The golden rule of data visualisation

The golden rule of data visualisation

March 27, 2019 DATAcated Challenge 0

The most often situation when we use data visualisation is to present some information to a manager to support taking some decision. With this article I will share probably the most crucial and yet most neglected rule when data visualisation is done in this context.

The rule is very short and simple. It requires some time and effort to adopt it but once you get accustomed to it, your visualisations will become dramatically better.

So here is the rule that (probably) will change your life:

MANAGERS ARE STUPID

I was lucky to learn this very early in my career when I was working as a financial analyst for a manufacturing company. One of my duties was to implement a report for sales. I worked really hard, combined data from several sources, summarized it in several pivot tables and charts on several sheets. Finally I put a dozen of buttons for slicing and filtering the data aiming to provide the end user (my manager) with the opportunity to look at every little detail of the sales results – products, clients, regions, seasonality, currency, unpaid invoices, margins and many more.

But my manager looked at this report for a minute and nodding his head simply said: “This is not useful for me.” And then added: “You have to understand that managers are stupid. They want to see one page and understand everything.”

Of course, he didn’t mean that managers are illiterate people with low IQ. The meaning of this phrase is that managers desperately need to receive information that is in such a form that even stupid man can understand it. Or a 5 year old kid if you prefer.

Why? Because managers are:

  1. Very busy with many issues
  2. Overwhelmed with information
  3. Not into details on your level

With that in mind here are three best practices that I follow when doing data visualisations (they are perfectly applicable for any other form of communicating information – report, e-mail, analysis, even phone conversation)

1. Minimize details

Managers are not interested in every little aspect. What they care are trends, relationships, big picture.

So if you want to present some sales data, it doesn’t have to be on a monthly basis. Think whether quarterly or even yearly basis could present the trends better.

Another example is usage of whole numbers. I still see numbers like this 10124956.78 presented to executive directors. Did you manage to read it at a glance? Do you agree that “10.1 mln.” is easier to comprehend?

Why is this so important? Managers don’t spend precious time for distinguishing between details and important and can concentrate on the issues on the spot.

And I can bet that no C level manager on this planet cares about the 78 cents after the decimal point.

2. Use colours sparingly, traditionally and consistently

Sparingly means you should use 2 or 3 colours for your visualisation. If you need more – use variations.

Traditional use of colours follows generally accepted rules in the society. For example in Western countries green is generally associated as a positive signal. And if you use green to visualize actual sales are below budget this will simply delude the reader. Another example is using pink for men and blue for women, red for Boston Celtics and yellow for Chicago Bulls etc.

Consistently simply means that if you use dark blue for product A on the first chart or page you should use it in all other charts and pages.

Why is this important? Visualisations are easier to read and assimilate which saves time for analysing and interpreting the information

3. Eliminate visual noise

Visual noise are all elements in a chart that actually don’t present any information – gridlines, axes lines, borders, area filling, legend etc. As a general rile they should be avoided and even completely eliminated. This makes charts lighter and more visually appealing.

Why is this important? Visual noise distracts the attention of the reader from information. Sometimes it could literally cause a headache.

Conclusion

Creating data visualisation that is visually appealing and communicates information efficiently is not an easy task. However, if you think for a moment how to explain it to a person that is completely unfamiliar with the topic, like a dummy or a small child, it will become much easier. The three rules above are not the only ones you can use. But from my experience, they will perfectly work in 90% of the cases.

By: Kolyu Minevski

 

Leave a Reply

Your email address will not be published. Required fields are marked *