As a rule, Sherwin-Williams Co. prioritizes IT projects according to five criteria: their financial, strategic and business impact, and their technical and business difficulty. Higher ranked projects are given a higher priority; mandatory, “keep-the-lights-on” projects are given top priority. “That’s how we normally do it,” says Gary Marich, manager of application development at the Cleveland-based paint-and-coating company. “Then we have the bombs that get dropped.” A business-unit manager, for example, might convince the CFO that her project needs to be a top priority, or perhaps two directors with competing projects will each insist that the controller said theirs is the most important. When this sort of thing happens, Marich pushes back. He sends the managers back to their bosses, or he seeks them out himself and explains how their requests will affect other projects. “The goal,” Marich says, “is to keep the politics under control.”
Fifty-three percent of the respondents to this month’s survey on project management say that the project prioritization process at their company is politically driven. That’s an awfully high percentage, especially given the danger that executives at more influential business units will win funding for projects, whether or not they provide significant business value, while potentially more valuable projects will be killed, go unnoticed, or simply languish. The result can be higher IT costs, poor IT-business alignment, missed strategic opportunities and delayed projects.
No one we spoke with feels politics can be eliminated entirely. “Politics in projects is natural,” says Gopal Kapur, president of the Center for Project Management in San Ramon, Calif., “because projects change the status quo, and change means stress. The survival instinct kicks in. People want to make sure they get the most resources they can, and come out the least damaged.” And since IT is a finite resource, business-unit executives will always compete for the biggest slice of the pie.
Is it possible to minimize the effect of politics? Yes, insist those CIOs who say that their project prioritization processes are not politically driven. Notably, some small and well-focused companies—those, for example, in which a small team of top executives, who share a common set of values and measurements, make the project decisions—might be able to keep politics in check. That’s how it works at Insurance Auto Auctions Inc., says John Nordin, vice president and CIO at the $200 million auto-salvage company based in Schaumburg, Ill. The six top executives meet every Monday, look at the ROI and technology, consider the business and implementation risks of any IT projects on the table, and then focus on their customers by asking a basic question: “What am I doing today to sell a salvaged car?”
Other, larger companies typically need a more formal process if politics is to be minimized. At Advanta Corp., a $377 million Salt Lake City firm that provides credit cards for small businesses, both IT and non-IT projects come under the same three-part process (managed by the company’s project-management office), according to John Dana Smith, vice president of IT at Advanta Bank Corp. A committee of directors and vice presidents from different organizations meets weekly to make sure the project’s sponsors have done the ROI and risk analysis; a higher level executive committee of vice presidents and directors decides whether the project should proceed and how high a priority it should be given; and the chairman or another executive gives the nod. The chairman himself makes sure that no executive attempts to circumvent the process, Smith says.
At Photronics Inc., a $349 million firm based in Brookfield, Conn. that makes photomasks (a key tool used in the manufacture of semiconductors), the process covers only IT projects. Shawn Knox, vice president of information systems at Photronics, says that prioritization and approval resides in an eight-person committee composed of vice presidents, executive directors and directors from different business functions. Sponsors must use a Web portal to fill out a project application that also requires them to provide information on benefits, alternatives and risks. Attempts to circumvent the process are discovered and confronted. For example, Knox recalls a case in which an IT director who had failed to win approval to work on a new software development methodology attempted to do the project on his own. The project was discovered when other IT projects began falling behind schedule, and it became clear that precious IT resources were being diverted.
The politics of project prioritization cannot be entirely eliminated—at least not until there’s a way to banish self- interest from human nature. But CIOs at least have a fighting chance of minimizing politics if, as Kapur says, a clearly defined process aligns every project to strategy, an executive steering group establishes project priorities, and an education process is in place so that “people will see why this governance structure will benefit us all.”