A Loving Intervention for SoftwareRick Wartzman
About 15 years ago, Peter Drucker’s grandson gave him a little computer software demonstration. An architectural firm, where the young man worked as an intern, was using the program to design the heating, lighting, and plumbing systems for a new building. Grandpa was dazzled. Sort of.
“The software … can, literally in the twinkling of an eye, do the work that once took hundreds of individuals to complete,” Drucker noted.
Yet he also saw in his grandson’s display a shortcoming of the Information Revolution: While computers were having a huge bearing on the way things were done, they were having little, if any, influence in determining what to do in the first place. For the typical executive, “new information has had little impact on how he or she makes decisions,” Drucker asserted. “That is going to have to change.”
In fact, I believe it is, albeit slowly. This shift recently became vivid for me when I learned about a company called iRise. It makes computer-based tools to “test-drive” software applications in early development, before businesses have sunk too much time, money, or ego into projects. In so doing, software prototyping firms such as iRise are helping to solve a huge problem: the inherent inability of IT people and others in the organization to communicate clearly with one another.
“The producers of the data cannot possibly know what data the users need,” Drucker wrote in Management Challenges for the 21st Century. “Only individual knowledge workers, and especially individual executives, can convert data into information. And only individual knowledge workers, and especially individual executives, can decide how to organize their information so that it becomes their key to effective action.”
Traditionally, companies have handled this divide through a set of software requirements meant to translate a particular business need into computer code. These specifications are usually captured over many months in a War and Peace-sized document, following endless rounds of mind-numbing meetings between IT staff and those who will actually use whatever is being produced—an online payment system for customers, say, or a new inventory-management regime.
The trouble is, what’s ultimately built rarely lives up to what the user had imagined. “You’d get to the user acceptance testing—the last phase—and you’d have these aha moments: ‘This isn’t what I need.’ ‘This isn’t what I expected,’ ” says Rob Carter, the chief information officer at FedEx. At that point, it’s not uncommon for an organization to overhaul the entire project, pushing everything behind schedule and over budget.
Since bringing in iRise five years ago, however, those disconnects at FedEx have basically disappeared, according to Carter.
That’s because, with visualization and simulation technology, IT personnel can quickly provide an accurate, highly interactive rendering of how a software application is going to look, feel, and flow. This goes way beyond the static screen shots to which many are accustomed. Adjustments can be made immediately, before coding even begins. Back-end surprises are eliminated, saving time and dough.
“It enables us to show rather than to tell,” says Mike Capone, the CIO at Automatic Data Processing, which first brought in iRise a couple of years ago.
Why Drucker Would Approve
Drucker, I think, would have loved this formula for three reasons (in addition to all the efficiencies it brings). First, co-designing with the end users takes all the mystery and misunderstanding out of the central question: What does the customer value?
Second, involving the users more directly in the design means they wind up with a cleaner and simpler experience. The final product is “just more elegant,” says Carter of FedEx, which is utilizing iRise to help create a variety of software applications, including quality-reporting systems for the executive suite and the shipping and package-tracking services that its customers employ.
Such ease of use is important. Consumers living in an iPhone world are demanding this more and more. For those inside the company, meanwhile, simplicity is essential if they want their measurement and feedback systems to work. “Complicated controls … confuse,” Drucker wrote in Management: Tasks, Responsibilities, Practices. “They misdirect from what is to be controlled toward attention on the mechanics and methodology” of the control itself.
The third great thing about this approach is that it stimulates people’s thinking and, in this way, is moving business software in exactly the direction that Drucker wanted: from merely running operations to helping hone strategy and spark innovation.
ADP’s Capone, for instance, says that visualization has been instrumental in bringing to life his company’s next generation of products, which are no longer divided by function (payroll services, human resources, and so on); instead, the products now focus on tasks and activities that often blur departmental lines. “I don’t know if we could have done that if we hadn’t been able to show it,” Capone says.
IRise, which brings in more than $40 million in annual revenue, faces a number of competitors, including Axure Software Solutions, Balsamiq Studios, and Justinmind. But the biggest thing iRise is up against, according to its co-founder and chief executive officer, Emmet B. Keeffe III, is the mindset of executives wedded to the old, cumbersome way of designing software.
“Ninety-nine percent of the time we’re competing against a 30-year-old process set in concrete,” Keeffe says. “We’re trying to convince the CIO to start over, and there’s just a lot of resistance to that.”
Drucker, who decried “organizational inertia” while recognizing its incredible power, would appreciate Keeffe’s frustration. It’s too bad you can’t change human behavior with the click of a mouse.