Affordable IT: Staying on Budget

Most companies do a good job of tracking their biggest and most important IT projects, says Mike Metcalf, vice president of marketing at Artemis International Solutions Corp., which makes the Artemis project-management and budget-tracking tools. The problem, he says, is that hundreds of lesser projects are "crudely defined and poorly managed. Those are the ones that kill you."

Additionally, the difficulty with many IT projects is that they've never been done before--at least not at your site. Managers have an idea in their heads, but they don't know how to translate the job into IT time and resources. The full scope of the project may not be clear. Even in an infrastructure project, where no business units are involved, there's no sure way to know exactly how much time and money it will take to develop and deploy a new technology.

How To Avoid Futility"Every project is a practice in futility," says Douglas Davis, IS coordinator at Monical Pizza Corp., which operates about 60 restaurants. "First, everyone gives you the list of specifications he or she thinks you want--each person wants to convince you the project is really small. The true scope usually appears halfway through. But if you consistently overestimate on your budget plans, people tend to automatically assume you're overestimating."

Experts advise starting with known territory. For example, the hardware requirement for a given IT project is usually predictable, and you can use your IT inventory list (if you have the equipment) or the vendor's price sticker (if you're planning to buy) to calculate it. Packaged software also may have a predictable cost, though there may be variable costs associated with customization, testing and deployment.

The cost of IT staff time is also predictable--all staff salaries can be reduced to an hourly rate--but the number of staff hours required to complete a project is usually the project manager's biggest guess. "If you do your homework, the hardware will always come in on budget or below, but the labor will kill you," says a systems administrator. "Labor is always a wild card."

If your project involves implementation of a specific product, you may be able to get some intelligence on the resource requirements from your vendor representative. However, most vendors are reluctant to give the worst-case scenario, even after the contract is signed. Ask the vendor for a reference customer with a similar environment and project scope. Vendors usually underestimate the labor requirements for a project, while systems integrators usually overestimate, but you can usually count on your peers for an honest evaluation. If your vendor can't provide a solid user reference, look to user forums, newsgroups or professional associations.

As you evaluate technology costs, grill your business side associates to ensure that you understand and can meet their requirements. Many project budgets are overrun because the business unit decides it needs "one more thing" that wasn't included in the original budget. Also, talk with the end users of the new technology to be certain everyone understands and agrees with the scope of the project.Even after your due diligence, it's a good idea to insert some moderate wiggle room into your project budget. Most IT managers we interviewed said they include a contingency of about 10 percent, so that minor unexpected problems don't send the finance department into a panic. By the same token, most IT managers caution against padding the budget too much.

"If you continually overestimate your project needs just to come in under budget, you're tying up capital dollars that could be used elsewhere," says Christopher Clancy, regional director of technical services for the Cleveland Clinic Health System. "You can only do that so many times before you get a call to stop by the CFO's office."

Rather than overbudget, smart project managers qualify their line-by-line budgets by adding two components: a list of priorities and a contingency plan. The priority list, which should be developed with the business unit, outlines and ranks the project's goals. This list can be useful in assigning resources to each part of the project and helps you retrench if elements of the project cost more than anticipated. A logical next step, the contingency plan indicates which elements could be curtailed or cut if the budget is threatened.

Many experts advise project managers to gauge the contingency requirements by performing a risk assessment before setting anything in stone. As its name suggests, a risk assessment outlines the potential problems a project might encounter, as well as the potential impact on the business--or on other IT projects--if the project is delayed or runs over budget.

"You should evaluate not just the costs of the project, but the potential benefits and risks," says Georges Ataya, a member of the steering committee at the IT Governance Institute, which develops best practices for IT budgeting and project management. "Too many times, the project schedule and budget are driven by one technical manager who has made up his mind about how the project will go. But there may be other projects, business priorities or financial considerations that he doesn't know about. If you do your budget and time line in a vacuum, you're going to be in trouble down the line."This interdependence of project schedules and budgets often is not accounted for in project-management tools, such as Microsoft Project and the open-source Open Workbench, which are designed to aid in a project's planning, scheduling and monitoring. For this reason, companies are turning to "portfolio management" tools, such as Changepoint Portfolio Management and the Artemis suite, which provide a shared repository for project budgets and time lines. Portfolio management helps managers standardize the budget process, prioritize IT projects and see the effects of resource changes across multiple projects. (Click here for more on portfolio management.)

But with a starting price of $50,000, Artemis is not for all budgets. Smaller companies may want to take a look at the Information Systems Audit and Control Association's COBIT Quickstart, a set of guidelines for implementing IT planning, budgeting and project-management best practices that sells for just $50. COBIT (Control Objectives for Information and Related Technology), a broad framework for implementing IT governance, includes some guidelines on project portfolio management. More guidelines are in the works, say ISACA officials.

