Diagnose It Remotely, or This Needs Eyes on Site: Decision Tree

Why this matters

Rolling a truck for a problem you could have solved with a two-minute phone call wastes a slot another customer needed. Trying to fully diagnose over the phone something that genuinely needs hands and instruments on it sends a tech out with a wrong theory, or worse, talks a customer into doing something unsafe because you were guessing at what they were looking at. Getting this call right is a real skill, not a coin flip, and it starts before any camera comes out: if anything in the description sounds hazardous, stop triaging and give the safety instruction first.

Start here: any hazard signal ends remote triage immediately

Before you weigh remote vs. on-site, screen for these. If any are present, skip straight to the safety response, do not continue trying to diagnose remotely first.

• Any gas smell, or "rotten egg" smell.
• Smoke, an active burning smell, sparking, or visible flame.
• Standing water anywhere near an energized electrical source.
• The customer describing symptoms of a person being harmed (shock, injury, someone unconscious or impaired).

For any of these, tell the customer what to do right now (evacuate, do not touch anything electrical, call the utility or 911) and dispatch emergency service. This is not a remote-diagnosis decision.

If there's no hazard signal, ask these three questions

1. Can the fault be seen or heard clearly enough for a photo, video, or description to actually show it? Some faults are visual and obvious (a leak, a code on a display, a light blinking a pattern, a smell you can name, an unusual noise). Others are invisible without instruments (an electrical reading, an internal pressure, a subtle vibration). If the fault is the second kind, remote triage cannot confirm a diagnosis, only narrow the possibilities.

2. Is the fix, if there is a quick one, something the customer can safely do themselves? A tripped reset button, a clogged accessible filter, a breaker that needs resetting once (not repeatedly), a simple obstruction they can see and clear without opening anything: these are reasonable to walk a customer through. Anything requiring an open panel, disconnection, work at height, or handling anything under pressure is not.

3. Does the pattern match something with one dominant, well-known cause, or could it be several unrelated things? A single obvious symptom (one light blinking, one clear noise) with a strong single likely cause is a good remote-triage candidate. A vague, multi-symptom complaint ("it's just not working right") that could be a dozen different things is not; you will spend the call guessing instead of confirming.

Decision guide

If you choose remote triage

Move to a live video call over a phone description alone whenever you can; a customer's verbal description of what they are seeing is frequently wrong in specific, predictable ways (see the related article). Direct the call deliberately rather than letting the customer point the camera wherever they think matters. If, partway through the call, the symptom turns out to be more ambiguous than expected, say so and stop trying to force a remote diagnosis: "I can't confirm enough from here, let's get a tech out to look at this directly" is a legitimate outcome, not a failure of the call.

If you choose to dispatch

Send along whatever was learned during triage (any photo, video, code, or description gathered) with the job so the tech arrives with a head start instead of starting from zero. A failed remote-triage attempt is still useful information for the visit, not a wasted call.

References

• Trade-standard practice for remote visual triage and dispatch decision-making
• See related: Coaching a Customer Through a Live Video Diagnostic Call
• See related: What a Customer Consistently Gets Wrong Describing a Fault Remotely
• See related: The Remote Fix Attempt Worth Trying vs. Not (decision tree)