• Sean Keenan

Breaking down the different type of information visuals

There’s a countless amount of research done that shows that humans are extremely visual creatures. We’re far more likely to remember and understand ideas and concepts presented in a visually stimulating way versus concepts we simply read or are told about. That’s part of the reason why when presenting data to groups, there’s usually a strong visual accompanying it.

A difficulty that arises from that though is the sheer magnitude of different avenues we have to present these visuals. To try and address that issue, author Scott Berinato uses this section in his book Good Charts to break down what he sees as the two core questions a visual must answer. These questions are whether or not the information presented is conceptual or data driven and is the visual trying to explore something, or declare something. These questions can then be put in a 2x2 matrix, which gives us four different types of information visuals.


The first type Berinato discussed is the idea illustration, or the conceptual-declarative model. In this illustration type, the main goal is to turn seemingly complex sets of data into easy to understand visuals. It takes concepts easily understandable to most people, such as metaphors and conventions like hierarchy and shapes. He calls this the “consultants corner” of the matrix, as a consultant is often tasked with trying to present this complex information as simply as possible to their client.

A good example of this type of chart is this one showing the decrease in air pollution of major cities.

Obviously there’s more to it than the information shown in this, but at a glance it’s very easy to grasp the overall concept. The large bars point downwards to help show that pollution is decreasing, as well as having the exact percentages at the bottom for a more exact measurement.


At first glance, concept-explorative visuals can seem odd, as it’s often not using “real” data and rather simply exploring an idea using a visual. In reality though, these types of visuals are fairly common and useful for brainstorming and exploration. They’re often rough sketches done on things such as whiteboards and notepads in group settings as a way to decipher through a bunch of complex, muddled concepts.

A great example of this that many people probably know about is the classic whiteboard strategy meeting used in a basketball game.

A typical coach's whiteboard found here

Often done during a timeout in the middle of a game, this is used as a way to quickly communicate to the team what play the coach wants to run. The X’s and O’x representing the players are easy to understand, with the lines helping show what each player has to do.

A typical sight you'll see at a basketball game. Coaches use this whiteboard to go over a multitude of different plays in a quick fashion


This type of information visual is easily the most complex and can be difficult to grasp. At its core, it’s trying to answer two specific questions, as said by Scott Berinato himself in Good Charts

1. Is what I suspect actually true? 2. What are some other ways of looking at this idea

It usually involves big, complex data sets and is the area that professional data scientists and others of that ilk thrive. The exploration aspect of it also opens itself up to interactivity, potentially using multiple sets of data as well as data that’s dynamic and frequently changing. When presenting this data, it opens itself up for unique visual presentations compared to the average bar or pie chart.

A good example of this is this chart which examines the correlation, or lack thereof, of immigration and crime rate

A sample of some of the many charts used for this data analysis

There’s a multitude of different complex data sets being presented here, all with a unique type of line chart to show them. In the full article, there’s also scatter plots showing even more information.


In comparison to the previously discussed data-driven-exploratory, this type of information visual is fairly simple and probably what first comes to mind when thinking of data visualization. This is often just taking basic data from an excel sheet and spitting it out into something such as a pie chart or bar graph. There’s few variables in these data sets and the message of the chart is simple to understand.

A good example I found is this bar chart showing the average inauguration age of Presidents

Quick and to the point

The main key of this graph is that there’s not much going on. It answers a simple question and presents the basic facts behind it.


There’s a seemingly endless amount of tools nowadays for presenting your data. Due to this, it’s key that if you’re ever in a scenario where you need to present your data, you understand the different types of approaches available and make sure you choose the one that best suits your data.

  • LinkedIn

© 2020 by Sean Keenan.