Inaccurate project budgets can be common in the tech sector. George Tillmann, VP and CIO of Booz Allen Hamilton, recalled a former client who consistently underbudgeted every IT project phase by 30 percent and, not surprisingly, always came in over budget.
“They never believed in the ability to predict costs or dates,” he explained. Management apparently didn’t require project leaders to be accurate, because the issue was never rectified.
Given today’s economy and business realities, such an approach is not viable. CIOs are not only tasked with delivering operational and financial benefits via projects, but also with making sure project cost estimates are accurate.
The best strategy, according to project management experts, is fairly straightforward—it’s a matter of CIOs asking the right questions and managing expectations. Questions should include:
- What’s the track record of the person preparing the budget?
- Where’s the project contingency?
- How does the budget relate to the project requirements?
- What are the project lifecycle costs?
Trusted person, trusted budget?
Many IT executives’ project approaches have been based on trust of the project manager. While it seems like a sound approach, it’s not a very viable practice, according to project management experts.
“Unfortunately, it’s then the individual you trust, not the budget,” said Ed Setar, IT director at the Project Management Institute.
That doesn’t mean the expertise or character of the budget creator isn’t important. There’s some value—if only to settle that sinking feeling in your gut—in finding out a little about the person who prepared the budget before you.
Tillmann recommends that CIOs probe the project leader’s project experience, and determine whether the individual was responsible for calculating the actual costs of previous projects. Project leaders should be able to show achievement in budgeting accuracy.
“Estimating is hard and you’ll be really bad at it when you start,” explained Tillmann. “The only way you’ll get better at it is if you have a feedback loop.” That includes reviewing problems and issues to avoid making the same mistake on the next project budget. If project leaders fail to do that, each estimate they make will be like their first—whether they’ve been a project manager for two years or 20.
If you can ascertain that a project leader uses a feedback loop and that his or her budget planning has improved with experience, you can make a safer bet on the project leader’s estimating accuracy.
Contingency here, there, and everywhere
Another question you should ask is where and how the project manager built contingency into the budget. A good rule of thumb is to tack 10 to 15 percent onto every project. But if, unbeknownst to the project manager, the project team added 10 percent to the estimate for every task in the project, and the project manager lumped on an additional 10 percent at the end, the project’s timeline and estimated costs will be way off.
In the best approach, contingency should be added at the end of the estimating process—otherwise no one will know where the allowance is. “Lie only once and lie last,” advised Tillmann. The bottom line is that it’s the CIO’s job to find out where the fudge is and if you don’t ask, you won’t know until it’s too late to undo the damage.
Project requirements should drive the budget
Simple as it may sound, the budget must relate to the project requirements. This assumes that a project scope exists and is of sufficient detail to create the estimate.
“If the project managers have worked carefully on defining the project scope and these have been carefully projected to a Work Breakdown Structure, then there is a higher degree of accuracy for such projects,” said TechRepublic member Haris Rashid, a management consultant in Islamabad, Pakistan.
Still, items in the project scope, such as the time and effort it takes to get a user community involved in project development, do get overlooked or underestimated.
“IT will develop a budget for X hours of functional requirements, specifications, programming… but won’t do detailed involvement of users, or the costs associated with the investment of the user community’s time,” explained Setar. That means the user community issues aren’t factored into the budget.
Budgets also typically shortchange testing and production deployment costs and labor, often because it’s not viewed as a big risk. Yet, too little testing can lead to the widescale deployment of buggy product. The firestorm of customer complaints that follows can hurt your company’s bottom line significantly. If you shortcut testing, you may end up having to go to an executive board to explain that a project will be late because you must conduct more thorough testing.
A project’s lifecycle costs
Another project cost “gotcha” happens when project managers and CIOs fail to consider the long-term impact of a project.
“There’s too much emphasis on upfront, one-time costs,” said Tillmann. “People aren’t concerned enough about the ongoing costs, but those are the ones that will really bite you on the rear end.”
Lifecycle costs can come from a variety of places, including:
- Project impact on the help desk and systems administration staff
- Operational drain from extending software licenses
- Future capital expenses related to additional hardware and administration tools
You need to know what events could drive up the project cost over time—from potential staff increases and network traffic issues to government regulations.
The one lifecycle cost most often overlooked is depreciation. A new project may have an upfront cost of $1 million for software licensing, new servers, and an army of disk drives and tape drives. But the same project might have to be depreciated $100,000 a year for the next 10 years, creating a serious drag on operations. That’s not news a CFO likes to hear.
Lastly, CIOs owe it to themselves and other company executives to set expectations for project budgets. That’s exactly why you have to make it clear to the project manager that the initial estimates will be refined and revisited at multiple points along the project.
At a minimum, budgets should be revisited after the analysis and design phases. The choice then is to live with the increased costs or cut the project back. The last thing you want to do is have the powers-that-be believe the project will cost $1.2 million and have it come in at $2 million.