(Illustration by Adam McCauley)
Improving practices in any complex organization is a challenge. It can be especially difficult when an organization has a rigid hierarchical structure. The opinions of lower-status employees are often disregarded, and staff members may generally be afraid to make mistakes as they put new policies into practice.
Melissa A. Valentine, however, has found ways in which hierarchy can be utilized for organizational change. A management science and engineering professor at Stanford University with a background in healthcare issues, Valentine researched how a cancer center attached to a university hospital both succeeded and failed at improving organizational practices. She spent several months observing parallel efforts to change routines and improve patientcentered care, collecting more than 800 pages of field notes, transcripts, and archival materials to develop a new theory on organizational learning. While she found that within groups, rigid hierarchy could impede learning, it could also be useful in promoting learning across groups.
Navigating the maze of cancer care provided an ideal test case for Valentine. It requires relentless advocacy on the part of patients and their families, no matter how well-heeled or educated they are. Keeping on top of the information flow and communication among the different diagnostic and treatment departments is a monster task.
The cancer center that Valentine studied functioned around two different hierarchies, one medical and one operational. The physicians were employed by the university medical school rather than the hospital, while other employees reported up to a senior hospital administrator. Each clinical area was headed by a department manager, and there were many administrative groups, such as business services, insurance authorization, and IT. Accomplishing any of the stated goals would require significant interplay and coordination among the various groups, perhaps a readjustment of roles, and even an expansion of resources.
“If learning is as local, social, and contextual as one view has it,” Valentine says, “my question was, how can you synchronize it across an entire organization?”
In the field experiment, a team of quality-improvement personnel acted as change agents to achieve two different learning initiatives. The goal of the first, the “Admin” initiative, was to help more patients get appointments with the doctors and get them sooner. The goal of the second, the “Navigator” initiative, was to streamline administrative work to better serve patients and proactively help them coordinate the logistics of their care.
Valentine noted that although the initiatives were launched with similar fanfare, their methodologies quickly diverged. While the Admin initiative succeeded in effecting observable changes in the work practices of several operational groups and in their interdependencies, the Navigator initiative ended with little evidence of change.
What made the difference? In the successful Admin initiative, relevant managers were directly involved in renegotiating production and performance expectations for their groups and were asked by the consultants formally to commit to those expectations. In one example, the goal of getting 80 percent of patients scheduled for appointments within one day of referral was surpassed by 10 percent in just four months. By contrast, in the Navigator initiative, the consultants asked managers and doctors for ideas and feedback on overarching values but, notably, did not engage them in any backand- forth discussion or renegotiation, or ask them to make formal decisions or commitments. Even though new work practices were designed, they were not implemented.
The Admin initiative’s managerial method is illustrated by the team’s success in co-locating the relevant administrators in the same “cell.” Several groups and managers were involved. The facilities managers helped the other managers and the consultants discuss and understand the regulations that had to be adhered to in their plans. Each group understood what its and the other groups’ obligations were in support of the new goal, and why it was imperative to work closely together in the same location: When they put their new plan into place, they could continue a dialogue to troubleshoot and smooth out the process. The patterns of interaction seemed to depend on the managers’ having the authority, capacity, and ultimate accountability for solving emergent problems that arose in their groups, Valentine observed.
“The study demonstrates that for organization-wide learning to occur, separate groups across the organization need to synchronize their learning with one another,” says Katherine Kellogg, the David J. McGrath Jr. Professor of Management and Innovation at MIT’s Sloan School of Management. “People in formal management positions can facilitate this global learning … by acting as brokers who synthesize their local group’s interests with the organization’s goals.”
Melissa A. Valentine, “Renegotiating Spheres of Obligation: The Role of Hierarchy in Organizational Learning,” Administrative Science Quarterly, 20, 2017, pp. 1-37.
Read more stories by Marilyn Harris.