“Digitale Wissenbissen": These 3 stumbling blocks trip up software projects
An incredible amount of cool technology is waiting to be integrated. Medium-sized companies have enormous potential to gain a real innovative edge through their own software projects. But the reality is different: 75% of all software projects fail – they exceed the budget, blow the time frame, or don't deliver the desired results.
Why is that? And are there any magic levers to prevent failure? In this episode, we uncover the three common pitfalls that trip up projects – regardless of whether they have a budget of $10,000 or $10 million.
Key points from the episode
1. Grace-based collaboration – the invisible project killer
The core project team is in place, the skills are there, the budget is there. But then the team needs input from other departments: APIs from the IT team, legal clarifications, data management. The problem? No one has formalized WHEN and WHY these teams need to deliver.
- Other teams know they are supposed to help, but have their own roadmaps and priorities
- Escalation paths often lead to senior management
- Matrix organizations exacerbate the problem
2. Agile chaos instead of agile processes
Many large projects label themselves “agile” without understanding the basics. Agility is the opposite of chaos – it requires clear structures, roles, and processes.
- Projects start without requirements AND without a strong product owner
- Stakeholders show up sporadically, give new directions, and then disappear again
- Hybrid models without clear responsibilities are often doomed to fail
3. The greenfield temptation
Development teams often push to rewrite everything instead of extending existing software. “It'll be much faster with the new framework!” – a dangerous fallacy.
- The effort required to catch up with the legacy system is massively underestimated
- Every “obscure” button is used by someone
- Migration strategies are planned far too late
You don't need magic levers or better tools—you need better decisions and structures. If you want to master large projects, you first need to think about organizational structure, processes, and risk management, not Jira versus Trello.
