Why digitalisation projects in mid-sized businesses really fail

When a digitalisation project fails, people usually point at the technology afterwards. The software was faulty, the provider too slow, the system too complicated. In the projects we've seen up close, though, technology was almost never the real cause. It was the symptom. The decisions that tip a project over are made earlier, and at the moment they're made they usually look entirely reasonable. Here are the five causes we come across again and again, and how to recognise each of them before it gets expensive.

1. Software was bought before the problem was understood

The classic starts with a trade fair, a recommendation or a convincing product demo. The tool is decided before anyone has written down which problem it's actually supposed to solve. After that, the business is adapted to the software rather than the other way around, and staff build themselves Excel side-ledgers to plug the gaps. How to spot it early: if project meetings talk more about the software's features than about your business's workflows, the order is back to front.

2. The people meant to use it daily weren't asked

Decisions are made at the conference table, the work happens elsewhere. When the clerk who has known the process for fifteen years sees the new system for the first time at the training session, the project is at risk, regardless of how good the software is. Not out of spite, but because nobody planned for her special cases, and special cases are half of everyday life in businesses that have grown organically. How to spot it early: ask the two most experienced staff in the affected area what they think of the planned system. If the answer is "this is the first I'm hearing of it", you've found your problem area.

3. Everything at once instead of one thing after another

The wish is understandable: if you're going to make the effort, do it properly, and in two years the whole business is digital. In practice, the big push overwhelms everyone involved at the same time. The project grows so large that nobody has an overview anymore, the first visible results are months away, and at some point the mood turns, because a lot of money has flowed and nothing in everyday work has improved. How to spot it early: if the project plan contains no point where something concrete runs in live operation after six to eight weeks, the project is cut too big.

4. Nobody really carries the responsibility

On paper there's a project manager. In practice he does it alongside his actual job, the provider points to the specification, management asks for the status every few weeks. When something then gets stuck, everyone is a bit responsible and nobody entirely. Problems are left lying, not because they're hard, but because it's unclear who is allowed to solve them. How to spot it early: ask the question "Who says stop if something is heading in the wrong direction?". If the answer is longer than one sentence, responsibility isn't settled.

5. After go-live, the project is "done"

The launch is celebrated, the provider says goodbye, and three months later it becomes clear what's really missing in live operation. No system fits perfectly on the first try, that's normal. It becomes a problem when nobody is left to fine-tune: small annoyances persist, staff get used to workarounds, and after a year the expensive system is only half used. How to spot it early: before the project starts, ask what happens in the three months after go-live and what it costs. If there's no concrete answer, support ends with the invoice.

The pattern behind the five points

All five causes share the same core: it's not technology that's missing, but honesty in the planning. Honestly clarifying which problem is to be solved. Honestly involving the people it affects. Honestly admitting that a small working step is worth more than a big plan. And honestly naming who is responsible, before and after go-live. The good thing: none of it costs money. It only costs the willingness to ask these questions before the first invoice is written. If you have a project ahead of you, take the five recognition questions from this article into your next meeting, they're uncomfortable, and that's exactly their purpose.

Wandelwerk Tech builds custom software for mid-sized businesses, with clear responsibility before and after go-live. If you're planning a project or one is currently stuck, talk to us.