Easy scales

Three data communication lessons from the book Making Numbers Count

1,546.

If I tell you that you slept for a total of 1,546 last year, would this number mean anything to you?

I received this exact figure in my Oura app. Oura is a smart ring I wear that tracks my fitness and health levels, especially sleep. Last month, they sent me a data story that was called the anniversary report. The 1,546 figure was at the forefront of it. And it was confusing.

First of all, it’s hard to grasp. Is it good or bad? How should I interpret it? Second, when dividing it by the number of days there are in a year, I got a surprising result: an average of 4 hours of sleep per night. I take my sleep very seriously, so I knew that something was off. After doing some digging, I realised this number was for half of the year only. You see, I purchased the ring in May 2022, so the anniversary report included only the second half of the year.

I was reassured that my actual average number of hours slept per night was slightly above 8. But it also got me wondering: why not communicate this small figure, instead of the large and abstract 1,546?

Improving these types of data communications is the premise of the book Making Numbers Count by Chip Heath and Karla Starr (shoutout to Neil for the recommendation!). It’s a good book. I enjoyed reading it. After a while though, the lessons and examples become a little redundant. So to spare you time, I summarised my favourite techniques from the book in this edition of The Plot.

Divide

Focusing on one concrete item instead of the abstract total makes the numbers much easier to grasp. While no one really knows (or cares about) how many hours they sleep per year, most of us do take an interest in the length of our sleep per night. This could also apply to many other units, human or not: one citizen, one employee, one deal, or one game. Here is an example from the book:

🥱 Throughout the first 18 years of his career in the NBA, LeBron James scored over 35,000 points.

🎯 Throughout the first 18 years of his career in the NBA, LeBron James scored an average of over 27 points per game.

35,000 is way too large for a human brain to really visualise, however, anyone who watches basketball knows that consistently scoring 27 points is mighty impressive.

This technique can be especially effective with more frightening data:

🥱 There are about 400 million civilian-owned firearms in the United States.

🎯 There are about 330 million citizens in the United States, and more than 400 million firearms, or enough for every man, woman, and child to own 1, and still have around 70 million leftover.

You’re probably already familiar with a slightly different version of this approach—using ratios instead of units of one. Florence Nightingale is considered to be one of the first people to have employed it in her data communications. Here’s an example:

🥱 In the first 7 months of the Crimean War, 7,857 troops died out of 13,905.

🎯 Nightingale’s translation: We had 600 deaths per 1,000 troops.

Simple and impactful, isn’t it?

Compare

Defining something new in comparison to something the audience already knows can get your message across much faster. Turns out, we’ve been doing this for centuries. There are historical examples of us comparing certain measurements to the human body for easy scaling. And it seems like the word mile simply means a thousand steps.

Here’s an example from the book to illustrate this concept:

🥱 The geographic area of Pakistan is about the size of five Oklahomas.

🎯 The geographic area of Pakistan is about twice the size of California.

And one more:

🥱 The amount of meat recommended as part of a healthy meal is 3 to 4 ounces.

🎯 The amount of meat recommended as part of a healthy meal is 3 to 4 ounces, which looks about the same size as a deck of cards.

Comparisons like these help us visualise the numbers. As someone who grew up in Europe, I don’t know how much 3 ounces represent. I do know however what a deck of cards looks like. As you’ve probably gathered, the trick with this technique is to choose a comparable that is easy to grasp. Five Oklahomas is a few too many. Similarly, if you compare your company’s revenue to the number of bags of chips you can buy with it, it probably won’t be very helpful. So choose your analogies wisely.

Now that you know that, here is one last—and super cute—example for this section.

🥱 Hummingbirds weigh about 3 grams and consume between 3 and 7 calories a day, making their metabolism nearly 50 times faster than humans.

🎯 A hummingbird’s metabolism is so fast that, if it were the size of an average adult male, it would need to consume slightly more than a Coke every waking minute—67 cokes an hour, for 16 hours a day.

Now you really feel it, don’t you?

Personalise

The more personal a data point is, the higher the chances that you’ll remember it. You may forget an anecdote about a stranger but will recall a piece of gossip about your cousin. Well, turns out you can use this technique in your professional data communications, too. Let’s look at a few examples from the book.

🥱 There’s a 20% chance of experiencing a mental illness in a given year, and a 50% chance of being diagnosed with a mental illness in your lifetime.

🎯 To a group sitting at a conference table, say: for every 5 people, 1 of you will be diagnosed with a mental illness this year. At some point in your lifetime, either you or the person across from you will be diagnosed with a mental illness.

And now you’ve got chills, don’t you? Here’s one more example I love.

🥱 Average earnings in Kenya are about $7,000 per year, compared with $68,000 in the United States. Kenyans spend about 50% on their income on food.

🎯 If you spent the same portion of your weekly income on food as Kenyans do, 7 days of eating would cost you $650 for dishes like cornmeal porridge and potato pea mash. How easily could you pay your other bills if food sucked up that much of your resources?

I don’t know about you, but I feel like I’ve just travelled into the life of a Kenyan family for a few seconds. That’s how powerful your wording can be.

I know it’s not always easy to make comparisons and adjust scales in day-to-day data communications. But I do hope these examples will inspire you to think twice about it.

Thanks for reading The Plot. 💤

See you next week,

—Evelina

P.S. The figures and ratios quoted in this newsletter were taken from the book Making Numbers Count and may be out of date.

---

Weekly gem 🤩

Here in Paris, we’re still buzzing from the fabulous dataviz exhibition curated by Marthe Viallet for INSEP. I shared some of the showcased pieces yesterday, but you should follow Marthe for a more immersive online experience!

---