- Lack of effective governance
- Poor communication between business stakeholders and IT
- Not enough input from folks who will actually use a new system or technology
- Changes in project size, budget and scope
- Changes in key personnel such as project managers and executive sponsors
- Internal politics
- Infatuation with technology that is cool, though not necessarily useful
- Not enough courage to kill projects with poor prospects for success
Bruce Benson recently offered a list of five bad project practices on his Project Management Tools That Work blog, all of which involve how organizations collect and manage their data.
They could easily be subsets of the categories on my list, especially in the areas of governance and communication. The best thing about Benson's list is he includes good suggestions for solving the problems.
Benson's No. 1 reason is good data is hard to find, with Benson observing that presenting selective data and overly manipulating it can hide overall trends that are important to project success. His suggestion: Remain cognizant of the fact that data must show the "big picture."
The No. 2 reason is good data is locked away. He offers the example of committees that create controls for the kinds of reports that can be created and the data that can be included in them. While I'm sure this is ostensibly out of a desire to avoid too many different versions of the truth, Benson suggests a better approach is making data widely available so it can "be vetted by the masses and hence can be more accurate and useful for managing all projects."
Reason No. 3 is too much noise from throwing data around. Here, Benson discusses the issue of folks who manipulate data to reflect the picture they want to present rather than reality. He suggests displaying objective data right next to the "approved" reports to show a range of views. If anyone balks at this, it can be presented as a risk analysis. To truly solve the problem, the raw data behind reports should be available to anyone for subsequent analysis.
Some people do not believe data is reason No. 4. The main reason this problem exists, says Benson, is because a lack of good data analysis has hindered prior projects. He suggests BI pros must be patient and demonstrate success with smaller and less critical projects to build confidence in the data. (I'd add that BI pros should make as many clear connections as possible between good analysis and good decisions, so folks can see how the data helped.)
Benson's final reason is too many conflicting efforts to "improve" the organization. He relates the tale of a chief quality officer who wanted to improve the defect management process by changing how defects not attributable to software were entered into the system, a change that canceled out improvements that had been made in reporting defects. He says that remaining mindful of the issue should go a long way toward solving the problem.