image by Tobyotter via Flickr

As a boy, it wasn’t the history, the architecture or an admiration of the faith involved to build these churches.  Those were concepts beyond my ability, knowledge or frankly interest at the time.  What I have come to realize only in the past couple of years is that the windows were meant for me. At the base level, I needed something that could grab my attention and hold it. What I have discovered is that from this standpoint, I am no different than the illiterate masses of the Middle Ages or Renaissance.

I discovered that hundreds of years ago, with a need to engage the European population and educate them on scripture, someone decided it wasn’t the job of masons, who built structures that would last for centuries, but storytellers and designers who could make kids, like me, stop and look.  This was the intent all along as stained glass windows were referred to as “biblia pauperum”, which meant “poor man’s bible”.

Now, with two years under my belt at Juice and hundreds of churches visited, it is interesting to apply the history and beauty of stained glass windows to the field of data visualization and presentation graphics. I now have a better handle on the true value of a designer.  For “design” to work for me, in any type of artistic endeavor, the designer should make me feel that it was designed specifically for me and make it beautiful at the same time to help lengthen my otherwise short span of attention.

As the noise about data visualization and data storytelling grows, it is nice to see that current leading experts in the field also value (and have not forgotten) these 2 design principles provided by our European ancestors.

Consider these two examples:

Design for your audience

• In a recent blog dated 5/10/13, Stephen Few highlights this important customized approach as the 2nd of 7 tenets for best practices of quantitative Data presentations.
  

Beauty’s role in dashboard design

• Back in November of 2009, even before I joined the team, Juice published a frequently downloaded “Guide to Creating Dashboards People Love to Use”. The guide noted that “modern web design has moved on to seek a union of utility, usability and beauty. We must find a similar union when displaying data in business.” (bold and italics added)

What will we learn from these impactful stories, built and told on stained glass sanctuary walls?  Will we preserve the most important principles found on those magnificent etchings? Today, our stories are accessed and downloaded from cloud-based applications and displayed in high resolution graphics on state-of-the-art devices. Yet our challenge is the same: capture the attention and imagination of our viewers – in a user-centric and aesthetically pleasing way.

Hello – I’m a professor at Georgia Tech and I’ve been working in the data visualization research area for over 20 years. My friends at Juice asked me to write a short guest blog entry providing perspectives from the academic data visualization community and exploring ways to foster more industry-academia collaboration. I’ve found that we don’t work together often enough, which is too bad because each side has a lot to offer to the other.

I personally have benefited from business collaborations in many ways. Since data visualization research is so problem-driven, industrial interaction provides an excellent way to learn about current problems and data challenges. In my graduate course on information visualization student teams design and implement semester-long data visualization projects. I encourage the teams to seek out real clients with data who want to understand it better. Some of the best projects over the years have resulted from topics suggested by colleagues working in industry. Additionally, I often employ guest lecturers such as the guys at Juice to come and speak with my students and provide their own insights about creating visualization solutions for clients.

I hope that in some ways my class is benefiting industry as well and helping to train the next generation of data visualization practitioners. Students learn about all the different visualization techniques and their particular strengths and limitations. They also get hands-on practice both designing visualizations for a variety of data sets and using current “best practice” tools and systems. The course has become a key piece of the Master’s degree in Human-Computer Interaction here at GT.

Another opportunity for interaction is academic research forums such as conferences and workshops. Coming up this October in Atlanta is IEEE VIS, the premier academic meeting for data visualization research. VIS consists of three conferences: Information Visualization (InfoVis), Visual Analytics Science & Technology (VAST), and Scientific Visualization (SciVis). Last fall, the meeting garnered over 1000 attendees for the first time.  VIS is an excellent forum to learn about the state of the art in data visualization research, see the latest systems from commercial vendors, and just rub elbows with like-minded friends and colleagues.  Recent papers at VIS presented tools such as Many Eyes and D3, introduced techniques such as Wordles and edge bundling, or just pondered topics such as storytelling and evaluation.  And the meeting has much more than just research papers – It also includes numerous workshops, tutorials, panels, and posters. This year for the first time we have added an “Industrial and Government Experiences Track”. This program is designed to highlight real world experiences designing, building, deploying and evaluating data visualizations. The presentation mode for this track will be posters on display throughout the meeting with multiple focused interaction sessions. Each submission should include a 2-page abstract about the project and a draft of the poster. They are due on June 27th.  More details about the track can be found on the meeting home page.

I hope to see many of you at VIS in October here in Atlanta!

