Developing your IT project: a well-crafted plan is the key to success.

Corporate Websites and intranets have made huge strides in the last decade. Internet technologies have become responsive and reliable, with features improving by leaps and bounds. Companies are selling more of their wares online and are able to accomplish more with less staff because of improved efficiencies brought about by information technology. However, even with robust technologies and experienced IT staffs at the ready, failed projects continue to clutter the servers of corporate America.

Q. With all of the technology tools and innovations available these days, how could your IT project possibly fail?

A. Lack of planning. Failing to properly plan a project is an issue within any industry, but in the information technology field, the ramifications are amplified. A positive or negative experience is accentuated because an IT project usually touches so many people: customers, management, staff and vendors.

To ensure your IT project is a success, take the necessary steps to create a well-defined project plan. The plan should take into account your technical and business goals. You should include key players from your organization as well as expert IT resources that can help guide you through the process of building your project.

So what should be included in a project plan? It will differ slightly for each IT project, but you should define and document the following:

1. Identify ALL of the project stakeholders. Gather valuable input from every possible end user of the proposed project during the planning stage. Beyond the people perceived as the potential "customer" for the project, this could also include upper management, marketing, finance, IT, operations.

2. Define measurable business goals. Be as specific as possible when establishing the success criteria for your project, or else how will you prove to yourself that you've succeeded? This could be as straightforward as an ROI (return on investment) calculation, or it could be something more complex involving corporate policies or identity. Use definitive statements like:

* Increase online sales by 25 percent for the first quarter by getting 2,000 more site visitors per month.

* Reduce the sales process by three days with an online order entry form integrated with the distribution center.

* Free two employees for other tasks by connecting the document review process directly to multiple departments.

3. Understand the competitive landscape. Know what the payoff is before you begin development, and how the project will affect your standing among your competitors. Use your detailed, measurable goals, and re-evaluate throughout the planning process.

4. Clearly define the technical objectives. This may be one of the most daunting tasks of creating a project plan: development of a project scope document. Defining technical objectives requires an in-depth knowledge of software architecture and the ability to mesh technologies and business goals together. The project scope document should contain detailed directions for building your project. If you don't posses the skill set to properly create this document, enlist help from others that do.

5. Select the proper technical architecture. What's the right technology platform for your project? It's dependent upon several factors such as budget, existing technology infrastructure, ongoing support staff, etc. Be open about what is the best technology platform based on real needs, rather than what's being touted as the "next big thing."

6. Take a phased approach. A phased approach, also known as iterative development, allows you to adjust the project during its development period. The one certainty about technology projects is that things will change and evolve. Give yourself flexibility by planning for this natural shift to occur--don't commit to a year-long development cycle for a monolithic technical plan that promises to solve all of your problems on time and on budget, with no wiggle room in the schedule before some crucial corporate deadline.

7. Build a pilot. A pilot isn't a fully operational piece of software, but it is a visual representation of your initiative. It's a lot easier to communicate functions to end-users with pictures than with a long tick-list of features. And of course, it's easier to adjust the visual model of your project based on feedback from end-users, rather than trying to retrofit requirements after you've invested a large sum of money towards an expensive software solution.

Dan Reynolds is president of Boldface Technologies Inc. in Dearborn, a member of the Detroit Regional Chamber.