A truck roll is only wasted when the visit produced nothing that could not have been achieved remotely, or when it had to be repeated. This guide sets out how to find those visits in your own data, which categories are genuinely avoidable, how visual triage changes the decision, and how to estimate the value honestly.
Define the visits you are trying to remove
Before targeting a reduction, separate the categories. A first visit that fixes the fault is not waste. The visits worth attacking are those that produced no work, those that produced work someone else should have done, and those that had to be repeated because the technician arrived without the right skill set or parts.
Find them in your own records
Most service systems already hold the evidence, usually in closure codes and engineer notes.
- Jobs closed as 'no fault found' or 'customer instructed'
- Jobs closed with no parts consumed and short on-site duration
- Second and third visits to the same asset within a short window
- Jobs reassigned after arrival because of skill or access
- Jobs that turned out to be out of contract or another party's responsibility
What visual triage changes
Each of those categories comes from the same root cause: the decision to dispatch was made on a verbal description. Seeing the asset, the fault and the surroundings before scheduling converts some of them into a remote resolution, some into a correctly equipped visit, and some into a redirected request. It does not remove the need to attend genuine faults, and it should not be sold internally as if it did.
The workflow, step by step
The change is narrow and sits between intake and scheduling.
- Request received through any channel
- Guided capture link sent to the person at the asset
- Structured evidence returned when convenient
- Reviewer decides: resolve, request detail, dispatch, or redirect
- Dispatch decisions carry the skill set and parts list with them
Costing a visit honestly
A defensible cost per visit includes labour for the full door-to-door duration, vehicle running cost and mileage, the overhead loaded onto a field hour, and the opportunity cost of the work that engineer could not do. Exclude anything you cannot evidence. If internal figures differ between finance and operations, use the lower one — a conservative case survives scrutiny.
Estimating the value
Multiply the monthly volume of the categories above by the share you believe triage could genuinely influence, then by your fully loaded visit cost. Treat the result as a range and as an estimate, not a saving. The honest question for a pilot is not 'how much will we save' but 'what proportion of these specific requests changed outcome once we could see them'.
What goes wrong
Three failure modes recur: capture that asks for too much and gets abandoned, review that arrives too late to change the dispatch decision, and a triage step that is optional and therefore skipped under pressure. All three are process problems rather than technology problems, and all three are visible within the first weeks of a pilot.