Disarming Hidden IT Project Time BombsBy Guest Author | Posted 2014-11-19 Email Print
WEBINAR: On-demand webcast
Next-Generation Applications Require the Power and Performance of Next-Generation Workstations REGISTER >
How can you tell whether an IT project will be a boon or a bomb? And how can you spot the land mines in time to change course? Three disciplines can help.
Next, traditional estimation methods assume a stable environment, but corporate environments are anything but. Too many potential changes are outside your control, such as strategic realignments and/or key personnel losses that can dramatically affect resource availability and slow your project to a crawl. Also, external changes such as customer need shifts or competitive market entries can alter your target space, and therefore the functionality and scope of your initiative.
Finally, complex projects take time to complete—usually more than 18 months. During that time, projects are subject to the so-called “unknown unknowns”: the types of events that are extremely difficult to anticipate, yet are entirely possible during long projects. We’ve seen projects suffer delays due to hurricanes, tornadoes and blackouts. However, the original project plans did not allow for any of those events.
3. Simulate your risks.
Traditional estimation begins with brainstorming the sequence of tasks required to complete a project. Unfortunately, it also implicitly assumes that “unknowns” will not occur, and that, as we’ve seen, is a bad bet to make.
How then do we estimate projects on the high side of the tipping point? By using a different approach—particularly one that assesses risk in terms of probabilities. In our work, we’ve found Monte Carlo simulations (algorithms that rely on repeated random sampling to obtain numerical results) to be very effective at both compensating for the unpredictable and accounting for complexity.
The simulations are simple to perform. Your project team develops best-case, worst-case and most-likely duration estimates for each task. Then, using any number of inexpensive Monte Carlo software packages, you simulate the project multiple times. (We’ve found that doing 5,000 simulations works well.)
The primary result is a readout of possible project outcomes from which you can choose the duration that best fits your risk tolerance. Then plan your project accordingly. Also important is a sensitivity analysis report that details which tasks are the most risky. This allows you to mitigate the risks of each task before you take action.
Several years ago, American Express used this approach before an overhaul of its online membership rewards presence. The financial services giant wanted to completely rework not only the look and feel of its site, but also the underlying functionality available to its consumers.
The project would demand decommissioning existing systems, while others would need to be revised and new systems added. Initially, size estimates focused purely on scheduling, which determined the project would take approximately 33 months to complete.
Fortunately, company leaders recognized that earlier projects this big had underperformed in terms of schedule and budget. So they tried a new approach to forecasting and structuring large projects: Monte Carlo.
The simulation predicted that the project was more likely to consume 45 months—a full year more than expected. Armed with this knowledge, American Express used a series of risk-mitigation steps and completed the project on time and on budget.
Risk is inevitable, especially when it comes to large IT projects. Yet it's far better to know just how much and what type of risks you’re dealing with before it’s too late.
Therefore, insist on estimates that measure effort level (e.g., person-months) so you’ll know on which side of the risk tipping point the project falls. If it’s on the high side, use a risk-simulation approach such as Monte Carlo to understand the true schedule risks and restructure your project plan accordingly.
Follow this simple yet disciplined approach, and you can avoid costly surprises and be much more confident that your project will be completed on or very near schedule and budget. Even more important, you can rewrite what might have been an IT project horror story into one with a happy ending.
Ron Bisaccia is managing director at Alvarez & Marsal, a global professional services firm that delivers performance improvement, turnaround management and business advisory services.