The colored map and detailed drill-down show 37 measures, yet they forgot to include one important freedom: your freedom to communicate data with ease and create interactive infographics in minutes. Don’t worry, we’ve got a heaping-helping of info-liberation. So before you send an angry e-mail to your congressman, take a look at what we put together with Slice in under an hour:

Like the Freedom site, we want users to be able to choose a metric and be able to see which states are freedom-loving and which are freedom-hating (I’m look at you, South Dakota, with your anti-bachelor party policies). That’s our new “map slice” in action, which can color states based on our data or overlay colored bubbles to visualize locations.

To add even more data exploring fun, we created a visualization to let you compare two states side by side.  Check out how North Dakota totally dominates California on freedoms.

Having flexed my information visualization freedom muscles, I’m off to ride my bike without a helmet while drinking a 32-ounce soda.

Choose a time below and sign up to watch our live webinar.

For the East-coast lunch-eating friends: May 1, 2013 12:00 PM EDT

For our West-coast lunch-eating friends: May 2, 2013 3:00 PM EDT

From Edward Tufte’s Visual Explanations, a diagram based on Salman Rushdie‘s description of the Indian epid Kathasaritsagara or Ocean of the Streams of Story.

The hot new concept in data visualization is “data storytelling”; some are calling it the next evolution of visualization (I’m one of them). However, we’re early in the discussion and there are more questions than answers:

• Is data storytelling more than a catchy phrase?
• Where does data storytelling fit into the broader landscape of data exploration, visualization, and presentation?
• How can the traditional tools of storytelling improve how we communicate with data?
• Is it more about story-telling or story-finding?

Many of the bright minds in the data visualization field have started to tackle these questions — and it is something that we’ve been exploring at Juice in our work. Below you’ll find a collection of some of the best blog posts, presentations, research papers, and other resources that take on this topic.

Note: I’ve excluded a lot of excellent sites and articles that use the phase data storytelling, but treat it as fresh way to talk about data visualization.

1. Blog Posts

Storytelling with Data: What Are the Impacts on the Audience? by Nick Diakopoulos
“I realize there’s a whole lot of inspiration out there, and some damn fine examples of great work, but I still find it hard to get a sense of direction…We need to know what makes a data story “work”. And what does a data story that “works” even mean?”

A Data Scientist’s Real Job: Storytelling by Jeff Bladt and Bob Filbin
“In short, we’re tasked with transforming data into directives. Good analysis parses numerical outputs into an understanding of the organization. We “humanize” the data by turning raw numbers into a story about our performance.”

Coffee & Empathy: Why data without a soul is meaningless by Om Malik
“The idea of combining data, emotion and empathy as part of a narrative is something every company — old, new, young and mature — has to internalize. If they don’t, they will find themselves on the wrong side of history.”

Look ma, no story! by Moritz Stefaner
“Tools have no stories to them. Tools can reveal stories, help us tell stories, but they are neither the story itself nor the storyteller. Portraits have no story to them either. Like a photo portrait of a person, a visualization portrait of a data set can allow you to capture many facets of a bigger whole, but there is not a single story there, either.”

Discussion: Storytelling and success stories by Andy Kirk
“I just wanted to share my view on the distinction I personally make between the two main types of visualisation function: exploratory and explanatory”

The secret to storytelling is in the editing by Garr Reynolds
“Although it is a film about the role of editing in filmmaking, the lessons and principles are applicable to other creative work such as writing, and storytelling of all kinds, including presentations.”

Visualising data: can you see stories? by Chris Twigg
“Narrative can on the one hand be broken down into a set of universal laws and principles that may transcend mediums. Stories have temporality in common (they deal with time) as well as causation (they deal with cause and effect of something). On the other hand there are the more media specific narrative affordances as for example in the way that film, opera, novel and data visualisation – because of their physicality and the dimensions open to them – would be able to give a different ‘staging’ of a story.”

Data Visualization as Storytelling: A Stretched Analogy by Zach Gemignani
“For practitioners of the craft, connecting our work to stories feels satisfying — it is a call to raise our standards and an opportunity to enhance the influence of our field. Stories evoke images of rapt audiences, dramatic arcs, and unexpected plot twists. Unfortunately this analogy is a stretch.”

Why good storytelling helps you design great products by Braden Kowitz
“It’s not uncommon for designers to confuse a beautiful looking product with one that works beautifully. A great technique for creating smarter, better products is to approach them using story-centered design.”

How might rhetoric inform information design? (Quora) and related blog post by Stewart McCoy

2. Presentations

How to Tell Stories with Data (Really) by Edward Segal
PDF version

Visualising Workflow: Findings Stories and Telling Stories by Andy Kirk

Storytelling with data visualization: Questions and challenges by Albert Cairo

Storytelling with Data by Jonathan Corum

3. Research Papers

Visualization Rhetoric: Framing Effects in Narrative Visualization by Nick Diakopoulos (Summary, Research Paper)
“We carefully analyzed 51 narrative visualizations and constructed a taxonomy of rhetorical techniques we found being used. We observed rhetorical techniques being employed at four different editorial layers of a visualization: data, visual representation, annotations, and interactivity. The five main classes of rhetoric we found being used include: information access (e.g. how data is omitted or aggregated), provenance (e.g. how data sources are explained and how uncertainty is shown), mapping (e.g. the use of visual metaphor), linguistic techniques (e.g. irony or apostrophe), and procedural rhetoric (e.g. how default views anchor interpretation).”

