This article originally appeared on LinkedIn.

Flipping through the pages of my college alumni magazine sometimes feels like stepping into a bucolic utopia, where everyone collaborates selflessly for the good of ol’ U.

Stepping into real campus meetings, though, is a different experience. Within a university, colleges and departments rarely collaborate effectively. They tend to strive for complete independence from one another and wince every time central administration tries to engage with them. This dynamic leads to the fragmentation, redundancy and misaligned incentives I wrote about in my last blog post.

While several proven solutions such as shared service centers and outsourcing can correct fragmentation and redundancy, it’s very hard to effect change on campuses. Nearly every school I've encountered is plagued by a lack of alignment, trust or accountability.

But these dysfunctions can be overcome, as I've seen firsthand while working with universities on their fiscal and management challenges. Three tools to break through the dysfunctions are clarifying roles, creating a culture of accountability and working with star teams of star players.

Lack of alignment usually stems from a feeling that “you don’t really understand me and my priorities.” Departments fear that the central administration will take over a function and then provide “one size fits no one” service.

At one major research university, for instance, many departments managed their own unique contract with the same learning management system (LMS) vendor. Each unit had an independent software license, a different software version and server to run it, and its own employee to manage the system. The setup was fragmented, redundant and inefficient. As part of a campus change initiative, all the departments agreed to have the central IT office manage a single contract with the vendor. Central IT renegotiated one license, put all units on the same software version and server, and had a single employee manage the whole system. The result: substantial savings and a better user experience.

What hadn’t been clarified, though, was who had decision-making authority over classroom technology within departments. A year after the change, when central IT informed the departments that it would be switching LMS vendors, the departments were irate. They demanded their individual contracts back—and got them. The savings were erased, and trust eroded. All agreed that they would never trust central IT again. If, at the outset, the university had established which party had decision rights over vendor selection, the collaboration would have succeeded.

As this experience illustrates, university staff often conflate making decisions about what to do with performing the tasks associated with the decisions. They believe they need to operate the server in order to choose the system—a flawed belief based on a lack of role clarity.

Besides role clarity, it’s critical to improve trust and accountability. People worry that the central function isn’t actually capable of delivering the agreed-on service level. They also fear that they have no recourse when such failure happens. These concerns can be addressed by clearly articulated and documented service-level agreements negotiated between the functional service provider and its customers. These agreements should spell out the expected level of performance and should be tracked through service-quality dashboards. Publishing the dashboards for all to see helps overcome suspicion and distrust about how decisions are being made.

Individual accountability also matters. Universities tend to be culturally averse to providing critical feedback to staff, and trust often flows more through relationships and familiarity than competence. At one university, of the more than 6,000 performance reviews on file from the prior couple of years, fewer than 10 people were rated as not meeting expectations. When my colleagues and I interviewed campus managers, it was clear there were far more than 10 underperformers on campus! The university needed metrics for evaluation that everyone can understand and apply consistently.

Finally, reversing the financial decline will require the combined efforts of top talent from throughout the university. This team shouldn’t be those who are readily available, but rather star players who are capable of filling roles outside their current role, as my colleague Michael Mankins has written. They’re more likely to bring a fresh perspective to the problem. Put simply, A players get A grades.

If universities are serious about getting their fiscal house in order, they’ll have to convince department fiefdoms to work together. Assigning clear roles and clear accountability, then harnessing a team of A players, is essential to enforcing the truce.