and 75% of what we see. We also process images 60,000 times faster than text and tend to focus on visual information. What's more, we have the unique ability to remember images with at least 90% accuracy.
From there, we can write that smartly presenting data appeals to the heart as much as the mind.
Combining the persuading power of insights from data with narrative engagement makes data storytelling compelling.
Its role is to make the bridge between data analytics (generating insights and visually presenting them) and communication (the art of storytelling).
How to start?
The winning starting point is to focus your attention on the human elements of a dataset to pinpoint more clearly what is at stake (why should your audience care).
Once this first objective has been clearly defined, take the time to ask yourself the following question:
Is what I present going to enable my audience to make a decision, or am I going to enrich a discussion that will lead to a decision?
If the answer is no, you need to review your objective.
A successful presentation should always put your audience in a situation where they can make a decision (even if that decision is to validate your point of view).
What’s next?
Developing a solid and convincing data story is a step-by-step process:
1. Understand your audience
The first stage determines the rest of the process.
The general idea can be understood at a glance.
However, be careful: understanding your audience takes a little time.
But it is worth the effort.
The more you understand your audience, the better equipped you are to tell your story appropriately.
Let’s start by investigating the following classic 5 W’s questions.
Here is a list of questions you can use to prepare yourself:
-
Who are the stakeholders?
Primary stakeholders: Who is it about?
Secondary stakeholders: Who is affected?
Who benefits?
Who loses?
Your audience will often be someone other than the stakeholders but the decision-makers. Taking a moment to measure the impact of your insights on stakeholders will enable you to gauge the need to integrate the human element into the unfolding of your story.
Pay attention
Remember that the recipe for a good story lies in balancing the emotional and the rational.
Who will be making decisions based on my data?
Who are they?
What is their level of data literacy or familiarity with the topic?
As a data storyteller, your main job is to ensure the information is as easy to process as possible.
In this area, it's essential to consider your audience's level of data literacy.
If your audience is new to the subject and the type of data you're dealing with, you'll want your data to be simple, concise and close to what they do.
On the other hand, an audience made up of peers will be more interested in explanations of processing methods, calculations and assumptions. They will also expect you to address them using professional terms.
Finally, a management committee will expect a clear, structured, simple, and brief presentation of insights, where the answers to the challenges and the options to facilitate decision-making are clearly formulated.
Who might have differing perspectives or interests that should be considered?
-
Why is this data important to the audience?
Why does this story matter in the context of their goals or challenges?
Make sure your insight is relevant to your audience. Is it aligned with their current priorities or goals. If not, you will have to clarify why it is pertinent to their needs. It may seem obvious, but it is worth checking before you start preparing your story.
Another situation requiring special attention is when your insight creates discomfort or is against an established point of view.
In such situations, you must show diligence and an eye for detail.
Even if your audience doesn't always like all your ideas, they need confidence in the figures you communicate.
So, ask yourself, why should they trust your data and your conclusions?
Further, if your insights challenge the current situation, you must emphasize the benefits, address expected concerns, and align your insights with their goals or values to make it more acceptable.
Finally, avoid being defensive if your insights contain bad news or information that could be perceived as negative or critical. Instead, focus on the data that will help mitigate the problem and offer a perspective that allows you to move forward.ext goes here
-
What actions do I want the audience to take after hearing this story?
What data or metrics will be most relevant and impactful for them?
At this level, think twice. Make sure that your audience is fully aware of the metrics you are using. A fairly common trap, known as the ‘curse of knowledge’, lies in no longer being aware of what you know and what you consider to be common knowledge.What key insights or messages do I want to convey?
What does each of them mean for my audience?
What are the consequences?
What questions are the audience likely to ask?
-
When will the audience interact with this data (during a meeting, in a report, on a dashboard,…)?
How much available time do they have?
When should key insights be revealed for maximum impact in the presentation or report?
What is the timeline for their decision-making process?
-
Where will the audience access this information (e.g., in person, remotely, via a document, on a website)?
Where will the data be most effectively communicated (charts, tables, infographics,…)?
Where are potential gaps in the audience's understanding that need to be addressed?
2. Build your narrative flow
Objective: Make your point, feed a discussion and get the decision you desire
How to get there?
Let’s consider the following 2-step scheme:
Storyboarding
Select relevant information that matches your audience's expectations and allows you to make your point.
Then, define the order so the audience feels inspired to act.Writing your Story (special thanks to Brent Dykes)
One common way to structure your story is the following (based on the famous Freytag’s Pyramid):
-
At the beginning of the story, you set the scene.
