A taxing project


A seeming lack of clear leadership has put a project by the Canada Revenue Agency over budget and off schedule.

In 2000, the Canada Revenue Agency (CRA) launched a project to replace an outdated computer system responsible for processing a federal value-added tax. The project hasn't added much value for stakeholders, however.


Originally budgeted at $98.5 million, the project is now closer to $200 million because of additional training and testing. Its original 2004 completion date had to be pushed back to October 2006—and even that deadline may not be met.

The problems began in 1991, when, after only 18 months of development, the CRA—then called the Canadian Department of National Revenue—went live with the original computer system. Because of its quick turnout, according to a Canadian Press article, the system suffered several fundamental deficiencies including:

It did not communicate with other tax systems run by the agency

It did not correctly calculate penalties and interests on overdue or excess payments.

The CRA also said the system corrupted data and mismatched filing periods on an estimated 740,000 to two million master records. Because of these shortcomings, the CRA has had to pull out their pencils and manually calculate data, which made for plenty of corrections and clean-ups, according to Canadian Press.

The responsibility for the CRA's new computer project has been jointly shared by the assessment and client services and IT branches. “The Canadian Revenue Agency should have learned a lesson from the 1991 project,” says Oliver Lehmann, PMP, director at large (case analysis) of the PMI Troubled Projects Specific Interest Group. “A project of this importance, size and complexity needs a single project manager who is well-educated and strong enough to make the right decisions—just like a ship needs a captain.”

The project relies on committee leadership, according to a CRA audit report published in January 2006. There is no project manager with accountability on top, Mr. Lehmann says. “Dual leadership too often means there is no leadership at all,” he says. “Having two individuals in command may result in disputes and resentment. Often, however, one of the heads wants to consult the other person before taking responsibility and making a decision needed right away.”




Related Content

  • PM Network

    The 20 - percent solution member content open

    By Smith, Ronald B. Most recently baselined project plans are inaccurate because they have unrealistic start dates, finish dates, work hours, costs and/or durations. Poor estimating during the planning phase is a big…

  • PM Network

    Measuring delay member content open

    By Rajagopalan, Sriram Project managers can evaluate the cost of schedule delays and proactively manage resources by envisioning delay as a pseudo-resource, measuring the difference between actual finish and baseline…

  • Scheduling optimization with line of balance and start-finish relations / Ricardo Viana Vargas, Felipe Fernandes Moreira. member content open

    By Vargas, Ricardo Viana | Moreira, Felipe Fernandes The start-finish relationship between activities are often misunderstood; partly due to the lack of understanding regarding the true meaning of logical relationships. This paper, which has the…

  • PM Network

    Imposed deadline syndrome member content open

    By Heerkens, Gary Contributing editor Gary Heerkens encourages practitioners to learn to generate bottom-up schedule and cost estimates that deliver realistic goals to senior management, responsibilities that seem to…

  • Project planning-- must it be a Sisyphean task? member content open

    By Brokman, Yaniv | Pilovsky, Yuval A typical technology development project is composed of, over the course of more than a year, thousands of activities including the work of dozens of engineers from a variety of disciplines.…