Stephen Few the Godfather of Visual Data

The Godfather of Visual Data

Stephen Few

Is the Godfather of visual data. In his seminar at Copenhagen Business School (CBS) on monday 8th of April, arranged by Inspari, Stephen Few talked about making sense of data and using this sense to gain knowledge and in turn use this knowledge as wisdom.
Wisdom is by many scholars seen as the ultimate state of data transformation though they each have their take on what “The Wisdom hierarchy” should look like (Cleverland (1982), Ackoff (1988), Bellinger (1997), Wu (2000) and Rowley (2007)). This transformation is the essens of Business Intelligence, and how companies and organizations can evolve from gutfeeling-made-decisions to fact-and-hard-evidence-made-decisions. The following blog is a summary of “The Godfather of Visual Data”, from the CBS Seminar.The Wisdom Hierarchy also known as DIKW

The problem with data

Through evolution data has evolved from wall paintings to written paper to electronic data which have exploded the amount of data. Stephen Few showed that if all the data was printed it would render the world treeless 12 times, and if all the data was stacked into books it would be 13 stacks from earth to the sun. In just six years amount of data has gone up 10.000%. This trend does not seem to turn so in the future there will be even more data. This really put things into perspective – but also point towards the problem. How do we make sense of that amount of data so that our knowledge can be used to make wise decisions.

Mr. Few points out that we can no longer be experts at anything it might no longer make sense the way we are educated with focus on memorization and regulation of information – that because data is accessible at all times, anytime and anywhere, and we need to be able to handle it differently – we need to make sense of it differently. To summarize he showed us the following video, and concludes that education needs to lean pupils how to

  • Ask good questions -> to get good answers
  • Access information -> from appropriate sources
  • Analyze and authenticate -> to determine fact from opinion
  • Apply it to -> real world problems
  • Assess the outcome and process

This is in short called “Information Fluency” and you read more about it on “21st Century Fluency Project


Visual Thinking

According to Mr. Few does data not become information until we make sense of it. To do this we need to learn the right skills argumented by the right tools. Two of the right skills are

– statistical thinking and
– visual thinking

To understand visual thinking Mr. Few presented another short video about a woman called Temple Grandin. Temple Grandin is a female autism that works with cattle. Because of her disability she has developed her visual thinking much more than people without her disability. This has in fact lead her to be able to visualize and see patterns and make sense of the information that she is presented with, that others could not.


From here Mr. Few went on and divided sense making into three modes.

Mr. Few argues that there are three modes of making sense:

  • Verbal: A B C
  • Numerical: 1 2 3
  • Visual: by image

In his presentation he underlines the need for recognizing the visual mode of sense making, as a vital skill that should be taught to everyone today. Furthermore, Mr. Few highlights the need of a need to develop greater balance between the three reasonings to better understand and filter the data that we are presented with. This will help us understand data and filter it. By being better at visual sense making users and viewers of data will have a easier time communication and understanding large and complex amounts of data.


There is no better way to convince an audience about your arguments if you can exemplify them. To introduce the notion of tools Stephen Few talked about the difference between human and computer thinking, wanting to show that humans should do what humans are good at, and computers what they were built for. The example given was that you in your mind needed to add lines of math written in text. Now I can’t remember the exact numbers but Stephen Few walked us through it, and in the end asked what people had added up. More than two-thirds of the audience got a figure that was way of the actual result. He used this example to show that computers are great at addning numbers. Afterwards he showed a captcha

Captcha – Only humans reads this

image and said that no computer would yet able to recognize the text, as it is patterns in the image that writes the text.
This underlines his fact that people show do what people are good at.

The example led to the next phase of the seminar in which Stephen Few presented different ways of presenting data. To start he showed a table with four rows and ni coloumns. The table had a lot of figures. Next he showed the exact same data but presented in a graph. The point being that it is easier for people to recognize a graph and with that remember large amounts of data, as we can remember patterns – instead of trying to remember 36 different figures. Also the time it took to analyse the two different perspectives of the same data is a factor, as the later gave great insights into the numbers in a few seconds whereas the first need much more time to get the same value.

