Data visualization

In a paper I co-authored for EPIC 2007, I wrote about a pie chart that I used to depict the research findings from my first ethnographic research study inside SAP in 2005.  With the exception of my dissertation research, which I had ‘distilled’ into a 300-page book, I had no experience in presenting complex, qualitative research to an executive audience in the corporate setting.  Like a typical consulting engagement, I had about 80-100 hours worth of findings, but less than an hour to make a compelling statement about what I had found.  I used the pie chart as a door-opener for an executive conversation, and got myself a seat at the table with a larger management team to discuss my findings in more detail.

I was a little pained by the visualization, because as an anthropologist I knew that the pie chart grossly oversimplified what I had found.  However, it made the major themes of my research accessible to a broader audience who, frankly, might have not listened to me at all if I hadn’t taken the approach I did.  I still grumble and joke about that pie chart, because it is in wide circulation inside of SAP to this day.  In the most unfortunate circumstances, it gets used to build a business case with people really not understanding the nuances.  In the best cases I still get asked to provide readouts on what I learned.

The pie chart wasn’t my first brush with visualization, though.  I studied Whole Systems Design for my Masters degree, and I started my PhD at Temple in the Anthropology of Visual Communication track.  But even earlier than that, I remember reading The Synopsis of the Four Gospels, which lays out Matthew, Mark, Luke, and John’s versions of story of Jesus side-by-side so that they can be compared in detail.  As a scholar of religious texts, it’s absolutely fascinating to see the common elements, the gaps and variations between one version and the next, and so on.  The other tool that had (and continues to have) a great value for me is Quicken.  I first started using it to manage my finances in 1996, and the pie chart called Insights (depicting total spend by major budget category) was instrumental in causing me to change my spending patterns.

In the past year or two, I’ve grown increasingly interested in how to represent user research findings in a visual way.  It has paralleled my involvement in new social media, as well as my increased access to SAP executives.  Below I’ll explore why there seems to be a growing interest in visualization in business today, and how the opportunities and benefits of User Experience, Social Media, and data visualization are interrelated.

Data visualization & social media

In addition to a User Experience team, I run a small team called Emerging Solutions, which is focused on identifying new technology trends and introducing them inside of SAP, demonstrating their business value through proofs-of-concept, and eventually pilots.  Jen Robinson runs this function for me, and because we work in a knowledge management organization, we felt it made sense to explore new social media (wiki, blogs, micro-blogging) first, as they were the emergent capabilities most aligned with our organization’s charter.  While Jen was teaching me about Facebook and Twitter, she also insisted that I watch a TED talk by Jonathan Harris, about his project We Feel Fine. Harris and a colleague work together to scrape content from publicly available blogs, and provide a compelling visual display of the content related to feelings.  That content has also been analyzed alongside weather and census data and other factors to look for interesting patterns.  For me the talk raised a whole new level of awareness about what is happening in the blogosphere, and furthermore, blew my mind about what could be done with information that is now publicly available on the Internet.

More recently, my growing interest in social media and social tagging had me thinking about tag clouds and other ways to show the compelling elements of largely text-based content in new ways.  I uploaded my bookmarks from my browser into http://del.cio.us to see the patterns in my bookmarks.  I won’t share the details of that here except to say one word – shopping!  During that period I had just started this blog, and (again, the tie back to knowledge management), I was curious what a tag cloud would show about what I was writing about.  I later came across a tool called Tag Crowd, which created a visualization of the most frequently used words on a website or in the text of a document.  Wordle took it to the next level by providing alternate fonts, colors, and layouts – so it’s lots of fun to play with!  I run it against my blog today, it looks like this:

wordle4

I find it valuable – and at times generative – to look at data from a new point of view.  For example, I’m glad to know that the word work appears so prominently (indicating I use it a lot) since I consider myself to be an anthropologist of work. But I didn’t know that I used the word really so much.  It’s bad enough that it’s a speech tic – but in my writing too?  Bummer.

Why now?

More recently, I have really come to appreciate some of the interactive visuals that were created by the New York Times.  This one represents people’s moods after voting in the presidential election, and it seems to take it’s inspiration from Harris’ project.  I think this Transcript Analyzer from the Presidential Debates is fantastic as well.  By rolling my mouse over a grey transcript snippet in the fourth column, I am able to view an excerpt of what Obama had to say on the topic being discussed in that moment:

debate

I would just LOVE to have this for each of the user research studies my User Experience team conducts!  It allows the viewer to read, view, and interact with the content of the debates.  This particular visualization reminds me of the Synopsis of the Four Gospels that I mentioned earlier.  One of the things I find particularly interesting about what the New York Times is doing is that they are leveraging the capabilities of IBM Many Eyes to enable their readers to interact with and create visuals from the Times data.  You can read more about that effort here.

To answer the ‘why now’ question, I think that there was a huge public interest and focus on the elections.  Visualizations like the ones above makes the information more accessible to a new / broader audience, and furthermore, new computing capabilities and publicly available tools like Many Eyes help to make it possible to interact with the content in new ways.

In one of Business Week’s Eye on Design columns in August, author Maria Popova wrote about the growing interest in data visualization, and why she believes it’s happening.  Consistent with what I said earlier about the relationship to User Experience research, she says:

Data visualization isn’t about using all the data available, but about deciding which patterns and elements to focus on, building a narrative, and telling the story of the raw data in a different, compelling way. Ultimately, data visualization is more than complex software or the prettying up of spreadsheets. It’s not innovation for the sake of innovation. It’s about the most ancient of social rituals: storytelling. It’s about telling the story locked in the data differently, more engagingly, in a way that draws us in, makes our eyes open a little wider and our jaw drop ever so slightly. And as we process it, it can sometimes change our perspective altogether.

I would argue that visualization allows people to be self-reflexive, to understand themselves in relation to a broader world (represented by complex datasets).  I think this is one of the most critical and compelling parts of engaging in new social media – seeing your communities and your networks, and having visibility into what is happening within them.  At the same time, visualization allows both laypeople and executives to look at complex problems in new ways, and to interact with content that might otherwise be prohibitive to engage with.  That’s a big reason that I find it to be such a powerful tool for User Experience.

In a subsequent post I hope to write more about what we’re doing in this space, though I may be a bit challenged to do that while respecting the confidentiality of the work I’m doing inside of SAP.

One Comment on “Data visualization

  1. Pingback: 2010 in review « Natalie Hanson, PhD

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: