Lights, Camera, Systems Development!
By
Eugene Reagan
Senior Consultant
It takes a large number of people and a great deal of
money to make a movie. Yet, scores of films are produced each year
that bomb at the box office, fail with the critics or go straight to
video. No one sets out to make a bad movie. They all begin with what
seems like a good idea. Often the basic premise is wrong. The
decision-makers simply didn't understand the audience or the idea.
Just consider Ishtar or Waterworld.
So, how do people get involved in major enterprises
that, in retrospect, were doomed to fail?
This question is especially pertinent in the
technology-driven business that the insurance industry has become. Why
do so many major systems-development projects fail, totally or in
part?
According to the Standish Group of West Yarmouth,
Mass., in 2000, almost one quarter of all U.S. commercial software
projects were cancelled. The cancelled projects cost firms $67
billion, and overruns added another $21 billion, equaling $88 billion
in lost productivity on projects that were supposed to reduce costs,
increase revenue or improve operations.
Is it a failure to understand the idea or the
audience? Many times, yes. Frequently, there is no clear understanding
of the business need to be addressed. In a surprising number of
instances, systems developers have little understanding of the real
business problem that should be tackled. Superficial analysis or poor
communication can often lead to development efforts that do little to
address the true needs of the user audience. Often this takes the form
of overkill -- complex system projects with the latest, coolest
technology rather than simple add-on products or enhancements to
existing systems.
Is it a problem of execution? Quite often, yes. The
problem or benefit may be fairly well defined but the system
development effort may be under-funded or otherwise handicapped.
Project managers sometimes do not understand the complexity of the
newer technologies that may be employed. As a result, resources may
not be allocated properly, and the best-laid plans will fall apart.
The following tips describe some of the specific steps
that should take place early in the development process to help make
smart choices and good plans.
 |
Challenge business requirements. What is the real
user need? Do the requested changes or improvements truly address
the underlying business processes? Often requested enhancements are
only band-aids to address minor symptoms of much larger problems. |
|
Pit vendors against each other. It is a fair
question to ask, "Who is your competition?" Let each vendor tout
their product and critique their rivals. It all provides more
information for the client to evaluate. |
|
Talk to existing customers. Unless you consciously
choose to be on the "bleeding edge" of technology, let other
companies take the risks and then learn from their experience.
Review their selection process. What did they like or dislike about
the different vendors they reviewed? Do you have common issues? What
success or problems did they have with implementation? Are they
achieving the projected operational benefits? |
|
Perform a rigorous cost / benefit analysis. The
analysis of development and operating costs and potential benefits
cannot be too detailed. A thorough, detailed understanding of the
facts provides the confidence needed to make complicated decisions.
A thorough understanding of the projected costs and expected
benefits is key to determining project priorities. It enables the
project to be managed at a high level through the budget process. |
|
Analyze available development resources. Does
existing development staff have the skills and time to undertake a
major project? Will extensive training be required? Many times
organizations rationalize that training will be reusable and provide
value for future projects without validating that premise. What will
be the impact on other projects? Too often, companies are forced to
rely on the same key developers for all significant projects but
development staff should not be stretched too thin. |
|
Concentrate on business deliverables. What will the
business organization and process look like during each stage of
development and after complete implementation? Is the business
prepared to make the necessary changes? If the organization is not
truly committed to making major changes, significant benefits will
not be achieved. This lack of commitment should bring into question
the validity of many complex projects. |
