Related Links

Web Seminars

We hear it many times after disaster strikes: the warning signs were there, but the decision-maker didn't see them. Or, sophisticated technology had been installed, but no one had properly learned to use it. Such failings are often chalked up to "user error" or "problems between keyboard and chair."

In catastrophic failures that have involved analytics, we've come to see that sophisticated models won't do us much good unless decision-makers are able to interpret, understand and act on the results appropriately.

Bad outcomes related to analytic models often arise because designers have not considered how people interact with the information being delivered. Designers might be too enamored with the power of their model, its options and parameters. They may have forgotten that users are pressed for time and need to focus on daily duties.

As a result, users might not be able to answer basic questions such as, "What information should I pay attention to?" and, "Now that I've seen this information, what should I do?"

If users can't easily answer these questions, the interface and the analytics are plainly not meeting their needs. A more user-centric analytics design approach is needed.

What Users Need

Good analytic interfaces facilitate what the military calls situational awareness: a grasp of events relevant to a given situation, how those events relate to each other and an ability to project the effects those events will have on the situation. A situationally aware pilot uses the information presented in the cockpit to determine what course of action might be needed to respond to a situation.

The analytics interface is the decision-maker's cockpit. Well-designed "analytics cockpits" are built with characteristics that promote situational awareness, including:

• Role-based design. A telecommunications CEO makes decisions unlike decisions made in the company's network operations center, so role-based interfaces to their analytics need to support different types of decisions. Where interfaces are tailored to roles, users won't be asking, "What does that thing do?" or "Why do I care?"
• Less is more. Good analytics interfaces show the information most critical to the user - not every piece of information that might be available for analysis.
• Sensory cues direct attention. Good interfaces exploit people's abilities to perceive patterns based on position, size, shape, color and movement. These properties highlight important features that might otherwise be lost in a table of numbers.
• Interfaces suggest actions. Analytic dashboards alert users to potential performance issues and provide actionable information. Good interfaces provide context to interpret results that suggest what the user might do next and provide mechanisms such as clickthrough to facilitate an explanation and further analysis.

Designing User-Centered Analytics

Just as cockpits could not be designed without understanding what pilots need in order to fly an airplane, analytic interfaces should be driven by an understanding of what users will do with the results. Here are some principles for obtaining that understanding and designing interfaces accordingly.

Let the Users Lead

User-centric analytics follows the approach of other user-centric designs: start from user needs and work backward to design the interface that supports those needs, ultimately to the analytics that will drive that interface. Even when users cannot specify in advance what they really want, it is critical to involve them early and often as analytic interfaces are designed. Users are likely to feel about interfaces the same way Supreme Court Justice Potter Stewart described obscenity – they can't define it, but they know it when they see it. Users are even better gauges for bad interfaces – if enough users believe an interface is unsatisfactory, the designer is well-advised to accept their judgment.

Beyond consulting users, analytics designers should consider user preferences even in choosing basic analytic approaches. Users who are comfortable adopting an analytic approach are likely to be confident that they can explain and defend the analytics (that it is not a black box). Simple models have a greater chance of adoption than complex ones, and linear models have a greater chance of adoption than nonlinear ones, even at the sacrifice of accuracy.

If you are contemplating giving users the ability to set analytics modeling parameters, determine if they want to set those parameters and that they know how to do so (or at least give them default values).

Other benefits arise when users are involved in the design of analytic interfaces. They can help identify early wins the designer may not have thought of and might provide useful introductions to other potential users and their communities. A user who feels a sense of ownership in interface design can become an advocate for the technology respected by other users. Users of different abilities may point out accessibility considerations, such as how and when color is used so color-blind users get the same information from the intensity of the display.

It is always best to avoid relying on a single user for design. Vet judgments and suggestions with several other users to be sure that the input is representative of the intended user population.

A Picture is Worth a Thousand Numbers

The right visual display can make it much easier to understand the results of complex analytics and increase user adoption. Because of our human ability to understand relationships quickly based on size, position and other spatial attributes, the eye can summarize what might otherwise require thousands of numbers to convey.

As an example, Figure 1 represents an analytics interface at a large consumer products company. It shows the effectiveness of trade promotion investments in distributors of products offered by the company. Each dot represents a distributor, with the horizontal axis showing the amount of the investment in rebates offered through that distributor and the vertical axis showing the profit or loss from that investment. The vast majority of investments are small, with correspondingly small profits or losses.