In the next part of his seminar Stephen Few show various examples on how not present data and how to present data. The essens being that you should pay attention to what is important for the viewer to understand. Is it detailed numbers, or a understanding of how economy has developed over the last few months? Making sense of data presenters should make sure to understand the two essential activities of presenting data properly in reports or interactive dashboards:

  • Multiple views: Data should be displayed with the audience in perspective – it has to make sense for their purpose
    • Multiple vies of chang through time
    • You want to see the data in different perspectives
    • No ‘one’ view will the the full story
  • Comparisons: Visualizations should use the same foundation – if they are to be comparable
    • Find ways to visualize data to help compare it
    • Look for rich ways to compare data in order to explore and make sense of the data

Data needs to tell rich stories in a very rapid way, to make people understand data in a way that make sense. Stephen Few’s Motto:

Eloquence through simplicity

Finding the simplest way to show complex data. This is in my opinion probably also one of the hardest skills to learn and continuously adapt to the audience. Stephen Few says that data visuliazation builds on understanding the brian and the way we think and have been taught to think. He exemplifies by showing the audience of to sets of circles. One looks as if they are sticking out of the canvas, and they other as there is a whole in the canvas. He asked us which was which whereafter he flickered to the next slide and asked the same question. This time it looked as if thethe circles had swapped place when in fact he had only turned the gradian in the circle up side down. With his example Stephen Few argues that from early time human has seen the light coming from above, which is why we percive an element as bulging in or out when it is darker in the bottom and lighter in the top. On this Stephen Few notes that gradiants should not be used in backgrounds – as they have no purpose and only serves as a confusing element of the presentation.

Visual perception works according to it own rules
– the light comes from above

Data analysis requires rapid interaction

The Godfather of visual data doesn’t want us to manipulate data, but the perspective in which we view data. To illustrate this he showed the following triangleData Analysis: Rapid interaction - Think, See, Modify - Interact

With this he wants us to look at the different aspects of data visualization. It is important to present the data in a way that makes it possible to interact with it. I dont remember the exact explanation that he provided here, but he refers to Ben Shneiderman’s Mantra:

“overview first, zoom and filter, then details-on-demand” (read it here: The eyes have it: A task by Data Type Taxonomy for Information Visualizations)

Ben Shneiderman is an expert in the field of data visualisation and is very cited in his field. While writing this article I also found “Shneiderman’s ‘Eight Golden Rules of Interface Design’, which I definitely think is worth a read too.

Attention to detail

When visualizing data it is important to focus attention to what is important and augument memory to make sure the audience remembers what you want them to remember. In the godfather of visual datas own words:

“Focus attention and augment memory”

Again examples are really good and Stephen Few, uses the following example to show how the mind of people are easily distracted from observing. Try and watch the following video, and count the number of times the the ball is passed from one player to another.


Did you pass the test? Not everybody does – infact people are often so focused on a single task that their mind do not ‘see’ other things. While watching the video at the seminar I remembered another pretty cool awarness test. Try and watch the follwing:


Memory is limited! Apperantly the mind can only keep 3 to 4 chunks of information simitalionously, which is why you need to focus your attention to what is important. According to Stephen Few the working memory (the brains RAM) pulls from longterm memory or imagination. External aid can augument memory and help people to remember larger chunks of information. Stephen Few divides this into three areas:

  • Quantity
    • Greater quantity through visual encoding
    • It easier to visualise a whole line representing data rather than the data it self. We are comparing data, so we do not nesecarily need the details of data
  • Dimension
    • More dimensions through multiple visual encodings and small multiples
      – Stephen Few shows a comparison of multiple graphs showing gender in populations over 10 decades
  • Perspective
    • More perspectives through coordinated views
      – Stephen Few illustrates this with a story of three blind men who feels an elephant. Because they feel the elephant in different places they have different perspecive of the elephant. When the men are talking about the what they felt – they each saye something different – this goes to show that different perspective gives different perceptions – which is the same with data.

Coordinated views expand what we can connect and compare and help us to remember and understand information better. Finding and understanding the stories in data is not enough we must also tell these stories. This is illustrated with another example with Hans Rosling of – expert in telling the story about numbers in order to make the world a better place.

[ted id=92]

The Godfather of Visual Data Stephen Few rounds up his seminar with a quote by Leonardo da Vinci, which I personally is a great fan of, and strive to practise in my work and presentations.

“Simplicity is the ultimate sophistication” – Leonardo da Vinci

How can Infomation be transformed into Knowledge and apply this to make better decisions?

The value of information depends on how it is used
– use it wisely

All in all it was a great seminar with the godfather of visual data and I learned a lot. I will definitely be buying his book and reading a lot more into his seminars and articles. I definitely agree that the way we illustrate and show data is important  and as pointed out by Stephen Few, the amount of data is only increasing which is why we need new and better ways to navigate and filter through information to find what is important to us.

Thanks to Stephen Few for a great seminar!

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply