Chris Farnum: Jared, I'm so excited that you're excited about my infographic, because I was really excited too. "Look, I made an infographic! That's what I'm here to show you guys. What do you think? Isn't it awesome?"

[laughter]

Chris: We can even get a little bit closer. Oh. Well, all right. I guess maybe I should tell you a little bit about why. Backing up a little bit. I live in two worlds right now. I work at ProQuest. I rejoined ProQuest in 2009 in order to work on a project for a search platform for students and teachers and faculty members, usually at colleges but also sometimes in special libraries like government or business, to be able to search for academic journals, magazines, scholarly research, firsthand resources and so on.

It's something that you would get via subscription of an institution you belong to. If you're a student at a university, it's very likely that you've been exposed to one of our products. But we're not a household word like Google and so on.

I worked on that for a couple years. My current project is one called InTota, which is based out of Seattle. There's a company called Serials Solutions that became part of ProQuest a few years ago. They make tools so that libraries can do a better job of managing their electronic collections and giving access to not just stuff from ProQuest, but from lots of different vendors.

They have a suite of tools for helping you get from a citation you find in one source to full text in another and also some other solutions, like a solution called Summon to search across all those things and present it in a nice, friendly, Google-like interface.

In order to make all that happen, you need to have a good administrative console that your customers, who are librarians in this case, can log into and manage their electronic collections. It's a lot like managing your library, your whole entire physical collection. That's the space we want to move into. I've been working on that for about a year, year and a half.

Previously with the ProQuest search platform, where it's same audience, same kind of customers, the decision-makers who buy this product are the same. They're librarians, library staff people, administrators, and so on. There's that commonality. But very different products, really. One's for managing your collection. The other is for searching your collection.

This is a project that's been through a lot of this. We've had new leaders come in. We, in December, had a little bit of a change and refocused on our objectives. We even had UX team members join from a small external agency based out of Silicon Valley. They're in the same time zone, by the way, unlike me. I'm a little bit challenged in that way, working remotely.

We started to do things like try to go back and pick up on some UX process stuff that the new folks had missed and get them up-to-speed on the product objectives. We had to go back to things like what are the personas and what are the user's tasks and think about design direction overall.

As they were coming on board and we were meeting for the first few times and trying to get a feeling for what we're doing, I realized I had an idea. That idea was that we have a robust analytics programs for the ProQuest platform, which is that thing I described first, which is searching journals and magazines and so on. We use things like Webtrends and ForeSee, and we have this growing data set, and we have a digital analyst who has been working on a series of reports and really making that program come together.

Since it's the same target audience, if we could just get down to the part that's the use of librarians and all these things that we're measuring in our analytics programs, then maybe I can cross-pollinate the InTota project. I can find out some important things that you might want to know when you're starting a project, about demographics -- what kind of devices do they use? Do they use social media? Anything I can really glean from what I can find out about librarians. There's a whole lot of data I can't use because it's very specific to the ProQuest search interface, but there's other data that I could take advantage of.

I had put my idea in the form of that little rebus diagram that you saw, and I even added it to the infographic.

Digging for analytics gems. One of the things, like I said, about that digital analyst is he sat right next to me for a number of months. He just last week moved to a different floor and a different cube, but we're going to stay close together. His name is Ferris Khan, and he's a talented analytics lead at ProQuest.

Some of the programs that he administrates involve Webtrends data collection, and that includes things, of course, as you know, like clicks, pages, sessions, refers, and so on, and also ForeSee's satisfaction survey program. Are you all familiar with the ForeSee product? You basically know what it is? Great.

Their primary measure is satisfaction, but users who are intercepted wind up taking surveys after they've hopefully done a few things on your site. They're asked, "Do you want to take a survey when you leave?"

As a customer of ForeSee, we get to design the questions and not only do we allow questions but people can put in open ended comments, as well.

This gives us this sort of infrastructure of analytics. It gives us all kinds of charts and tools and things to analyze, dashboards that we can use to go in and look. A lot of this is done by Ferris, but he's also provided some training to the UX team and others in the organization through ForeSee so that we can get in and muck with it ourselves a little bit, although we're all pretty new at it.

Finding information is a little bit like finding a needle in a haystack, of course. Ferris does monthly and weekly reporting where we get a lot of great insights but it's kind of at this stakeholder high level and so on.

And so, a lot of the questions you have as a user experience designer are more granular. Like "Did the click on something?" Or, "Is this design that we just launched working better than the one that we just got away from?"

It really helps to be able to have conversations and to work with someone who can help you navigate the data and all of those stacks of information.

My approach in digging for gems here was that, of course, I wanted to ask for help. Ferris told me that one of the important things you do is create hypothesis and you're brainstorming questions about, "What do you want to know about these users?" Because I knew I wanted to ask questions about librarians and how they use ProQuest and some of their characteristics, but through talking with him we were able to get at some of these questions.

Again, training and access to the tools was good. Ferris and I sat in the conference room with a projector running and he could take me on a tour of the ForeSee tool and show me some things that were very helpful in terms of narrowing down to the right persona. Because based on the questions asked in the survey I'm able to do that, to pick different audience groups.

I also needed to try to gather all this data and analyze it. It was all good info. I was collecting things like charts and graphs and statistics and so on on my questions from ForeSee but it's kind of dense. Charts are charts and so on. What I really needed to do was to try to find some way of communicating this to the team and give them something that they would actually read.

I began to think of...Well, in past deliverables I've had to do an executive summary and come up with a way to get the bullet points to people. And so, I had a thought about this as I was looking at cool things that come across Twitter and also things that circulated around our organization. The idea was that an infographic is a great way to summarize stats, to create a message that gives people takeaways and, "Hey, maybe I should try to make one myself," which I really hadn't done that before.

I'd drawn lots of wireframes. I've been doing that for more than a decade and a half now. But I hadn't really tried this infographic idea before. There's lots of models out there. I was actually kind of inspired by one in particular that went around our organization from Project Information Literacy which had done a large research study on academic users.

What I noticed about this was, "Hey, I could probably do something like that. It looks like the graphics are fairly simple and digestible. A lot of it has to do with things like getting the fonts right and getting your message right."

I kind of felt like, "Well, I'm going to give this a shot," so I wanted to back up and educate myself a little bit and figure out what I could do to draw more inspiration and try to get it right. I wanted to step back and study this a little bit.

With some suggestions from Ferris and also looking around a little bit, I found there are sites like Visual.ly which is a showcase of infographics so people can upload and submit their own. It's kind of like SlideShare in that way or YouTube where you can share your infographics. They subset down into particular categories like education so I was able to find some examples that way.

There are also tools out there to help you create infographics like Many Eyes from IBM. Now the thing about Many Eyes is that the idea is that you have a lot of data and you upload it, and then you use their tools to manipulate it and create a chart.

But I had already gotten charts for ForeSee. I had already sort of done that step, and I didn't feel like I wanted to go and take the effort of taking data out, reuploading it, and then remanipulating it. What I wanted to do is to get to the infographic piece and summarize what I'd already been finding out.

But they also have a blog which has some advice from the pros and some great tips for getting started, and so I was able to glean that. By the way, along the way, I found an infographic that mocks infographics in general, which was kind of funny. It's very self referential.

On the surface when you're zoomed out, it looks like a really nice infographic, and then you see irony dripping from it as you get a little bit closer. They're really harsh on things like banner ribbon headlines like this one here in the middle.

Oh, OK, I thought wow, infographics are so mature now that maybe they're driving designers nuts, and there are haters of infographics, which I suppose is true. The 90 percent rule, right? I tried to glean what I wanted to do in terms of goals for making an infographic.

I came up with a couple of tips that I wanted to share with you, which are things like analyze until you get to headline stage. If you message that you're trying to communicate is too complicated, it's not going to fit in an infographic. Save that for the pages afterwards.

A lot of people dive into the details in your report or your collection of graphs that include afterwards. But communication's the goal when you're designing an infographic, and that communication has to be quick. Remember, these are people possibly with short attention spans who have to build something really quick.

The rest of the team doesn't have time to read a long report. You need to give them something on page one that's going to work. Use just enough color imagery. That kind of works for especially because again, I am not an artist.

But it's also true because it's not about your beautiful picture you're making. It's about communication again. Simplicity is important, and also, you can apply your information architecture skills to organize this in a way that's going to be meaningful and define themes and organization schemes that make sense.

As I've already kind of pointed out, pick a way of doing it that kind of fits your abilities, such as they are. The method I chose was I used sort of a comfortable standby tool, Vizio for me, which is easy to use. But pick your own, feel free.

For me, it was a case where I wanted to constrain myself with basic colors and white background, which would make things easy to cut and paste. That's not the only way to do it, but it's the way that I chose. And then extremely simple shapes that I could create in Vizio or that I could use from clip art, but try not to make it look too much like clip art because again, I think the 90 percent rule is even higher in that case, for clip art. It's scary stuff out there.

I want to run through sort of a chain of process that I did for one particular piece of my infograph. The question that I was doing in this model was should we consider mobile and responsive design for this particular back end office tool, and which devices matter?

This was one where I was only able to get just so far with data that we had existing from the Proquest platform, but it was still helpful. I had to come up with a hypothesis first. Ferris told me, and this is an important point, to start out by just making a hypothesis, even if it's wrong.

Because in your scientific method classes, if you remember that, then you start out and experiment by creating a hypothesis, and you can prove or disprove it. Now or in the near future, it's going to be important to support mobile use of InTota by librarians.