Narrative Visualization: Telling Stories with Data by E. Segel and J. Heer
(Abstract, Research Paper)
“We systematically review the design space of this emerging class of visualizations. Drawing on case studies from news media to visualization research, we identify distinct genres of narrative visualization. We characterize these design differences, together with interactivity and messaging, in terms of the balance between the narrative flow intended by the author (imposed by graphical elements and the interface) and story discovery on the part of the reader (often through interactive exploration).”

Storytelling: The Next Step for Visualization by Robert Kosara and Jack Mackinlay
“Presentation and communication of data have so far played a minor role in visualization research, with most work focused on exploration and analysis. We propose that presentation, in particular using elements from storytelling, is the next logical step and should be a research focus of at least equal importance as each of the other two.”

What Storytelling Can Do for Information Visualization (PDF) by Nahum Gershon and Ward Page
“Effective presentations using the storytelling approach require skills like those familiar to movie directors, beyond a technical expert’s knowledge of computer engineering and science. Creating a presentation is not just a matter of being literate in visual media and storytelling but depends on a frame of mind that caters to other modes of human information processing and thinking.”

The Enchanted Imagination: Storytelling’s Power to Entrance Listeners
“While storytelling has flourished, there has not been a concomitant surge in research of the art form. One element of storytelling has remained nearly unconsidered, and it is, perhaps, the most profound and influential characteristic of storytelling: its power to entrance those who listen.”

4. Tools, Examples, and Other Resources

Hans Rosling’s TED Talks
“What sets Rosling apart isn’t just his apt observations of broad social and economic trends, but the stunning way he presents them. Guaranteed: You’ve never seen data presented like this. By any logic, a presentation that tracks global health and poverty trends should be, in a word: boring. But in Rosling’s hands, data sings. Trends come to life. And the big picture — usually hazy at best — snaps into sharp focus.”

Robert McKee, Godfather of Storytelling (Wikipedia)
Rather than simply handling “mechanical” aspects of fiction technique such as plot or dialogue taken individually, McKee examines the narrative structure of a work and what makes the story compelling or not. This could work equally as well as an analysis of any other genre or form of narrative, whether in screenplay or any other form, and could also encompass nonfiction works as long as they attempt to “tell a story”.

Stories Through Data
Exploring storytelling in data visualization. A collection of visualizations sorted by Chris Twigg’s narrative analysis framework.

13pt Information Graphics
Gallery of examples from the studio of Jonathan Corum, an information designer and science graphics editor at The New York Times.

A free and collaborative taxonomy of Data Storytelling tools by Philippe Nieuwbourg
“To summarize my investigations around data storytelling tools I created a mind map. This map will be an up-to-date taxonomy / ontology / typology, of software available on the market, to create stories around data.”

DataStorytelling.tv
“An independent website, dedicated to storytelling around data.”

Pixar’s 22 Rules of Storytelling
“Give your characters opinions. Passive/malleable might seem likable to you as you write, but it’s poison to the audience.”

Bob Beamon’s Long Olympic Shadow by Kevin Quealy and Graham Roberts (NYT)

I don’t like to knock our own stuff, but a little healthy introspection is always a good thing. Consider our popular ChartChooser tool. In spite of it’s carefully crafted name (it was core of an ad campaign akin to peanut butter: Choosy chart choosers choose ChartChooser — no, not really), we’ve come to believe that ChartChooser isn’t so useful for the “Chooser” part; it is useful because the “Chart” part is nicely formatted, downloadable PowerPoint and Excel charts.

Here are the filtering choices for ChartChooser:

I’ve been at this a while and I still don’t always know how to connect what I’m trying to express with words as vague and broad as ‘Composition’ or ‘Relationship’.

It isn’t entirely ChartChooser’s fault. Basic chart types are by nature broad and flexible in their usage. How can we make it easier for someone to make that leap from their question to a visualization that best answers it?

We believe one part of the solution is to make visualizations more purposeful. That is, create re-usable ways of expressing data that are carefully designed to answer common questions that people pose about their data.

While it’s true that everyone’s data is unique, what we’ve learned is that in most cases, the things they want to know about their data aren’t so unique. The same sets of question patterns show up time after time. It’s almost like a game of Mad Libs:

• Which are my top performing _plural noun_?
• Which _plural noun_ are the most significant outliers when measured by _ measure_ and _ measure_?
• Which _plural noun_ have improved or declined the most over the last _time period_?
• How does _singular noun_ compare to _singular noun_ across my important performance measures?

