Data visualization has great promise. People who see a demonstration of a data visualization application immediately see its potential to provide insight into their information. Ideas abound as to what types of analysis data visualization could be used to enhance.

Yet, while many pilot applications have been developed, data visualization has yet to achieve mass deployment. Why? What have been the obstructions to providing instant insight into information? There are several factors that enter into the equation. One is the platform issue. As data visualization first appeared in the early 1990s, it required a sophisticated graphics workstation running a UNIX operating system. Sun or Silicon Graphics workstations were implemented on analysts' desks, but were generally not widely deployed throughout the enterprise. The need for specialized hardware and a UNIX architecture turned away some potential data visualization users. Within the last year, however, visual applications have been developed to run on Windows 95 or NT PCs. While at NationsBank we have been doing data visualization applications for two years, it was only recently I could run them on my laptop. The cost of producing applications has been another deterrent to mass deployment. While some ad hoc landscapes can be used to visualize any multidimensional data, the best results usually come from customizing the visual presentation based on information content and meaning. It usually requires four to eight weeks to develop a visual application unique to the data at hand, which is not significant based on traditional operational applications, but is very significant when compared to the cost of developing Excel charts or graphs, for example.

Another obstacle seems to be that proponents of data visualization have had difficulty quantifying its benefits. Since analytical work is not like automating daily operational procedures, data visualization's benefits tend to be more in the area of revenue generation rather than cost avoidance. That "return on information" is very valuable, but hard to quantify up front, which leads us also to issues with the socialization of data visualization. The early adopters of data visualization technology have been innovators. Data visualization pioneers are people working on difficult, complex, mission-critical problems for their organizations, as opposed to daily processing problems. As such, any successes are usually kept confidential. This makes it difficult to assimilate data visualization into the information technology mainstream.

And within an organization, data visualization requires a new way of working with information--another socialization issue. There is an awareness and education cycle to get the organization prepared to accept data visualization. Not everyone in the organization is immediately prepared to use visual representations of data. It takes time to ready users for the change, have them use the visual applications, get their feedback to make the applications more effective, make enhancements, help users use the applications in their day-to-day work and show the applications to others.

In addition, people need analytical skills; effective data visualization users are multidimensional thinkers. While it seems intuitive for analytical thinkers to assess data by multiple dimensions (such as product, time, geography and profitability), not everyone thinks that way. Spreadsheets have helped people organize their information into two-dimensional matrices of rows and columns and have achieved great success in doing so. Spreadsheets are nearly ubiquitous. When information is organized dimensionally, however, as in OLAP and data visualization, the potential for understanding it grows exponentially. That third dimension of information can often be grasped if people can visualize a cube. But dimensions beyond three can be difficult to comprehend. It can be a significant socialization issue to stimulate large, heterogeneous business audiences to think multidimensionally. The increased use of OLAP tools will help socialize multidimensional thinking within the organization. But until multidimensional analysis skills are pretty much universal, we can expect some obstacles in those business areas where the skills do not already exist.

Just as people need to become proficient in choosing the graphing method that is appropriate to the data involved, they also need to become proficient in selecting the method of visualizing data. Whether it's charting data or visualizing it, knowledge of the data and practice in using charts or visualizations will produce success over time. Experience in the rate of adoption of new technologies in general would tell us that ten years is the average amount of time it takes from initial research until widespread adoption. If that holds true for data visualization, in four to five years we can look forward to a time where the cost of development has come down, where benefits are more widely understood and where the socialization issues are less prominent. As people become more analytically oriented and begin to think multidimensionally, data visualization will become a necessary tool to deal with massive amounts of complex data. With most obstacles out of the way, they will then have unobstructed vision into their information!

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access