In a previous, Project Management Dashboard, I wrote specifically about a dashboard for project managers. In that post, I made the point that a dashboard should tell a clear, crisp, story that simplifies any complexity without loss of information. Doing this, creating this clear, crisp story really brings together two different topics I have discussed in recent weeks.

In my post From Data Literacy to Transliteracy, I made the point that data literacy is the ability to transform data into information, noting that the transformation process puts the data elements into context. This was part of the discussion concerning project management dashboards, that we are putting into context the status of the project. Expanding this concept to a general dashboard design principle, if you are going to build good dashboards, you need to be able to transform data into information, i.e. become data literate. The problem, however, is that data literacy is no longer sufficient. We all now live in a world of big data. As trite as this may sound; it is true. An increasingly larger volume of data is coming at us faster from a greater diversity of sources. This drives the data literate to a higher level, to become transliterate. (For more information on transliteracy please refer to the above-cited article.) This means that in a big data environment, to build dashboards well we cannot be content with data literacy, we must strive for transliteracy. Let me explain…

If we are to create a dashboard that communicates a story that could easily be understood the images on that dashboard must be easily understood. So, how do we do that? Let’s look at an example:

I have heard visualization experts describe this image as the best use of graphics ever. What do you think? At first, I did not like this image, mainly because you need words to understand what it is showing. So, let me give you some words. This is a map created by Charles Joseph Minard, a nineteenth century pioneer in information graphics. It presents the losses of Napoleon’s losses during the 1812 Russia campaign. It shows the course of his troops’ advancement and retreat, the gold being the advance and the black being the retreat. The width of the line represents the number of troops. The bottom section represents the time and temperature. Got it now? Take a moment to study the graph.

As I said, when I first looked at this image I didn’t care for it. Then I started to study it. Without knowing the details, it tells the story quite well. As I look at the beginning of the bold gold line, I can see in my mind’s eye the more the 400,000 troops marching, self-confident, ready to engage the enemy. Then I look at the narrow end of the black line, I can see a beaten band of survivors limping back to where they began. Also, notice that every time the retreating army crosses a river the black line narrows. Bearing in mind that the width of the line represents the size of the army, we need not know a great deal to understand that when a retreating army is crossing a river they are subject to attack by their pursuers. The point to Minard’s graph is that he displayed six different dimensions of information in a two-dimensional image, he tells a crisp story that is relatively clear. Although the graph needs a bit of explanation, we need to give him a bit grace. He as working nearly 200 years ago with just one medium. Now here’s the bit where transliteracy comes into play.

A key characteristic of transliteracy is the ability to communicate meanings and new knowledge by using different tones, genres, modalities, and media. The diversity of data that is inherent to big data is not just in terms of input, but OUTPUT as well. Yes, we are receiving all sorts of new data from much more interesting sources, but if we in the business intelligence and data science world intend to keep pace, we need to present the information we have drawn from data in much more new and interesting ways. We need to leverage new tones, new genres, new modalities, and new media to communicate.

There is an opportunity cost to being a business intelligence or data science professional. We will not be as knowledgeable of a particular subject area as users. Let’s use finance for an example, the time we spend developing systems, learning new methodologies and technologies is time not spent delving into the nuances of finance. The reverse is also true, the time a financial analyst spends learning his or her craft is not time spent learning systems. So, we cannot anticipate every question they may ask nor can they necessarily develop a system that will answer all their questions. Unless… unless we can provide them an environment in which they don’t need that special knowledge. Unless we provide them a way to interact with the data. One of the key points in good dashboard design is creating an environment in which users, all users, can explore information in ways that at the time of dashboard construction we may not have anticipated. Now here comes the challenge…

With the Internet of Things (IoT) we now interact with systems in new and different ways. Yet, our dashboards are still the same old bars and charts. Frankly, I am sick to death of bars and charts. Pie Charts. Spider Charts. Rose Charts. Where is the innovation? How far have we really progressed from Minard? Here is the challenge for all of us; let us advance to the next level. Let us advance beyond simply rendering two-dimensional images on the screen to leveraging new media, new methodologies to genuinely empower our users.