Describe what is happening without naming a purchase.
Replace “we need a new system” with an observable statement: where work slows, what information is missing, how often the issue occurs, and who is affected. A specific description makes it easier to test explanations rather than defend a preferred solution.
Trace one real example.
Follow a recent case from start to finish. Where did the work wait? Who needed information? Which decisions required an exception? Speak with the people doing the work and compare the intended process with what actually happened.
Separate symptoms, causes, and constraints.
Late follow-up might reflect workload, unclear ownership, poor information, or a tool problem. More leads would affect each situation differently. Identify the evidence for each explanation and the constraints any solution must respect.
Test a small, reversible change.
Where practical, try one intervention with a named owner and a clear observation period. It might be a decision rule, a clearer handoff, or a limited process change. Decide in advance what would count as improvement and what would cause you to stop or revise the experiment.
Bring expertise to the unanswered question.
Define what an outside specialist needs to resolve and the evidence you already have. A focused question makes their time more useful and helps you evaluate whether the answer addresses the underlying problem. The appropriate outcome may be a better implementation plan—or no new purchase at all.