Creating visualizations from data can be a powerful and intriguing way to present findings. But way too many design teams sit on vast amounts of data. They also spend entirely too much time making static images rather than interactive tools.
In his virtual seminar, Data Visualizations that Pack a Punch, Brian Suda outlines different types of meaningful data visualizations, from charts and graphs to more interactive models. He also discusses the importance of using the right tools and newer technologies and higher resolution displays as they emerge.
The audience asked a slew of great questions during the live event. Brian comes back to chat with Adam Churchill and tackle some of those questions in this podcast.
- How do you approach accessibility challenges, such as color blindness?
- How do you communicate that the data you’re presenting is “fresh”?
- Is there a good way to demonstrate the ROI of good visualizations?
- What can you do to encourage people to start exploring and using data?
- Are there any examples of companies currently using visualizations well?
- Should you try to do this in-house or is it better to outsource to an agency?
- What is the best way to get started?
Adam Churchill: Hello, everyone. Welcome to another edition of the SpoolCast. Recently Brian Suda joined us to present his awesome virtual seminar “Data Visualizations that Pack a Punch.” Brian’s seminar, along with 110 others that teach the tools and techniques you need to create great design, is now part of the UIE user experience training library, soon to be unveiled as UIE’s “All You Can Learn.”
In this particular seminar Brian shows just how his work with companies such as PriceWaterhouseCoopers is defining just how powerful great data visualizations can be, especially when they’re tied to marketing or social media campaigns, and that raising awareness, conveying meaning, and getting users to interact.
Hey, Brian. Thanks so much for making some more time for us, and welcome back.
Brian Suda: Not a problem. It’s always fun.
Adam: If we can start this way, people that weren’t with us for your virtual seminar, can you give us an overview?
Brian: Sure. The seminar was about an hour, hour and a half long, of me talking about some of the interesting data visualization techniques that companies are using today and looking forward to into the future.
Specifically we wanted to ease people in who aren’t familiar with data visualizations and talk about the different types. From basic charts and graphs, all the way up to more interactive things, into these new ideas of these simulations and how we can simulate and actually interact with the data further than just clicking on buttons and checkboxes, all the way up to what’s happening in the future.
We wanted to give people a really good grounding and understanding of what is out there and what the capabilities are. We also talked about workflow. Any big company now is really in this inflection point where how do we change our workflows so a lot of the data visualization work that we do, and time and effort we put in, can be repurposed for both print and for web and potentially other needs as well.
That requires quite a lot of thought and purposeful design put into it, especially nowadays with more and more devices with high-resolution screens. We talked a lot about the difference between rasterized and vector graphics when it comes to data visualizations.
Along those lines we also discussed a lot of the how do you plan your campaign around your data visualization? A lot of companies will spend months working on actually getting some sort of product out there, but then they tend to fall down and forget about how to actually market that product once it’s produced.
I know a lot of people would finish the thing and then send a massive press release out. In the past it worked, but there’s always the risk of other news or something trumping you that day and then all the time and effort you put into, it’s just crickets on the website because no one’s actually visiting.
We talked about sort of not only planning out the production cycle to generate this data visualization, but also some sort of production cycle to promote it, as well. We discussed some of the ways that’s possible.
We also got into a bit more than nitty-gritty where we talked about some of the more common data visualizations that small and big companies have access to now, which, in the past, seemed quite daunting.
I think a lot of people tend to forget that an HTML page will work offline just as well as a Word document. We discussed data tables, we talked about sort of the next logical step from a data table into dashboard design, how to create a beautiful and useful looking dashboard for your company.
Then we talked about maps. Maps used to be one of these things that were really, really hard and you might have only had one or two choices for the look and feel. Within the last year or two, it’s really exploded in the flexibility and the design and having custom maps just for your need. We touched on some of the ways that people can get that in there.
Then finally, we wrapped it up with some hopefully inspirational ideas on, not only the data visualizations that you have access to today, but what I’m seeing, very tip-of-the-iceberg, what’s happening now. It’s going to be really big in the next year or two.
Adam: Brian, there were a few people that had questions about challenges such as color blindness and other accessibility issues with data visualizations. Any tips or thoughts on how to make those effective?
Brian: Sure. One of the big things that we try and always think about is the accessibility features. When you start getting into these really complicated data visualizations or graphics, we tend to forget about how they were created in the first place.
A lot of people might come from designing on the back of a napkin and then transferring that into something digital. My background is in programming, so I tend to think about it from the database, data store, how it’s structured, and then move it to the design.
A lot of the times when people are thinking about accessibility or how many people are going to actually have problems looking at my chart, one of your biggest consumers of accessible ready data and design are search engines. So it’s been a big thing which we’ve always considered.
On the topic of color blindness specifically, that’s one of those things that affects I think one out of every 10 males. There are a few really nice apps out there. One I tend to recommend is called Color Oracle. And it’s a really nice little app that is for all Windows, Mac, and Linux. It just sits in the tray and you can actually change the entire color of your screen.
For a brief moment you can actually see the world through someone else’s eyes. And it’s really useful for spot checking every once in a while your design to make sure you’ve got enough contrast and color differentiation. When you say something like, “Take a look at the red line” or “This line needs this in your key,” you can actually tell the difference.
Adam: What is a good way to make it obvious in your visualizations that the data you’re working with, or that you are telling a story with, is fresh, that it’s new, that it’s up to date?
Brian: This is always a pet peeve of mine as well. A lot of the time I’ll visit a web page or something like that and it’ll be a great article or a great tip, but there’s no way to tell when the text was written. The simplest and most obvious way to fix this is just put a date on it. If an article said this was published in 2011, I might go and seek out a better tip or some better advice.
If it was published a week ago, then I can probably trust that the numbers are correct. The same goes for any charts and graphs. People tend to forget really simple things like put a headline on it to tell you what it means and put a date on it somewhere to tell people when the data was collected and generated.
Adam: Is there an easy way to convince clients or the people that you are reporting to on the project that these visualizations that simplify data are going to offer up a good return on their investment?
Brian: That’s always a tricky one. I think a lot of people are afraid to spend the time and money on developing a lot of data visualizations. Maybe upper management thinks it’s a fad or it’s pretty pictures. But in this day and age there’s a lot of discussion of big data and how that’s effective in big business. Even in small businesses.
If you want to figure out if your data visualization is creating a good return on investment, you really need to be smart and tie it to some sort of business metric. Maybe if you’re an e-commerce site, your goal is to generate repeat buyers or change the price of the cart — have people checking out with more expensive items.
The simplest way is as you’re generating any sort of data visualization, either for the customer or internally, you have to be able to tell if that is affecting some sort of business metric. Things like dashboards make it easier to make business decisions.
If it’s easier to make a business decisions, you can make more decisions per day and it’s quantifiable. It’s important that when you go forward on designing some sort of visualization, there has to be a why. Why are you doing this in the first place? And tying it back to some measurable business metric.
Adam: Brian, Jeff asks a question about this concept of opening up the big data within your organization, and actually getting people to look at it and use it and try to understand it. What’s your experience with what’s been the key factors to determining the success or failure of that. How do you train people or encourage them to explore and use the information at their end?
Brian: That’s a really good question. It’s difficult to encourage people who don’t necessarily want to. If they’re just not interested, it’s hard to get them interested. Historically a lot of charts and graphs tend to just put people off. I think that’s why we see a big explosion in these colorful, cartoony infographics.
While those might generate some attention, and maybe that’s your goal, that’s fine. But they don’t seem to have either good shelf life or not a very deep dive into useful information. They don’t let people explore things.
In the presentation, one of the big things that I’m harping on is that you want to shoot for building a tool rather than just a pretty infographic. If you can build a tool, that’s a central hub which you can drive people to. It becomes a resource for people.
One example might be world GDP data. Every country in the world has some sort of GDP data that they might release annually or quarterly or something like that. You could easily do a beautiful infographic about this every year, but the shelf life of that probably falls off quickly after it’s produced.
If you can build some central tool that either academics or journalists or anyone who’s interested can go to and play with and click around and interact with, but also then choose their own story. They might be able to say, “I want to look at this country over this country and then generate some charts and graphs from that.” I think that has more sticking power.
Finally one of the ways that we talked about in the presentation to get people interested and generate interest might be to just think outside the box. We talked a lot about simulations, which take your interactive graphics and interactive visualizations, and take it to the next level where people can input their own feelings. “I bet this is going to happen,” and then they can play it out and see where it goes.
The ultimate simulation is a video game. We talked about games like SimCity. SimCity is a massive data visualization of what’s going on in the city, from transport to taxes, to health and fire and police services. That is a really interesting way to get people engaged in understanding by letting them into the driver’s seat to play around with the data. They don’t even necessarily know that they’re working with a data set because they’re just playing a game.