Our goal is to draw straight, obvious lines between questions like these and a visualization that directly and simply expresses an answer.

If you consider the last data Mad Lib question above, our match-up visualization is a good example: compare two things side by side to see relative performance. The Match-up was inspired by the traditional tale-of-the-tape graphics that you used to see in boxing matches.

Like a lot of our visualizations in Slice, we’ve added a number of key features that really help the user quickly understand and explore the data. Here are a couple examples:

We’ve put together a whole collection of these purposeful visualizations, such as a funnel visualization for sales conversions and other processes; a leaderboard for ranking top items across a bunch of measures (try it free here), and a comments visualization for reviewing and exploring survey verbatims, tweets, and other descriptions. And we’ll be making more. What questions do you ask of your data?

As we started our journey to pay tribute to the beautiful game, we came to realize much like the recent Sloan Sports Analytics Conference that fútbol is just starting to get its data on. Check out this view of team performance created in Slice.

This data comes from the MLS Soccer Team Stats page.

Since we’re bound for disappointment, why not have some fun with data. Our resident Iona grad, Michel, put together this slick Sliceboard that ranks the tournament teams by offensive and defensive stats. Notice how his school pours in 81 points a game as the 2nd most prolific offense in the tournament. They also manage to have the 2nd worst defense. Sorry Michel, defense wins Championships.

We’re not sure whether everyone came for the spread of pies or data visualization discussions, but either way the Juice 2013 Pi Day at our Atlanta office was a good time for all. As you can see, we decided to kick off the small talk visually.

With a wide variety of backgrounds, people could find a discussion area of their choice. Here are some topics we covered.

Data Storytelling

Data Storytelling is rising up the charts of trendy data visualization topics. I shared our thinking about what the term means to us (something we’d started with this blog post). Think of your role telling data stories like you are a safari guide. You can lead your audience around to see the most interesting sights in the park, but be willing to go off the planned course as the interests of your audience dictate. We pulled some lessons from Pixar’s rules for storytelling, discussed the importance of influencing both your primary audience as well as their audience, and connected these elements of data storytelling to how we designed Slice.

Meanwhile, next door…

Information Design Trends

We had an informal open discussion on trends we’ve seen and what they mean going forward. What are the challenges of responsive design to visualization; particularly as it relates to reporting vs. exploration. What does the rise of lifestyle data mean for the future, and for your privacy? There were generational differences in comfort with tracking with our younger attendees preferring to turn off and drop out. Finally, who doesn’t want to talk about “BIG data”? Is Hadoop for you? What will we do with the sextillion bytes of data (one billion terabytes) humanity will generate in 2013? How will this mountain of data get transformed into something people can act on and relate to?

In between, people cast their votes on these questions. What’s your answer?

Here are some notes and resources we jotted down. Feel free to add to them: http://bit.ly/JuicePiDayTrends

Thanks to all who came, and, if you couldn’t make it, join us next year!

In hopes of sparking some new thinking, here are a half-dozen dashboards that demonstrate innovative designs and features.

1. ThinkUp: Dashboard as News Feed

This clever social media dashboard focuses on the changes and news-worth data updates. Each timeline element that flows down the page can be expanded to show details that supports the headlines.

2. SumAll: Trends, annotations, and goals

SumAll’s recently released tool for trended business data is beautiful in its execution. The interface puts multiple metrics on the same chart to make it easy to see how trends correlate to each other (even though they are on different scale). SumAll does a nice job with allowing user-created comments, setting goals, and anticipating how users will want to read the data. If you design dashboards, it is worth signing up to absorb some of the nice design touches.

3. AppFirst DevOps Dashboard: Key metric trends and thresholds

This dashboard is a more traditional real-time operations dashboard. It shows all the key metrics together along with trends relative to goals.  AppFirst has some nice features that you can see in the video including 1) dropping metrics one on top of another to see correlated trends and, 2) “smart thresholds” to highlight points in the historical trends when the metric fell out of an acceptable range.

4. Analysis-One: Creative layout for easy comprehension

Many stoplight-style dashboards can be jarring to look at. This radial approach provides the high-level performance warnings in a subtle and pleasing way.

5. Square: Filtering

Mike Bostock of D3.js fame is the brains behind this dashboard for Square’s retail sales. He used his Crossfilter approach for super simple and fluid filtering.

6. Tweetping and Tron, the Movie show us dashboards of the future.

If you want to build a dashboard with a style that will still work 20 years from now, check out the dark backgrounds, tiny fonts, and animation in these dashboards.

Tweetping

Tron Dashboard

If you’re interested in more well-designed dashboards, Quora has a good discussion on the topic here.