How to run an effective, non-blaming Operational Health Audit

The goal is not to catch people doing things wrong. The goal is to understand whether the organization’s systems — designs, schedules, training, standard work, materials, and decisions — reliably produce safe, high-quality, on-time, cost-effective, and dependable results.

Before you go

• Define the scope: be specific about the process steps, equipment, or product families you will assess.
• Gather data: recent incidents, defect logs, downtime records, on-time delivery metrics, maintenance logs, and any recent audit or inspection reports.
• Invite the right people: one supervisor, 1–3 frontline operators or technicians, and a representative from maintenance or quality as appropriate.
• Plan time: a rapid audit can be done in 2–4 hours; a deep-dive may take several days with data analysis.

During the assessment

1. Observe, don’t interrupt. Watch the actual work. Record what people do, not what procedures say they should do.
2. Seek evidence. Use data where possible. Confirm whether documents reflect actual practice. Ask to see recent examples when someone mentions a problem.
3. Ask open, curious questions. Use phrases like “Help me understand how this is supposed to work” rather than “Who did this?”
4. Map the handoffs. Identify where responsibility and information move between people or teams — these are frequent sources of failure.
5. Score consistently. Use the five-domain scoring in the assessment form. Tie each score to explicit evidence to avoid subjectivity.

After the assessment

1. Synthesize findings quickly. Turn raw observations into 3–6 clear findings: what’s happening, why it matters, and who is affected.
2. Prefer system changes. Look for fixes that alter process, training, visual controls, or decision authority over temporary workarounds.
3. Create short verification plans. For every proposed action, identify a simple measure and a check date (e.g., "defect rate for XYZ reduced to <2% within 30 days").
4. Assign ownership and a next-check date. Avoid open-ended recommendations without a named owner and timeline.
5. Share results in a huddle. Present the top risks and the 3 prioritized actions in the next daily/weekly operational huddle to get help and accountability.

How to avoid blame and encourage improvements

Use language that describes systems and design: "The process allows late parts to be used without inspection" rather than "The operator ignored inspection." Encourage collectors to name constraints (time, materials, training) and decision rules that led to the behavior. Celebrate people who identify practical improvements and make it easy for them to test changes rapidly.

Common mistakes and how to avoid them

• Rushing to fixes without root-cause thinking — require a short root-cause note for every high-priority finding.
• Reporting everything — force the team to prioritize the top 3 opportunities that will reduce the biggest risk or cost.
• Leaving out verification — every action needs at least one measurable check.

Use the assessment form to collect comparable data across locations and the prioritization worksheet to focus limited improvement capacity on the most system-changing opportunities. Re-run the audit after 30–90 days to confirm progress and capture new learning.