Adam: Brian, we actually ended the virtual seminar with this particular question. It’s a good one and it gets people looking around. What are some examples of some big companies that you’ve noticed that are doing some interesting, innovative, exceptional work with data visualizations?
Brian: Certainly at the moment a lot of the big newspapers. New York Times has been doing quite a lot of really interesting both infographics, interactive graphics, and they’re using it to really advance and tell a story. It’s not just supporting the story with a little sidebar with a nice bar chart or something. It’s really part of the story.
The most recent one I saw was Mayor Bloomberg and how he was rezoning New York, and they can talk about it. Then they have this really beautiful interactive map that spins around and shows you the developments from year to year. It’s actually really a part of the story. If you’re looking for some great inspiration, definitely newspapers like The New York Times, The Guardian, are really on top of this sort of stuff.
Another big company that has really embraced big data and data visualizations throughout the whole company has been GE. GE has a whole data visualization blog just dedicated to looking at the data within the company and some of their products and their verticals that they interact with. It’s really, really well done.
It shows that this isn’t something that’s only relegated to a small team of five to ten people. This is something that’s a big sea change even within a massive company like GE.
Adam: Patricia asked a question about trying to do it in house versus not really being sure, and outsourcing it to an agency. What are your thoughts about trying to do it in house, where you’ve got somebody with some web design skills or really there’s some agencies out there that specialize in it and maybe would do a better job? That’s the concern. That’s the question.
Brian: There’s always going to be a learning curve. If you wanted to do it in house, it would be a long term commitment. It’s going to be a little while before the people come up to speed on both how to generate data visualization but also the politics and red tape within the company.
Data visualizations, whose role is it? Is it part of the design team? Is it part of marketing? Is it part of some online team which actually manages the website? There’s always where does this weird thing fit into the work flow and the politics within the company. So bringing someone in house is definitely going to be a long term commitment.
On top of that, everyone in the company needs to be committed to the project. They need to realize that if you wanted to be even more transparent, or deal with the data, individual silos — the verticals in the company — need to keep the data in a format that the data scientist can actually work with.
If you wanted to be really guarding and sitting on your data and only give it to people in PDFs or not give it to them at all, that’s a recipe for disaster. On the flip side, hiring a firm on the outside can achieve quicker short-term goals. If you needed a data visualization this quarter and you don’t have anyone in house to do it, it might be smarter to outsource that.
It’s a great way to dabble in it as well. Maybe the company isn’t sure that this is what they want to do. They don’t feel that they have interesting data, or it’s too sensitive. Maybe if we start doing these things, our customers are going to get scared.
Certainly outsourcing, to see if this is something you want to do, is a great way to start. It’s probably a better way than bringing some staff in house and then taking six months to a year to find out that it’s just not going to work out.
Adam: Let’s talk about those skills and how people get them. There were a couple of folks that were wondering, “How do you get started in a career in information design or specifically with data visualizations? If you’re already doing some of it, where are some places to get inspiration?”
Brian: There are two ways to approach it. I think of data visualizations a lot like photography. Photography, is it the domain of the technical side where you’re actually worrying about shutter speed and aperture and the technical, lighting aspects of it? Or do you come at it from the more arts and design side where it’s composition and thoughtfulness and what makes up a good picture?
A lot of the skill sets overlap in the middle, but there are definitely specializations that if you’re coming from a coding aspect you need to know and if you’re coming from a design aspect you need to know. If you needed a checklist of important skills, I would say definitely some sort of HTML, CSS for web-based.
Then I would recommend some sort of programming language. You don’t have to be good at it. You don’t have to consider yourself a programmer, but when people give you really nasty, gnarly, ugly stuff, you need to have a way to convert it around.
Knowing a programming language even can save you loads of time as a designer because, as we talked about in the presentation, you can always, from code, generate SVG. SVG can be opened up in applications like Illustrator, and saves you loads and loads of time.
That would be my short list of skills that you would really need to understand before you can work in data visualizations.
Adam: Very cool. Well, your seminar was fantastic, and I certainly appreciate you coming back and joining us for a little bit more time on data visualizations.
Brian: Yeah, it’s always great. It’s an expanding field, and it’s really fun because it’s one of these new online topics that you can really be in the cutting edge and on top of, in front of the wave. It’s really an exciting place to be.
Adam: To our audience, thanks for listening in and for your support of the UIE Virtual Seminar Program.