You Missed Something on the First Diagnosis: Decision Tree

Why this matters

The repair did not hold, the symptom came back, or something on this visit flatly contradicts what you said last time. You missed it. This is not a worse version of your diagnosis and not a separate new problem - it is your original call being wrong. The danger now is the reflex to defend the first diagnosis by bolting a second guess onto it, which is how a one-callback job becomes a three-callback job. This tree is how you catch your own miss cleanly and re-diagnose without repeating the mistake that caused it.

Safety first: did the miss leave a hazard running

If what you missed is a hazard that has been operating since your last visit - an electrical fault you did not clear, a gas or combustion issue, a pressure or bonding problem - secure it before anything else. De-energize and verify dead, shut off and relieve pressure, kill combustion. Only then work the diagnosis. A missed hazard that has been live for days is the one that turns a callback into an incident.

Start here: confirm the miss before you re-diagnose

Do not assume the return is your error until you have ruled out the alternatives.

• Did the original repair actually get done correctly? A good diagnosis with a bad repair looks the same from the customer's door.
• Is this the same fault returning, or a genuinely different one that happened to follow? Same symptom does not always mean same cause.
• If it is the same fault and the repair was sound, then yes, the diagnosis missed something. Proceed.

Re-diagnose from zero, not from your first answer

The strongest trap is confirmation bias: protecting your original call and looking only for evidence that supports it. Set the first diagnosis down and work the problem as if you were walking in cold.

• Verify the primary symptom yourself, live. Do not trust the memory of last visit's reading.
• Test the cheap and common causes again, on purpose, even the ones you "already checked." The one you rushed is often the one you missed.
• Follow the fault to a measurement, not a hunch. "I replaced the part and it failed again" means the part was a symptom, not the cause. Find what killed it.

Sort it: wrong cause, or incomplete cause

Two kinds of miss, two different corrections.

• Wrong cause (misdiagnosis): you named the wrong root. You replaced the capacitor, but a failing motor was drawing it down; you cleared the local clog, but a collapsed line downstream is the real blockage; you swapped the breaker, but a ground fault in the circuit is tripping it. The fix is a fresh diagnosis to the true root.
• Incomplete cause (under-diagnosis): you found a real cause but not all of it. There were two faults, or a primary cause plus the condition that created it. Fixing one left the other to bring the symptom back.

Find the process gap that caused the miss

A miss is worth nothing if you do not learn why it happened, because the same gap will bite the next job.

• Anchored too early: you locked onto the first plausible cause and stopped testing. The cure is a rule to confirm before you commit.
• Skipped a step under time pressure: you shortcut a test to make the schedule. The cure is knowing which tests you are not allowed to skip.
• Trusted a bad reading: your instrument, or the customer's report, was wrong and you did not verify it. The cure is checking the tool and the story against reality.

Verify the corrected diagnosis before you act on it

Having missed once, do not swap another part on faith. Confirm the new root cause with a test that would prove you wrong if you were, then verify the repair under real operating conditions before you leave, not just at idle on the bench. The second miss on the same job is the one the customer never forgives.

Recap

1. Missed hazard still live? Secure it first.
2. Confirm it is actually your miss, not a bad repair or a new fault.
3. Re-diagnose cold; do not defend the first answer.
4. Wrong cause versus incomplete cause - correct accordingly.
5. Name the process gap so it does not repeat.
6. Verify the new diagnosis and the repair under real conditions before you leave.

References

• Trade-standard practice for root-cause diagnosis and post-repair verification
• Manufacturer documentation on failure modes and diagnostic sequence
• See related: Owning a Diagnosis You Have to Revise; The Real Fault Is Worse Than What You Diagnosed (decision tree)