Documents/WH25PP/2: Large-Scale Programs

PART II: Large-Scale Programs

EFFECTIVELY MANAGING LARGE-SCALE IT PROGRAMS

Other Information:

IT has transformed how the private sector operates and has revolutionized the way in which it serves its customers. The Federal Government has largely missed out on these transformations, due in part to its poor management of large technology investments. To address these execution problems, we launched the IT Dashboard in June 2009, allowing the American people to monitor IT investments across the Federal Government and shining a light onto government operations. While this unprecedented transparency was an important first step, it was not enough to simply shine a light on problems and hope that solutions would follow. Building on the foundation of the IT Dashboard, we launched TechStat Accountability Sessions (“TechStats”) in January 2010. A TechStat is a face-to-face, evidence-based review of an IT program with OMB and agency leadership. TechStat sessions enable the government to turnaround, halt, or terminate IT investments that do not produce dividends for the American people. As a result of more than 50 TechStat reviews, OMB now has a sharper picture of the persistent problems facing Federal IT. One of the most consistent problems lies in project scope and timeline. In TechStat sessions, OMB found that many current IT projects are scheduled to produce the first deliverables years after work begins, in some cases up to six years later. In six years, technology will change, project sponsors will change, and, most importantly, program needs will change. Programs designed to deliver initial functionality after several years of planning are inevitably doomed. Modular development delivers functionality in shorter timeframes and has long been considered best practice in the private sector and in some areas of government; in fact, both Raines Rules and the Federal Acquisition Regulation (FAR) advise agencies to plan programs in this way. Successful organizations using modular development base releases on requirements they define at a high level and then refine through an iterative process, with extensive engagement and feedback from stakeholders. To maintain the discipline of on-time and on-budget, organizations push out additional functionality and new requirements for major changes into future releases and prioritize critical needs and end-user functionality. Evidence shows that modular development leads to increased success and reduced risk. However, because this is a new way of thinking about IT programs for some groups within government, it requires additional training, templates, and tools. Many existing government processes – from planning to budgeting to procurement – naturally favor larger, more comprehensive projects. As such, far too many Federal IT programs have multi-year timeframes well beyond the now accepted 18- to 24-month best practice. The activities outlined in this plan attempt to address the structural barriers to implementing modular development consistently across government. Moving forward, Federal IT programs must be structured to deploy working business functionality in release cycles no longer than 12 months, and, ideally, less than six months, with initial deployment to end users no later than 18 months after the program begins. Program managers need to define each phase of the IT development lifecycle and rigorously manage scope. These timelines should encompass the entire process – from concept through requirement analysis, development, test, and delivery. Today, a number of agencies have implemented these modular practices successfully. The Department of Veterans Affairs now requires that large IT programs deliver working functionality every six months. The following practices will help achieve the promises of modular development:

Objective(s):