The Expanding Scope of Data Visualization
Increasingly, the term "data visualization" describes how we expect to consume information.
- By Stephen Swoyer
- September 3, 2013
A decade ago, the term "data visualization" was used (sometimes dogmatically) to describe the technologies, concepts, and methods used to enable or to optimize the display or representation of information. These days, it has a less specific, less technical, and less doctrinaire meaning. Increasingly, "data visualization" describes how we expect to consume information: i.e., in the form of interactive visual (as distinct to static textual) representations.
A new report from TDWI Research, Data Visualization and Discovery for Better Business Decisions, explores this change, highlighting how "data visualization" has come to describe a large category.
Now as ever, the term "data visualization" describes a sophisticated class of best-of-breed technology solutions. Visualization has, to some extent, become commoditized over the last half-decade or more, however: most vendors now deliver a visualization capability with their offerings, be it in the form of canned analytic visualizations; integrated data visualization features; or a dedicated visual business intelligence (BI) discovery tool.
One common use of data visualization is in so-called "display" or "snapshot" reports, writes David Stodder, director of TDWI Research for business intelligence. This describes the updating -- or translation -- of textual artifacts into a visual context.
"Many organizations are implementing dashboards to display basic reports, including on mobile platforms. Snapshot reports are typically scheduled rather than requested on demand, although some users create snapshots manually. The results are often stored for users in a cache or database as a 'snapshot' of a certain point in time," he writes. "Because users examine snapshots to identify changes in data over time, they must be provisioned and presented consistently so that the trends and comparisons drawn are valid."
Organizations are likewise translating existing visual artifacts (such as scorecards) to the dashboard context. They're translating related features -- for example, drill-down -- to a dashboard context, too, at least when it makes sense to do so. Depending on the maturity of their dashboard implementations, organizations might also expose more sophisticated interactions, such as root cause analysis, to users. According to Stodder's research, this kind of basic "snapshot" reporting is the most common way in which visualization technology is used: almost three-fifths (57 percent) of respondents to a recent TDWI survey said that they're currently implementing snapshot reports; nearly a third (31 percent) are planning implementations. Only 9 percent don't plan to do so.
Another use of data visualization technology is for so-called "operational alerting." This, too, comprises a kind of translation of an established BI concept -- viz., "alerting" -- to an explicitly visual context. BI tools have included alerting features for at least the last 15 years. Nor is the visual representation of alerts -- be it in the context of a dashboard, in an operational application, or in another, dedicated view -- anything new. "Color is often used for alerts on dashboards. Some tools enable users to perform visual analysis to discover why an alert condition exists, such as if sales have decreased for a certain product line," Stodder writes. This isn't what Stodder and TDWI mean by operational alerting.
To be sure, BI tools have long included alerting capabilities; the emerging field of operational intelligence (OI) goes BI alerting one better, however. OI tools monitor and analyze event information in real time or "right time." They look for anomalies or exceptions in order to alert users to incipient issues before they become full-blown (and invariably costly) problems.
"These systems focus on monitoring activities in business, distribution chains, manufacturing, networks, IT systems, and more for problems, threats, and other critical developments," Stodder writes. "Alerts signify that monitors have uncovered something important in (often) real-time data or event streams; some systems provide analytics for determining root causes and the best way to address situations." According to TDWI Research, just over one-quarter (26 percent) of survey respondents use data visualization to support operational alerting. Nearly two-fifths (39 percent) say that they plan on doing so. A sizable 27 percent say that they have no plans to do so. "This could reflect the current immaturity of the emerging operational intelligence field," Stodder points out.
Data Visualization for BI Discovery and Analysis
Data visualization first came to the fore as a force for visual discovery and analysis, championed by best-of-breed vendors such as Advizor Solutions Inc., SAS Institute Inc., the former Spotfire Inc. (now a Tibco company), and Tableau Software Corp. (Bis2 Inc. founder Andrew Cardno has also been a trenchant, eloquent, and iconoclastic proponent of a non-traditional take on data visualization, which he dubs "supergraphics.")
On top of this, almost all big BI vendors now market visual BI discovery tools: in just the last two years, IBM Cognos, Information Builders, SAP BusinessObjects, Microsoft, MicroStrategy, and SAS introduced dedicated BI discovery tools. (SAS is a long-time data visualization powerhouse, thanks to its JMP offering.) When knowledgeable people talk about "data visualization," they mean either best-of-breed or BI discovery tools.
According to TDWI Research, one-third of survey respondents say they're using data visualization to support discovery and analysis. Just under half (45 percent) say that they plan to do so. Visual discovery has been a mainstream force for (at most) the last half-decade; prior to that, it was a still-coalescing phenomenon, relying principally on word of mouth from stymied or exasperated business users. "The term 'visual data discovery' is essentially synonymous with 'visual analytics,'" Stodder writes, noting that "it applies to tools and practices that make it easier for nontechnical business users to interact with data. The tools enable users to engage in self-service data analysis through visual representations rather than the tabular results delivered by standard BI queries."
Increasingly, visual discovery tools implement what's called a "guided" BI discovery feature. This permits them to be used by less-sophisticated users, Stodder explains. "[R]ather than give users a complete blank slate, most visual data discovery tools guide users in selecting the right visualizations or even automate the selection," he writes. "Some tools include predictive modeling capabilities to direct users to examine what is most important going forward. Predictive modeling complements visual discovery, especially when there are large data sets to examine with many dimensions and variables."
You can download the complete, 36-page report here. A short registration is required for readers downloading TDWI research reports for the first time.