The Sad State of I.T. Project Management

By Pascal Forget

December 10, 2006

I love programming. I have been programming since I was 10. I am now 36. Like most developers my age, I started programming in Basic, then moved to C, C++, and now Java since 1997. I've been involved in a ton of software projects over the years, and like most of us who've been around long enough, I've seen my share of doomed projects.

I am saddened by the lack of gain in maturity of our industry over the last 15 years. Our profession is supposed to be hiring among the brightest individuals on the planet, yet most projects are toroughly mismanaged and end up over budget and very late, when they are not outright cancelled.

If projects are totally mismanaged, why is it so? Well, for starters, the industry has an overwhelming number of people who did not stay sharp in software development, and so today the only job they can do is to be a project leader or a business analyst. However, these two functions are critical in the success of a project; how could a project succeed if the manager does not properly manage client expectations, scope, communications, resource allocations, deadlines and project timelines? How could a project succeed if the analyst produces a half-baked specification lacking clear requirements for error conditions handling and such?

Still, with such a large number of "Project Managers" and "Analysts" flooding the marketplace, the going rates for these types of jobs are actually LOWER than the rates for developers. What a shame! Most companies do not realize that cost should not be a factor when hiring the best possible project manager and analyst for a project. The lower market cost for project management is largely due to the hiring of project administrators rather than managers. The PMI or Prince designations do not guarantee the depth and breadth of experience. As a result, too many project managers and business analysts doom a project before it even starts. Oh I can hear the screams of people offended by my comments. Well I do not intend to be complacent over unqualified people who should be in another line of business.

Skilled developers, not being crazy, keep the highest paying job, so they remain in developer roles, although many of them would have an overall better effect on the project's outcome if they were acting as project manager or analyst.

The only way our corporations can compete against emerging third world countries such as India or China is to work smarter, not harder.

The proper execution of a project begins before the team is even assembled; solution definition and estimating are the bedrock. Even the strongest PM cannot work miracles all the time even with the best teams.

A project should be completely spec'd out before development starts. Make the client sign with his blood that the specification is complete and accurate. Who would think about building a house without a clear and precise plan? Well that's the stupid thing we keep doing over and again in our industry.

Change control should be implemented. Scope creep is the worst enemy of both a project's timeline and budget, yet it is commonplace. Why do we end up with scope creep? Because in most projects the client is not exposed to sanctions if he "forgot" to mention you this now very critical function that's missing from the requirements, and you feel bad about it and since most analysts and project managers are not strong enough to be able to say no to clients, projects end up delivered horribly late and over budget.

In my career, I have had the chance to turn around ailing projects and to turn them into great successes. When company executives saw that their projects were going down the drain and realized that something radical had to change in the way projects were being delivered in their organization, that's when they put out job offers for motivated highly technically skilled go-getters.

Success is not about using the latest fancy methodology. It's about hiring smart people who are highly motivated and act professionnally. It's about not allowing projects to spin out of control.

University curriculum for computer science and engineering students should expose students to sound management practices. Not enough people today are familiar with Professor W. E. Deming's teachings . Deming taught top management how to improve design (and thus service), product quality, testing and sales. He questioned the corporations' culture and the way managers operated. Deming talked not about quality but about management. He said that management actions were responsible for 85% of all problems in developing better products.

North american corporations are currently losing the productivity battle against emerging countries. In a rapidly globalizing economy, survival of the fittest means that corporations need to adapt quickly and optimize their internal processes in order to survive in an increasigly competitive environment. Those who do not adapt quickly enough will die. In professor Deming's own words, "It is not necessary to change, survival (yours) is not mandatory".

It also means that workers need to realize that they too must accept to come out of their comfort zone and take ownership of the continuous quality improvement process. As the old saying goes, you can bring a horse to water, but you can't make it drink, no matter how thirsty you want it to be. And that's why good managers have to be great leaders. Good managers must set an example, be good listeners, be mentors, and help people to pull away from their current practice and beliefs and move into the new philosophy without a feeling of guilt about the past.

With regards to cooperating with the lower cost geographies, or with off-site subcontractors for that matter, services must always be delivered where it makes most sense. Using subcontractors introduces complexity and overhead that can only be recouped for certain size and types of projects. It is always a question of delivering in the shortest amount of time possible, at the lowest possible cost while minimizing schedule and technological risk.

For an outsourcing strategy to work , you need the corporation to be able to prepare reasonably complete blueprints before it outsources the work. In too many places I've worked, the marketing department drives development, sets impossible deadlines, and starts out with only a blurry picture of what is required; this leads to evolutive prototyping, which leads to wasted resources (time and money) as certain parts of the solution are repeatedly reworked.

Competing against India and the other high quality/low costs geographies must begin by not competing. Innovation and quality will remain key differentiators for another 3-5 years as will industry knowledge. Cooperation with the lower cost geos is key to the survival of the mainstream IT industry in the west.

The solution to the current productivity crisis in the north american I.T. industry is known. The only ingredient missing is strong commitment from top management to initiate a continuous quality improvement process within the organization.

Pascal Forget is the president of UBeans Technologies inc. He is a freelance I.T. consultant, a city counselor, an airplane pilot, a husband and a father. Pascal holds a bachelor's degree in computer science from the university of Montreal.