Good journalists are excellent storytellers, and so some journalistic techniques can help you tell your data driven stories.

The inverted pyramid

In journalism, the “inverted pyramid” refers to a metaphor used to structure the content of news articles. According to the inverted pyramid, the most important and interesting information in an article should already be contained in the first few sentences. Details can follow later in the text, and additional background information should be mentioned last.

Source: Makeemlighter, public domain

Using the inverted pyramid to structure a text leads to articles with some interesting properties:

• in a world where attention and time of the audience are scarce goods, readers get the take away message upfront. Without reading the whole article, readers get the essence, and can decide to read the rest of the article if they are interested in the topic, or are triggered by a “hook”, which is sometimes also part of the lede of an article (a lede are the first few sentences of an article).
• when an article with the inverted pyramid structure needs to be shortened, it can be “cut from the bottom”. Because the least important information is placed at the end of the article, it can be removed without major consequences: the most important information is still there, and the article can still be read and understood without any problem.

So, the lede should contain the main message of the charts. A helpful technique to determine what this message should contain is another journalistic technique, the 5 W’s. The 5 W’s explain the who, what, when, where and why of the story, and the theory states that any good news story should answer all 5 W’s (the how of a story is sometimes also added).

So how is the inverted pyramid relevant in data visualisation and in data storytelling?

The first take away is to add a descriptive title (see also the Chart titles module) and a good chart description or subtitle to communicate the main message of a chart. Even readers spending limited attention on reading the chart, will still be able to understand the key message.

According to the inverted pyramid, the less important information should come at the end. So things like methodological notes and less important metadata should be added below the chart (if they need to be added at all).

Information-seeking mantra

The most useful application of the inverted pyramid to data visualisation, however, is in multi-panel and interactive visualisation: it provides a good framework for deciding how graphics should be arranged and how interactivity should be structured.

Looked at it this way, the inverted pyramid is similar to the “visual information-seeking” mantra, coined by Ben Shneiderman in 1996.

“Interestingly, the journalism triangle more or less corresponds to the mantra of interactive graphics: overview first, zoom and filter, and details on demand” - Barbara Tversky in Data-Driven Storytelling

The mantra was developed with interactive and exploratory data visualisation in mind, but it also applies to explanatory data visualisation.

According to the mantra, overview should be provided first. In a story driven by data, a visualisation of the longer term trend, or the overall distribution or relations in a data visualisation can be shown to sketch the context and introduce the topic and the data. All data points are shown. The overview can also consist of a single key metric, like a global average or total sum.

Then come zoom and filter. The view on the data is changed so that items of interest in the data come into focus and uninteresting items are filtered out. Outliers are highlighted, correlations are shown, smaller interesting patterns are annotated, the total sum is broken down into its components, ...

The opening of a series of graphics in a scrollytelling module in How Hong Kong’s National Security Law Is Changing Everything, by Bloomberg, is a clear example of an overview: it introduces the topic and the data set of 150 people, which is fully displayed.

After the overview, the story uses zoom: specific groups of people are highlighted. Source: How Hong Kong’s National Security Law Is Changing Everything, bloomberg.com