Although poor initial estimates are the most common reason for budget overruns, some projects break the bank because of a lack of supervision. Many organizations use scheduling tools to track project completion but don't keep close tabs on staff time or other "creeping" costs that can cause a project to slip into the red. "IT people feel that the most important thing is to finish on time, so they sometimes disregard spending as they revise their plans to keep up with the deadline," Ataya says.

Budget tracking can be handled in many ways, from periodic status meetings to a shared Excel file to sophisticated tools that manage the budgets of interdependent projects, such as the products from Artemis and Changepoint. Although many of the IT managers we interviewed extolled the virtues of shared databases and online budget tracking, several others said they don't use any budget-tracking software--and don't feel they've suffered for it. However, all the project managers agreed that frequent meetings and status reports are essential.

"Be personally involved and stay in communication with all the parties involved," says Paul Chernoff, director of IT at Washingtonian magazine. "This helps me to prevent communication problems, make sure the project is aimed at the proper goals and spot problems early. In my worst experience, I didn't stay involved. We went over budget, and after a year, we still didn't have the project done."Many projects are too focused on arbitrary "milestones" that are poorly defined or do little to advance the project, Jones says. "If I'm building a Web site and I say, 'We're going to have all the drawings done by a certain date,' that's a milestone--but it doesn't mean much."

Instead of milestones, Jones recommends that projects focus on "steppingstones," or tangible achievements that advance the project. In the Web site example, a steppingstone might be the completion of the home page development. "That's a point where all the pieces have come together, the business sees some value, and you can assess your real spending against your budgeted spending for that phase of the project," he says.

However, project planning is a two-way street. Many IT professionals have been hung out to dry after a business unit added new requirements to their project. Be certain that your business-side counterparts are stakeholders and have as much to lose as you do if the project's scope gets out of control.

Change management is another key part of keeping a project on budget, experts say. In many cases, the business unit may ask for what seems like a small change in the middle of a project, and the IT department accepts it. But changes can have a profound effect on the project time line and budget--or even the entire scope of the project--and any suggested modification should be carefully reviewed by both technical experts and business management. Artemis has a simulation tool built into its software that lets managers mock up a change to a project and assess the impact to budget and resources.

Seeing RedThe two best strategies for keeping an IT project on budget are to get the initial estimates right and to monitor spending every step of the way. But what about poor Joe Project Manager, whose project overshot its budget despite his team's careful planning?

An organization that has overrun its budget has two choices, experts say: Ask for more money (and revise the budget), or cut corners to make up the shortfall. Virtually all the IT project managers we interviewed have tried skimping, and most regretted it.

"If you need to cut corners, you are changing the scope of the project," says Norb Callahan, CTO at Metropolitan Properties of America, an owner and operator of luxury rental apartments. "Presumably, you worked hard to develop the scope--you shouldn't make a change unless you've communicated that with management and the project users."

If you made a priority list and contingency plan with your business associates during the budgeting process, you will have a leg up when overruns occur. The priority list will highlight which elements of the project are expendable, and your contingency plan will help you modify the project to keep it on track. With a well-planned budget, you can make some adjustments on the fly, without asking for more money.

If you've been through all the contingencies and you're still over budget, revisiting your budget with senior management doesn't have to be the end of the world, IT project managers say. In many cases, the business unit or top management can find ways to eliminate some requested features or capabilities. In more critical projects, most top executives will choose to bite the bullet and spend more to get the project done right.In a few cases, the organization may decide to kill a project to avoid throwing good money after bad. Earlier this year, the FBI said it will scrap its $170 million Virtual Case File project, opting for an off-the-shelf alternative that will be implemented in 2006.

"That was a project that was considered critical, and it was well-funded," Jones says. "But you can't solve every problem by giving it more money."

How To Keep Your Project on Budget

• Do your due diligence. Most budget overruns are caused by poor estimates of time and materials. Do research, contact your vendor, and consult with other enterprises that have done similar projects.

• Align your budget with business priorities. Working with stakeholders, make a list of the project's goals and prioritize them. Be sure that the business side has a stake in keeping the project on budget, and will be held accountable along with you if there is an overrun.• Conduct a risk assessment. A good budget always takes contingencies into account. Take time to understand everything that could go wrong. If you can't budget for the contingencies, at least you can record the potential costs.

• Don't budget in a vacuum. In any IT department, resource availability is affected by a broad number of projects, and yours may not be the top priority.

• Stay vigilant. Once the project is under way, you need a consistent method of checking the budget for potential overruns.

• Don't be afraid to revisit the budget. Consult with business and IT management about ways you can revise the project to keep costs under control. It's better to ask for more money than to cut corners and finish with a poor project.

Tim Wilson is Network Computing's editor, business technology. His background includes four years as an IT industry analyst and more than 14 years as a journalist specializing in networking technology. Write to him at [email protected].Affordable IT is focused on the challenges faced when doing business on a tight budget. Learn more.