What causes software projects to fail?

The four most common causes

Unclear goal

"We are digitizing the order processing" is not a goal, but an initiative. Without a measurable goal, it is impossible to say at the end whether it worked – and during the project, it is unclear what is more important.

No one decides

When every question ends up in a committee, compromises are made that benefit no one. A project needs a person who can say, "this is how we will do it."

The scope grows, the budget does not

Each individual additional requirement sounds reasonable. In total, they exceed the framework. Without a place where wishes are collected and prioritized, this happens reliably.

Users are consulted too late

Software built for people without consulting them is avoided by those people. This becomes apparent only after launch – then it is expensive.

Early warning signs

• After four weeks, there is still nothing to look at
• Questions remain unanswered for longer than a week
• The requirements catalog is getting longer instead of shorter
• No one from the subject area has ever used the system
• When asked, "what is the most important goal?" there are three answers

What helps

Early on, something that can be tested. Nothing clarifies requirements as quickly as a system that can be touched.

A contact person with authority. The single most effective factor.

Collect wishes instead of implementing them immediately. A list visible to everyone that is regularly prioritized.

Users from day one. Not as opinion givers at the end, but as testers from the beginning.