When you title your book on data visualization, Information is Beautiful, it’s safe to say you are passionate about the topic. Truth be told, the data visualizations created by author David McCandless, a self-described “information designer,” are quite beautiful.

AFP recently spoke with McCandless, who will speak at its Annual Conference in October, from his home in London.

Learn more about the FP&A Luncheon here and be sure to register for the AFP Annual Conference by September 16 to save $200.

AFP: I found your data visualizations pretty astounding. Obviously, they don’t take 20 seconds to put together. Walk me through your creative process—perhaps using your “Top 500 Passwords Visualized” as an example. How did you put that together?

David McCandless: It was actually inspired by another graphic I saw about bespoke PIN numbers, four-digit PIN numbers. I saw the most common PIN numbers that people just pick and choose. I thought, what would an English password be like? So that one was quite a data-led image. A lot of my images are actually concept-led. I actually think of a question or I have a scenario I don’t understand or a subject matter I particularly want to tackle. And then I develop data visualizations based on my actual data later. This one actually began with the data. It’s like you can access.

Unfortunately, when data breaches occur, ultimately it would leak information. I actually compiled the data from several big data breaches and did analysis on the English language passwords. And then looking at the data, I saw that various passwords fell into certain categories. There are certain categories to passwords. I don’t know which version you’re seeing but I definitely did the analysis on the different passwords and tried to tell a story about the choices people make about their passwords.

That was what I would consider like a data analysis story which is coupled with a quantitative analysis. Finding those patterns, more human patterns in the data, and then we were staking it in a data story and then building up into an interactive as well online. I just thought that was intriguing. Also the subject matter just meant we all have a stake because we all basically have to generate passwords. I mean we use passwords. Do we think clearly about a password?

AFP: Earlier you said usually it’s the other way around for you. You conceptualize what you want and then you look for the data. Is that correct?

David McCandless: Yeah, that’s right.

AFP: That’s an interesting concept. I think a lot of treasury and finance professionals probably don’t do this. I think they probably look at the data that they have available at their fingertips and then they concoct some kind of a very rudimentary chart. But you, you did what you were thinking and then you’re saying, “Well, let’s dig for the data and see what we come up with. “

McCandless: First, my approach is very similar to what you might call design thinking, which starts mainly with a problem or some kind of emotional situation or an emotional response to a situation or a feeling of something not working. That way, the clear process is about solving a problem. It’s more of a design or an intake on things. That means the end result is generally more useful and more compelling to an audience. So it’s more useful for an organization because it’s not really about taking some data and making it beautiful. It’s actually about addressing some core question or problem or issue that might be at the heart of the organization or, more broadly, just something that we all wrestle with. That is just more useful. It’s beautiful and it’s useful.

AFP: But you’ve touched on something that probably a lot of folks struggle with, which is that they may actually be under orders from their bosses to create a certain type of chart. I wonder if you have an advice on how they can make charts more interesting or tell a compelling story without annoying the boss because the boss might say, “I want to look at daily revenue numbers, and you just got to tell it.”

McCandless: I mean, it works that way as well. I got some general advice about how to choose more interesting charts. I think that’s one of the things that’s happening in visualization at the moment. I see these as a sort of language. The charts you use are like the words of that language.

I think what’s happening with visualization is we’re going slightly beyond the everyday select school board if you like. Especially with traditional and almost clichéd ways of presenting data into a time where there are a few more richer way of expressing data in visual form, use that to charts or bringing in new design styles or you define quality to lift up the traditional displays. So more like detail and you talk a little bit and explain a lot about this.

Just expand the vocabulary and think of play, be a bit more creative in your expression of data. Yeah, maybe you can’t find a really strong concept or a really powerful question, but you can at least use the cliché level of what you present, make it more impactful, a bit more memorable, a bit more colorful and stylish, a bit more interesting just to look at so people just pay more attention.

AFP: When you’re working with business people, you’re consulting and advising them, what’s the most important lesson that you try to impart on them? What do they most often lack when you view their work?

McCandless: A little bit will just touch to just defaulting, to more cliché PowerPoint type language and appearance. I think also I see a lot of people that maybe don’t have a clear business goal for what they’re engaging in, especially when it comes to data visualization. The data box is sort of voluminous. There’s no end of data. I think if you open that box without a clear goal in mind or a clear purpose, you can get just overwhelmed and just go around in circles.

So I just encourage people to be very clear about what their goal is maybe from a creative point of view but also from a business point of view. That acts as a thread, if you like, to guide you through the data and to decide what to include or to exclude and to know when to stop. Because I think a lot of -– what I find anyway personally when I’m exploring a data set, it often leads to one question and it leads to another question and it leads to another question. You just kind of keep going on and on. So it’s good to have some containment and a clear goal, a clear direction maybe if you’re able to communicate. I mean, the goals might change depending on the audience. Just be clear about your goal and the audience and streamlining individual presentations so that it fits that.

AFP: You published one book titled Information is Beautiful, and then you published another book titled Knowledge is Beautiful. What’s the difference?

McCandless: That’s one of the themes of Knowledge is Beautiful is exploring the difference between data and information and knowledge and how you might be able to just convert one into the other because, obviously, there’s an increase in value it seems as you go up that chain and an increase in the ability, the sensibility and understanding.

I wasn’t sure so I was exploring those notions. If you treated each of those things, take the information or knowledge as a substance or just as a metaphor, how they might appear, how might different visualizations suite different modes, what is data visualization versus the knowledge smack [sounds like] the common core, and what kind of processes are involved in converting them.

I go into a little bit of detail around that. I’m actually thinking instead of three substances to information and knowledge, there are actually six. There are some in-between stage that actually we encounter. For instance, between data and information, there’s a mass substance called or we call structured information, so like databases, spreadsheets, and so on. So there’s an in-between stage.

But generally speaking, data is like particulates. You talk about data points. It’s quite so scattered and disorganized. And then you arrange it and structure it into information that becomes easier to digest and communicate. And then beyond that when you are able to richly connect lots of different types of information and be quite comprehensive about it, you get something that’s quite cellular in network that’s a bit more like knowledge. It’s what you would call knowledge.

Read an expanded version of this interview in an upcoming issue of Exchange.