People comprehend data better through pictures than by reading numbers in rows and columns. So by visualizing data, you are able to more effectively ask and answer important questions such as "Where are sales growing?" "What is driving growth?" and "What are the characteristics of my customers using different services?". With the ability to quickly answer questions, your data becomes a competitive advantage instead of an underutilized asset.
So what does it take to maximize the effectiveness of Data Visualization:
Make Visualization a business user activity:
Business users should directly explore data without needing to engage IT. This means operating software to acquire and analyze the data.
If people who need answers personally work with the data, they can do more than test hypotheses or confirm their current beliefs via a software developer. The process of exploring data presents a continuous stream of opportunities for unexpected discoveries to literally catch the eyes of the tool operator. Business users better situated than developers to identify those discoveries and capitalize on the knowledge.
Likewise, operating the software presents a different level of personal engagement. For example, if you’re driving your car, you’re generally paying better quality attention, than if you were in the passenger seat.
The other value of business users performing visualization is timeliness of insight. There cannot be many organizations in the world who can afford to have a team of developers who provide any necessary data without no delay. Generally business has to wait for their data. This can run from a few hours to months or years.
Make data available to the business:
Providing data to the business is challenging for a number of reasons: Ad hoc data access can impede operational use of data, delaying or interrupting revenue-producing activity. Operational data is frequently difficult to grasp reliably. Factors such as non-report-friendly schemas, arcane rules for row selection and data organization, unusable naming schemes, and problematic physical data architecture (such as needing data from multiple physical sources) are some factors.
In solving these problems, data warehouses and data marts are traditional, usable solutions. But, they are not always essential to perform visualization. For example, we’ve seen clients gain significant understanding of their business from a limited number of data extracts. If the business were comfortable with the tools of visualization, it would also be possible to provide just data, rather than a finished product for each request.
Choose a powerful, usable tool:
One of the most important tool features is straightforward, intuitive operation. It makes sense that, if business users can’t begin to operate a visualization tool as quickly as they can grasp using Excel, they will be way less likely to use it.
Another valuable feature area is simplicity and clarity of presentation. You want to be able to make a visualization that someone can see for a very short period of time and begin to have some insights. In all my years in the business, I have not seen meaningful questions answered with speedometer widgets, animated gadgets and the like I’ll fully admit that visual designers can do beautiful work that compels attention, clarifies meaning, and has value. Still, if you want to know when you had idle call center reps last week, you probably are not a skilled visual designer. You probably want to get your answer, and then act on it.
This next feature is difficult to characterize feature by feature. The tool needs to support “train of thought”. You want to answer your question, refine your question, move off in a different analytical direction, or report your findings, depending on what you find. Along the way, you don’t want to get distracted from your purpose or from seeing an unexpected result. Obviously intuitive operation, and clarity of presentation are attributes here. It also helps if the product generates results as rapidly as possible. You can keep your train of thought much better with a 2 second delay, than a 30 second delay. Another attention thief is having to be distracted by small details. For example the more properties you have to set for each question answered, the more likely you are to lose your train of thought.
The last important feature is simple, straightforward sharing results. It should be possible for business users to share their findings with coworkers without need for change management, DBA intervention, deployment windows, etc. At the same time, the usual enterprise requirements such as security and backups should be enabled.
When I worked in the Enterprise, I used to roll my eyes about this sort of article. It all seemed so nice, but unattainable. Now, however, high-quality tools do exist to support business users use Data Visualization to gain meaningful, usable insight. If you’re interested in more information, contact us at XCentium.