How to Find the Root Cause Before You Fix a Problem

Before you launch a fix, write down the cause you think you found and ask what evidence would prove it wrong. That pause can save a team from spending weeks improving the wrong process.

A visible symptom is not always a useful diagnosis. Missed deadlines might come from unclear ownership, shifting priorities, a slow approval, or work that was never scoped. The first explanation that sounds plausible is still a hypothesis.

Ask what changed

Start with the facts. What happened, when did it start, and what changed around that time? A new policy, staffing change, tool, workload, or handoff may help explain the pattern.

Talk to the people closest to the work before deciding what to do. They may see friction that does not appear in a status report. Ask each person what they observed, not what they think leadership wants to hear.

Keep observations separate from interpretations. "Three approvals were delayed this week" is an observation. "The team does not care about the deadline" is an interpretation.

Use the 5 Whys as a conversation

The 5 Whys technique is a way to keep asking how one condition led to another. It is not a contest to reach five answers, and it should not end with a person's name as the cause.

For example, a customer response was late. Why? The request sat in a shared queue. Why? No one was assigned to review it each morning. Why? The team changed its handoff process, but the new owner was never defined. That chain points toward a process gap the team can inspect.

The chain may branch. One answer can lead to several contributing factors. Follow the evidence and check each answer with people who know the work. Do not force every problem into a single cause.

Test the explanation

Before acting, ask what evidence supports this explanation and what would make it less likely. Talk to someone who sees the problem differently. Consider whether changing this condition would probably change the symptom, and whether the fix could create another problem elsewhere.

If important evidence is missing, gather it before you commit. A small check with the people affected may be enough. A complex problem may need a longer review.

Set a stopping point for diagnosis. You have enough to test a fix when the evidence supports a cause, affected people recognize the pattern, and the proposed change can be observed without creating undue risk. If two explanations still fit, design a small check that distinguishes them. Do not present an unresolved guess as a finding.

Handle difficult causes honestly

Sometimes the cause involves a decision, incentive, or conflict that is uncomfortable to discuss. Do not quietly replace it with a safer explanation. Record what the evidence shows, decide who needs to be involved, and choose a responsible way to raise it.

Root cause work does not mean blaming a person. It means understanding the conditions that made the outcome likely. Accountability still matters, but it should follow a fair review of expectations, resources, and decisions.

Make the fix measurable

Write the cause in one plain sentence. Then define a change that addresses it, an owner, and a date to check whether it worked. Choose a measure connected to the original problem, such as fewer handoff delays or fewer repeated errors. If the measure does not move, revisit the diagnosis.

Managers often move quickly because an action plan looks like progress. Diagnosis is part of the work. Ask one more question, check with someone close to the process, and test the explanation before building the fix.