Harvard Business School professor Kristina Steffenson McElheran studies the effect of information technology on business process innovation. It's a topic, she is sometimes told, that is, well, less than exciting.
"I've had people say to me that studying IT use is like studying plumbing—every firm has it, so why is it interesting?" she says. "I think it's pretty sexy plumbing."
McElheran believes IT has the potential to completely transform the supply side of business by flattening hierarchies, shrinking supply chains, and speeding communications. The result: "People can spend more time thinking up new products and servicing customers, and less time checking boxes."
To get there, most firms must be willing to engage in radical change and completely rethink how they collaborate and compete. Few companies, especially larger ones, have done so. Those that have done the heavy IT and organizational lifting, such as Walmart, reap serious dividends.
"There's a tremendous gap between the most IT-savvy firms and the IT laggards," says McElheran, the Lumry Family Assistant Professor and a member of the Technology and Operations Management Unit at HBS. "The longer we take to figure this out, the more likely we're going to have a few of the best firms pulling way ahead based on this competitive advantage."
More likely to go radical
In her research, McElheran explored the characteristics of firms that engage in radical business process innovation versus incremental change. It's widely understood that when it comes to product innovation, market-leading firms are often more likely than laggards to pursue product improvements in a very incremental way—a practice that protects short-term competitive advantage but that makes firms vulnerable to innovative upstarts over the long haul.
Would this trend also hold true for companies engaging in business processs innovation? McElheran wondered—a question not much studied by researchers.
To find out, she drew on 1999 US Census survey data of manufacturing firms: approximately 35,000 plants in 86 different manufacturing industries. The sheer amount of data allowed her to test for incremental versus radical innovation among firms in a unique way. For incremental change, McElheran looked at a firm's "e-buying" activities to serve as a proxy. For radical change, "e-selling" was the stand-in.
E-buying—electronically procuring goods like pencils, staples, and toilet paper—is a simpler process, and less risky to adopt. E-selling entails a much more complex group of products and requires cross-departmental interaction, as well as an important group outside the firm's walls: customers. It is riskier economically and organizationally to adopt.
"The thing that struck me right away was that e-buying and e-selling for the firms in my sample were fundamentally different activities," McElheran says. She was also struck by how the reality of e-business innovation portrayed in the census data differed from what she had read in the press. "It was clear that firms were actually doing far less of it than you would think if you read the newspaper."
The e-business that was being done among survey participants matched the original theory: larger firms, market leaders, were more likely to adopt the incremental change of e-buying, while smaller firms were more likely to adopt the radical change of e-selling.
The reason, however, was surprising. E-selling is harder to adopt because it's a complicated process, which is naturally made more complex when it has to be implemented across a large, distributed organization.
"Large firms definitely had some work to do to adopt," says McElheran. "But it seems that the best firms had practices in place to deal with the complexity. The real stumbling block was that their customers had to meet them in the process. Even if a given manufacturer might be ready, if its customers weren't, the [firm] couldn't make the jump."
Full results of the research are described in her working paper Do Market Leaders Lead in Business Process Innovation? The Case(s) of E-Business Adoption.Missing ingredients
So what should firms do to upgrade their IT-based business processes? Unfortunately, there is no easy answer.
"It's not just about writing a check to an IT vendor," McElheran explains. "There's a complicated assembly of processes and people and organizational structures and supply chain partners that really have to come together."
The first step, she says, is "treating it as an innovative process. Understand that there's uncertainty and put processes in place to manage it." When firms invest in a new product, for example, they usually have a strategy in place in case it bombs. "The same should go for process innovation," says McElheran.Other ingredients:
Organizational understanding. A deep understanding of how the company functions and competes is essential, because IT needs to support essential firm processes, rather than vice versa.
Executive backing. Support at the very highest levels of the organization is crucial. "Don't treat it like plumbing," says McElheran. "I don't know a CEO who is deeply concerned about the firm's sewer system."
Long-term horizon. Transforming business processes takes time, so companies need to be willing to live in the unknown rather than going for the quick fix. "When you talk to people on the front lines, they know this is complicated and hard," she says.Communication breakdown
Investing in technology change without organizational change won't work either. For the Harvard Business Review article The Short Life of Online Sales Leads, McElheran and coauthors James B. Oldroyd and David Elkington measured how long 2,241 small- to medium-sized US companies took to respond to an online sales lead.
The majority—37 percent—did so within 1 hour; 16 percent responded between 1 and 24 hours; and 24 percent took more than 24 hours. Shockingly, 23 percent never responded at all.
A companion study shows that companies that respond to queries from potential customers within an hour are 7 times more likely to connect with a key decision maker than those that wait more than an hour, and 60 times more likely than those that wait more than 24 hours.
"It seems absolutely crazy," McElheran says. "These are the firms that already know they want to be fast. They've already invested in IT to make them faster. Why do firms that have the IT capability on paper not have the execution?"
Further research revealed what may be missing. "It has to do with the complementary organizational practices with which the IT has been deployed," she says. Deep hierarchical structures with long lines of communication muddy the signal.
"We talked to this manager whose organization is completely flat," McElheran explains. "He is on top of everything in that organization. He knows all the people. He oversees their training. He looks at the software every day. He runs experiments. He customized all the reports to tell him exactly what he needs to know for his particular business…and they are lightning-fast."
It would of course be impossible for one manager to do all that in a larger organization, and McElheran can't say for sure which part matters most. "Is it the flatness of the organization? Is it a process where the IT output is evaluated every day? Is it because the company invested heavily in customizing the IT?" Future research is aimed at answering these questions.
The competitive advantage
Despite the hard work involved, there are companies that get it right. "Walmart has done tremendously well," McElheran says. "One of the reasons Walmart has been able to compete so well on prices is that tremendous cost has been taken out of its supply chain and operations by using IT to keep inventory low but still not stock out."
That Walmart is secretive about how it manages to so finely coordinate its enormous global supply network is telling. "They're keeping it really hidden because it's a source of competitive advantage for the company."
McElheran looks forward to further exploration.
"I'm still at the very beginning of a very long and interesting journey," she says. "It'd be sad if I had all the answers right now."
[This article was provided with permission from Harvard Business School Working Knowledge.]