Some of the questions that arise from that, one of the questions would be are librarians more or less likely to use mobile devices than the rest of the population and which mobile devices do they use? Using the ForeSee dashboard in this case and then also for other parts of the question, bringing in some web trends data, I was able to do things like segment down just to the librarian roles.

I filtered so that I was only looking at surveys within a particular date range that people had said I'm a librarian or a library administrator when they took the survey. And that gets us to the right persona for InTota.

After I'd done that, then I would take a particular question from the survey and luckily, we had a question for this, which was do you own and/or use any of the following mobile devices. Select all that apply.

I had a nice chart I was able to pull out of ForeSee, and it was for a collection of surveys for the date range I picked. From September to January, I had 461 surveys, which is much better than going out and interviewing 461 librarians or sending out an ad hoc survey and waiting for the results to return. I could get them all from this one shot data that had already been collected.

I took the graph and kind of looked at it a little bit and played with the numbers and was able to come up with a few headlines from it, like we do need to design for mobile devices. Now, notice I didn't say when. Maybe it's not a first goal, which is kind of what the team wound up deciding from this.

But we figured that if 25 percent own an iPad, 18 percent own a Kindle, 23 percent own an iPhone and so on, and a lot of Android use as well, it matched what the rest of the population was doing. There was really no difference because I was able to compare librarians against everybody else.

And then looking at the web trends data, we were finding that visits by mobile devices was very small for the Proquest platform, but it was growing. And even though the number was small, it was growing exponentially.

I took that and plugged that piece into my overall infographic, and I did that for a number of other things, like social media use, use of search tools, what OS people were using, satisfaction scores, which again, is a big part of ForeSee, and then browsers.

You'll notice the librarians tend to be a little hard to please. They tend to not give us awful scores, but they just give us lower scores than everybody else. This was one area that I wasn't too surprised about, but it turned out that the majority of librarians were Windows users.

In fact, it was like 97 percent or above. I know it's a little bit hard to read my graph here, but that was the number. And over the half the people were using Internet Explorer who were librarians visiting Proquest.

Working with this new group of people from Silicon Valley, from outside UX people who sort of ride in lofty circles and so on, they had an interesting reaction to it. Here's my zoom-in on the numbers. But then the reaction was oh, no!

[laughter]

Chris: Windows, it's terrible. I'm a Mac user! I'm overstating, but what it turned into was that it really highlighted the need that we had to make sure that were going to not only design but test and look at these designs that we were making across multiple platforms and not stick solely to Mac conventions, Mac testing, and so on.

I think that this really did help the newcomers better understand the target audience. The things I was able to pull, with what information I already had, informed them and really helped out.

Again, mobile was something what was identified as important. Maybe it's not our first objective, but it's identified as future goal. Also, for questions like the social media questions we ask, it gives us an idea of design patterns that are going to be familiar to our target audience that we might reuse and modernize the site.

So it's question and comment time, and you'll notice my very un-ironic use of the ribbon banner here.

[audience laughter]

Announcer: Chris, thank you.

Chris: Thank you.

[applause]

Chris: Do you have a question?

Announcer: OK, Christina.

Chris: Hi, Christina.

Christina: Hi, Chris. Thank you very much. I have a question. You created this awesome infographic full of very useful information. How did you make sure people used it? It's like the 96-page usability report that nobody reads.

Chris: Well, a couple of ways. One was that, in the process of creating it, I brought it to the newly formed UX team, which is this blend of internal and external. I brought it to our teams for review. As they were giving me feedback, they were also digesting the information. For my small group, this UX team, I made sure it was in their face that way, but it also improved the product and made it better.

Another thing I did was, I made sure that when we did our Sprint demo, at the end of the Sprint -- because it was sort of like Sprint Zero -- then I was able to make sure that I got in front of the rest of the team, told them where it was on the shared drive, and published it that way.

I probably need to make sure that I call attention to it again from time to time when the moment arises. I haven't quite figured that one out yet, though. Other questions?

Male audience member 1: I have one. What was one of the more interesting reactions you got from people when you distributed this thing?

Chris: [laughs] Let's see. I think it was really that one that I got with the, "Oh, no! It's Windows and Mac use." For a town like Seattle, there are actually a lot of Mac users there, especially in the people who work in the city, as opposed to across the bridge.

[laughter]

Male audience member 1: Yes, I think that they physically divide Seattle into the Mac side and the Windows side.

[laughter]

Chris: I think that the reactions had a lot to do with the data in the messages I presented, and a little less on my format, which is probably a good thing. You want information architecture to be very usable but invisible. I wanted to communicate the information, so I think people liked the infographic, but I got fewer comments about the format than I did about the content, and that's really, I guess, the way I would prefer it.

Announcer: OK. Any other questions we have for Chris here? Or we send them away? OK, Chris! Thank you very much.

Chris: Thank you.

[applause]