- We will start by looking at what makes data stories hard.
- Then, we will move onto building pyramids
- Ordering the narrative
- And propelling the story forward.
Classic story telling is about entertainment. Making people laugh, cry, and maybe scream.
Most commercial story telling are sales pitches.
Or, a really big, memorable idea.
Like purple cow.
But telling data stories is hard. We've all seen a bad data story.
Slide after slide of boring data.
The client's bored.
And, the presenter is getting more and more anxious.
At McKinsey they have a special term for such a data story. They call it the Anxious Parade of Knowledge, where the presenter is sharing data in the hope the client gets something from it.
We don't want this.
We want this.
To understand how to get there, we need to dig a bit deeper into what makes data story telling so challenging.
The first hard bit is that facts need to be found in the data. We can't just create them.
I'll link to some webinars about that at the end.
The data can be confusing.
The facts can be ambiguous and weak. Is that just a smudge or a monster?
And our audience needs to use these facts to plan.
It follows from these last three points that the audience isn't just there to listen. They need to stress test any results. They need to make sure that the data is robust enough to rely upon when doing their planning.
And the document needs to be stand alone. There needs to be enough detail in it so that when it is forwarded to a colleague, the colleague can also stress test it.
And the document needs to be re-purposeable. It needs to be able to cut up and have facts and conclusions dropped into other reporting.
How do we solve all these problems?
Step 1: Build recommendation pyramids
Step 1 in creating a story with data is to form your data into pyramids.
What's a pyramid?
We have a key recommendation or conclusion at the top.
It's a summary, or a consequence of the blocks underneath it. Each block itself is also a summary or a consequence of the points underneath it. And ideally the blocks are ordered in a useful way
Using a pyramid structure helps people understand and remember - they understand because it shows how things fit together
They remember because
The information is chunked
Presenting information in a way that fit with the associative structure of our memories.
For data stories, the best type of pyramid is a recommendation pyramid, where all the data leads up to one or more recommendations.
Here is a recommendation pyramid for margarine. Please study it.
Elements of a pyramid
And this slide is just for reference. It spells out the details of a pyramid.
In particular, I draw your attention to all the different ways that we can order the facts in a level of the pyramid.
There are some options in how we create a pyramid. We can create an argument structured pyramid, like this one.
A grouping structure
Which is best? It depends on which works better for your project.
Step 2: Order the narrative
Once we have build our pyramid, we need to knock it down and use it to create a narrative, which is a linear path that goes somewhere. It's also known as a story arc.
A common way of analyzing stories is as the Hero's journey.
For example, with Star Wars, Luke is called to adventure.
He gets helped by Obi-wan.
He's almost crushed in the garbage compactor.
He becomes a bit of a Jedi
He blows up the Death Star.
Pixar's Story Spine
Pixar use something similar in their stories.
I used to work with a consultant who used this framework to present his results to clients. He presented a hero's journey, where he was the hero.
He'd start off by describing the brief. Once upon a time....
Every day he worked hard on the project. Until one day he discovered something.
Because of that he did a spectacular analysis and found something else.
Then he did something else even cleverer until finally he cracked the problem.
Ever since then the client's problem has been solved. The moral is that he's a genius, and should be given more work.
Narcissism aside, it's a terrible way to structure a story. The researcher should not be the hero.
To explain why, let's have a look again at the margarine pricing case study.
The researcher hero's journey as a presentation is shown shown at the bottom.
We present all the facts.
Then we present a summary of insights.
Then a recommendation.
This is a mouse trap
Earlier, we talked about how one of the unique aspects of telling data stories is that the audience needs to stress test the facts.
Consider the result shown in the top left.
If somebody presented this result to you on its own, you might think. So what, I don't really care, and not bother to stress test it. Like the mouse, you eat the cheese, not aware that this is a trap.
But, if you know that this fact is being used to support a recommendation to raise prices, it completely changes the way you look at it.
If your intuition is that raising price is a bad idea, you're going to attack this fact and make sure it's solid.
Lead with the recommendation
And this is how to structure the presentation so that the audience can truly stress test the data and buy into the recommendations, or, work out if the recommendations are wrong.