This mainly involves providing the essential background elements to set the scene for the rest of the story. Be brief and keep only the elements that are absolutely necessary to understand what is at stake.
· What is the status quo?
· Why is it important?
On the other hand, don't neglect the human element. Your data story will always be based on data from people who are of interest to your audience - customers, employees, investors, partners and so on.
It's through them that you'll create the emotional connection.
-
What noteworthy observation or triggering event will radically upset the balance of power in the players' lives, thus revealing the potential problem or opportunity?
This observation will capture your audience's attention and encourage them to listen to the rest of the story. -
The protagonists must react to the triggering incident, and these reactions always lead to more complex situations or problems.
You present what you've discovered in your analysis, organising them to create a climax.
-
This is the climax of your data story.
You reveal the main conclusion of your analysis. It may be something new and unexpected that changes the understanding of something.
It can also be a specific reason your audience should care (the ‘so what’).
Your challenge is to give your audience a sense of how to solve the problem or to catch the opportunity. -
At this stage, you will explicitly outline the different options available to your audience and make a recommendation.
This will put your audience in a position to decide or discuss the measures to be taken.
One last thing ….. Kill your darlings!
The most common pitfall at this stage is overcomplicating the story by not resisting the temptation to pack your data story with every piece of information. So, be careful.
3. Lay out the information
Once your narrative flow has been drafted, it's time to select the visuals containing the data you'll use to support the story's thread.
Whether your story takes the shape of a dashboard or a presentation based on a deck of slides, the following rules will help you enormously:
Choose the right graph formats
For example:Line charts are more effective for highlighting trends over time.
Bar charts are recommended for comparisons.
Pie charts highlight the different parts that make up a whole.
Scatter diagrams identify relationships between variables.
Heat maps are ideal for showing the relationship between two variables in a raster format using colours.
Layout: hierarchy, clarity, minimalism and accessibility.
Hierarchy
More prominent elements attract attention first.
Consequently, and quite logically, you will:Vary the data size according to the importance of each piece of information.
Play with colour contrast to accentuate essential information.
In the case of a dashboard, indicate the paths you want users to take.
Anchor your images with text, arrows, annotations, and other visual cues to guide users.
And if your data follows a logical structure, you'll follow it :-)
Design: Minimalism is the key.
Your audience should be able to discern significant patterns and trends at a glance. The only rule to follow is to keep it simple. Cluttered visualisations blur the message.
Here are a few tips to ensure optimal reading quality:No unnecessary labels
No excessive grid lines
Colours: define a strategy before you start.
Here, too, the rule is simplicity. Fewer different colours, more shades and a series of rules for use in line with the hierarchy of the information and their relative importance.Typefaces: Choose one of the most legible typefaces.
The size and style of the chosen font should once again prioritise clarity, producing an aesthetically pleasing display free from unnecessary distractions.No complicated layouts that confuse or mislead.
This doesn't mean that your visuals have to be boring. There are many ways of using colour, layout and typography to combine aesthetic appeal with effective communication.
Are you in any doubt?
Our design team can give you a hand.
Accessibility
Good practice exists regarding colour blindness and compatibility with screen readers to ensure inclusive access to information and comply with ethical and legal standards. We'll return to this in a specialised article.Testing and feedback
Ultimately, only a test with a few audience members will enable you to validate whether your device works and gather some enriching feedback.
How to ensure a validation before you present
When the stakes are high, validating what you are ready to present is advisable.
Two challenges are on the agenda:Pre-check with your audience
Having conversations with your audience about the topic ahead of time can give you an idea of their familiarity with it and collect possible scepticism about your point.Senior leadership buy-in
A preliminary discussion with someone with a high degree of influence over the topic will help secure buy-in from senior leadership before the presentation.
4. Share or present your insights
If you can access a Power BI Service, you can share reports or dashboards with other members of the same workspace.
If not, integrating Power BI and PowerPoint offers you a solution.
Here's how it works:
From Power BI, you can integrate visuals, dashboards and reports directly into PowerPoint slides.
Exporting retains the interactive elements, allowing you to explore the data and obtain information without Power BI.
You can rearrange, resize, and format visuals in PowerPoint for visually appealing slides.
Power BI visuals are stored in the presentation file, eliminating the need for an Internet connection or a Power BI service.
Final note
Whatever your job or role in the organisation, you are surrounded by data containing hidden treasures that are not difficult to exploit.
And you've realised by now: the work doesn't stop there.
Selling the findings of your own analyses is not something you can improvise.
And while the job is not complicated, it does require special attention.
I hope you have found this article useful.
Don't hesitate to contact me if you'd like to discuss it.