From Stakeholder Discovery to a Working Action Plan

The stakes

The expensive failure mode in a large organisation is not bad code. It is building the wrong thing confidently. A Big Four firm in Luxembourg had a project to move forward and a team to bring up to speed, but the stakeholders were not aligned on what was actually needed. Every week spent building or teaching on top of unvalidated assumptions would have compounded the cost, in the quiet way that only shows up months later.

The most expensive software is the software built on an assumption nobody checked.

The move

We refused to start with the build. The first work was with people, not systems: drawing out from each stakeholder what the project actually needed, then validating those requirements against how the team really works rather than how it was assumed to. That separation, real problems from assumed ones, is unglamorous, and it is the step most engagements skip because it produces no code. It is also the step that decides whether everything after it is worth doing.

What shipped

A concrete, sequenced action plan built on the validated requirements, owned by the people who would deliver it, not by us. Alongside it, hands-on training, so the knowledge stayed with the team rather than leaving when we did.

What it changed

• Stakeholders aligned on a single, validated definition of what was needed
• The genuine problems were on the table and the assumed ones were off it
• The team owned the plan and kept the work moving on its own
• The firm avoided the most expensive outcome: confidently building the wrong thing