Reading the book „Data Visualisation“ by Andy Kirk, I ran into an interesting chapter Trust vs. Truth. I liked the idea and the reasoning by the author. Thus, I decided to summarize this chapter with the addition of my own thoughts.
If you visualize data, you must create truthful and trustful work! Sometimes it is a challenge. Why? Because the interpretation of truth and trust could be different.
These two principles are very important, and they are the fundamental of every visualization. let’s take a detailed look at these two terms.


“Every number we publish is wrong but it is the best number there is” Andrew Dilnott, Chair of the UK Statistics Authority

We know the example with glass and water: The glass is half full is also half empty. Both claims are true, but which ones should we choose? It is your decision and you decide how to tell the data story. In these cases, you are faced with choices without necessarily having the benefit of theoretical influence to draw out the right option.

The truth in your visualization is an obligation and you must stay objective. Don’t express your own opinion on your chart. Never create the work you know to be misleading in content, nor should you claim something presents the truth if it evidently cannot be supported by what you are presenting.

The truth should not only be present as the result on a chart. You should present the truth already by data preparation. Be careful by combining different data sources, excluding and including the data, by rename dimensions or measures, by creating any calculation fields. Check the intermediate results and ask yourself whether the data still corresponds to the truth.
Here is a short summary of how to create truthful work:


• You should be well-informed about the topic you visualize. Perhaps, research work is necessary. If you create a visualization for customers, then you need a contact person, whom you can always ask questions.
• The statements in your visualization will make a correct impression if you minimize the number of assumptions applied to the data you are working with or by judiciously consulting your audience to best ensure their requirements are met.
• Be objective by formulating the context. In the case of absence, the truth, you need to be able to demonstrate that your truth is trustable.
• Don’t leave your users an interpretative scope by reading your context
• Bring information to the point
• Be well-intended and legitimate choices
• Indicate the source of your research


“Good design is honest. It does not make a product appear more innovative, powerful or valuable than it really is. It does not attempt to manipulate the consumer with promises that cannot be kept” Dieter Rams, German industrial designer.

Trust is a sister of truth. You need to create works that demonstrate your truth is trustable. Or rather: You need to create a lot of truthful works, get some recommendations and references and then your work will earn trust. It could be a difficult and long path. And another thing by the trust is, that it is hard to secure, and you can lose it very fast.
In data visualization, trust means that users can rely on what they see without rechecking the information. The user assumes that your work represents truth and thus they trust your work.

It may occur that a visualization can be truthful, but not viewed as trustworthy. There are a lot of reasons. In order to avoid this situation, I summarized a short checklist:


• Eliminate any sense that your version of the truth can be legitimately dispute
• Support your visualization with a context and with an indication of serious references
• Try to create an elegant design with the correct using of pictures, logos etc. (Elegant design means, that every single decision you make – every dot, every pixel – should be justifiable. Nothing that remains in your work should be considered arbitrary.)
• Your context must be grammatically correct
• Declutter your viz: Eliminate background lines, unnecessary pictures, and logos, be sparing with colors, put important information to the foreground
• Don’t forget to put references and data sources on your chart
• Care about the typeface, font size, and text color
• Use the correct color choice: think about your topic. Is your topic positive or negative? Think about how you transfer the mood to the users. Here is a good example: Iraq’s bloody toll
• If you have a possibility, you can also share your approach and how have you handled with the data. Tell your audience how data was collected and what kind of calculations have you applied to it. Have you removed or excluded any data and why?
• Be consistently by representing the information: Think about color usage and wording
• Ask your friend, colleagues for feedback. This feedback is worth its weight in gold.