It starts with the recommendations, and then builds out the detail.
Do not launch the iLock
Sometimes, though, if you have bad news, leading with a recommendation may not be the winning strategy.
Some clients will eat the messenger.
Others will say ""give me the facts, I don't want your opinions".
Regardless of the client, we still need to work backwards from recommendations, as there's no other way to stress test or prioritize data.
Decision tree argument structure
But, we instead present it as a decision tree structure.
Rather than us as presenters telling the client what to do, we just lay out a framework for assessing the evidence, and then walk through the evidence.
Here's an example.
The client is still seeing the data and its implications, allowing them to challenge it. But, they aren't being directly confronted, so it is softer.
But, of course, I know there's a lot of you who still want the mouse trap.
The nice thing about the mouse trap is it creates suspense. It's a bit like a who dunnit, hoping that the audience will stay tuned, waiting with bated breath to find out how it all ends.
But, now that the surprise has been taken away, we do have a problem that we need to fix...
Propel the story forward
If we aren't using a mouse trap, how can we propel the story forward?
The This American Life team, the inventors of the modern podcast, say we need to propel the story forward every 60 seconds.
So, what are the special effects we need to use every 60 seconds?
Some of these are widely known.
- Have interesting facts,
- A good structure
- Data viz
But, we will focus on the less obvious ones.
Most of you are probably already aware of the need to layout pages in terms of pyramids or hierarchies, but it's something that few do, so I will emphasize it
The title on this page tells the story.
If you want to get a bit more detail, you just need to read the first sentence.
If you want to dig a bit deeper, the next sentence.
If you really want detail, you can look at the visualization.
And if you are desperate for more detail you can look at the numbers.
If you look carefully, you will see that this slide is itself a pyramid structure
The heading summarizes the entire slide
The less important the information, the smaller and more remote it is.
Shock and surprise
I'm a big fan of shock, surprise
And maybe a bit if whimsey.
As a presenter you want people to feel your own positive emotion.
In the days of Zoom this is easy. You sit on the edge of your chair, and lean forward. You put a big smile on your face. And then this positive emotion just flows through from your voice.
That's what I'm doing now. You also want to inject emotion into your story. Share your opinions and feelings.
Recall our presenter with his anxious parade of knowledge? My talented colleague Claudia has drawn the emotion into this illustration.
Earlier we talked about the hero's journey.
We create excitement by emphasizing the gap between where the hero is, the abyss, and, where they need to get.
The difference between where they are and where they want to be is the story gap.
The bigger the story gap, the more exciting your story.
There's a big gap between getting compacted and blowing up the Death Star.
And, I used this technique at the beginning of my presentation.
The more I described how hard things were, the wider the gap.
In my stories I like to use Motifs - recurring images with sybolism. Hopefully you are aware that this webinar is about pyramids.
If you saw last year's Oscar Winning film Jojo Rabbit, you'll forever remember the shoes and their meaning.
One of the most powerful techniques for propelling a story is to focus on something concrete and specific.
If I say "I walked into the street and guess what I saw? A cloud" That's a bad story.
If I say "I walked into the street and guess what I saw? A UFO" That's a better story.
Boy lying down
"I was lying on the grass. I looked up and saw a cloud.
It was exactly the same shape as the clouds that I used to be able to see, when I climbed up on my bed and looked through the 5 inch window from my jail cell."
Now you are hooked. Let's do it for market research.
Carmen Quadros is 36. She lives alone and exercises a lot. She's a doctor. She lives in Buffalo New York, in a condo.
Last week she went into the Wegmans on Amherst St and she bought 9 giant Hershey's. That's right. 9. More than 4 pounds of chocolate.
You get the idea. The audience will be keen to understand why she did it and this will connect them to the dryer quant data to follow.
If you want to learn more about how to do this, check out the podcasts Revisionist History and This American Life, which generally start with a really specific and concrete story to hook you in.
And last but not least, promote interactivity.
Get the audience involved in finding the story themselves. Let them choose their own adventure. That both keeps them awake and gives them an investment.
Interactive reports like this one are a great way to do that.
What happens if I launch a sour gum?
So now you know how to communicate the stories in your data.