We've all seen IT projects fail over the years, and for various reasons. And after spending two decades as an IT manager and executive in the financial services industry, a new columnist at InformationWeek named “Coverlet Meshing” shared five unspoken reasons IT projects often fail.

Meshing’s take on technology projects and the world of business may be seen by many as cynical, but as playwright George Bernard Shaw once said, “The power of accurate observation is commonly called cynicism by those who have not got it.”

1. Technology ROI numbers are mostly fiction.

The most complex variable in the ROI equation usually goes ignored, Meshing wrote. The cost of business re-architecture needed to consume a new technology is something most organizations simply don’t consider.

Technology demands business transformation, but for whatever reason, this concept typically goes unnoticed until it’s too late. New technology also usually means training, cultural changes, and procedural changes around the business. As the size of a technology project grows, so grows the number of associated changes to business processes. These things should be taken into account before deployment, but even if this is considered, it is no guarantee of success.

2. ROI rarely drives the technology investment decision.

In most companies, Meshing wrote, determining the potential costs and benefits of a technology investment is an elaborate, dishonest marketing exercise aimed at persuading senior stakeholders that just one opinion should win out. Usually, this decision ends up belonging to the most senior executive, and his decision typically has more to do with his career than the success of the organization. In large companies, Meshing wrote, the health of the organization and what’s best for senior management rarely align.

Therefore, decisions on which technology projects to fund is often driven by competing executives, not engineering data. There's no simple solution to this problem, but identifying an illness is the first step toward administering a cure.

3. There's rarely any long-term accountability in technology.

This idea follows the same line of thinking as the previous point. Just as executives are incentivized to make selfish decisions, those decisions will often have effects lasting longer than an individual’s reign at a given organization.

The idea that one must continually change jobs or be considered irrelevant means that when a technology project is rolled out, the people responsible for the project are often long gone by the time it’s completed and the mess is in someone else’s hands. For this reason, IT plans need to be agile and able to adopt to a business culture that is constantly shifting, Meshing wrote.

4. Detailed plans are the enemy.

It’s fine to have details, but don’t set them in stone. Details planned too far in advance could set a project off in the wrong direction. “Project plans that extend past 90 days are as accurate as TV weather predictions,” the author wrote.

Technology that is available in years or even months down the road may be different than what was available when the project began, and the infrastructure or the users’ needs may have changed by then. A successful IT project redefines itself often.

5. Bringing in the big outside guns only ensures that someone will get shot.

And the person who gets shot is usually the outsider. For one thing, an outsider is an easy target. But when the shooting is over, the insider is left to clean up the mess. Usually, bringing in an outsider with a big name isn’t a way to meet a goal, it’s a way for an executive to mitigate personal risk and keep their career safe, Meshing wrote, adding that “the cost savings that the offshore outsourcers promise are rarely realized and come at the price of increased project complexity.”

Managing projects and realizing cultural change to meet goals is already difficult enough. Including offshore outsourcers and offshore cultures complicates the situation too much to lead to a successful project. What’s more is that bringing in an outsider also promotes investment in proprietary technology that is probably not good for an organization and could lead to a dead end.