The Last Job of the Day Runs Late: Decision Tree

Why this matters

The last appointment of the day running long is a different problem from a mid-day slip. There is no next customer to push back, but there is a tired tech, a family or business waiting to close up, and a decision about overtime, a next-day return, or pushing through that has to be made in real time with limited information. Get this wrong on a regular basis and you either bleed unplanned overtime every week or you develop a reputation for leaving jobs unfinished at the end of the day. Get it right and it becomes a routine, low-drama call instead of a nightly scramble.

Start here: how close is the job to actually done

The decision hinges almost entirely on how much work is genuinely left, not on the clock alone.

• Close to finished (a final test, cleanup, paperwork, the last few minutes of a task already well underway) is worth pushing through in almost every case.
• Meaningfully more work remains (a new problem was just discovered, a part needs to be sourced, the scope turned out larger than booked) changes the calculation entirely, and finishing "just to be done today" is often the wrong call.

Get an honest read from the tech on-site before deciding anything. A tech under pressure to wrap up sometimes underestimates what is left; ask directly, "realistically, how much longer," not "can you finish soon."

If the job is genuinely close to done, push through

1. Let the tech finish rather than stopping arbitrarily at the clock. A job stopped ten minutes from completion to avoid a few minutes of overtime creates a second trip, a second drive, and a customer who now has an unfinished job overnight, which costs far more than the marginal time saved.
2. Notify whoever needs to know that the day is running past normal hours, both the customer (if they were expecting a specific end time) and anyone at home or elsewhere depending on the tech's schedule.
3. Log the actual finish time and the reason. A late finish that closes out a real accomplishment is a fine entry in the record. It also feeds tomorrow's booking, if this job type keeps running past its slot, the estimate needs adjusting.

If meaningfully more work remains, decide push-through vs return-tomorrow on the merits

This is the harder call, and it deserves an actual weighing, not a reflexive answer in either direction.

• Favor pushing through when: the customer has no access constraint for tomorrow either way, the system is unsafe or unusable if left as-is overnight (no heat in freezing weather, a water leak, a security issue), or the remaining work is well-defined and unlikely to reveal a new surprise.
• Favor stopping and returning tomorrow when: the remaining scope is uncertain and finishing tonight risks a rushed, lower-quality job under fatigue, the customer has no urgent need for same-day completion, or completing tonight would mean a tech working an unreasonable number of hours in a row with real safety and quality risk from exhaustion.
• Do not let sunk cost drive the decision. "We're already here, might as well finish" is not a reason on its own if the honest read says the remaining work is substantial and the tech is running on empty. A rushed finish that creates a callback costs more than an honest stop and a return visit.

If you decide to stop and return tomorrow

1. Leave the site in a safe, stable condition, never mid-teardown with something unsafe exposed. If that is not achievable quickly, that itself may force pushing through regardless of fatigue.
2. Explain the decision to the customer plainly, including why: "We want to do this right rather than rush the last piece tonight, we'll be back first thing tomorrow to finish." Most customers respond better to this than to a rushed job that turns out wrong.
3. Prioritize the return visit at the top of tomorrow's board, not squeezed in wherever there is room. A job left unfinished overnight owes the customer the earliest reasonable slot the next day, and booking it as an afterthought undoes the goodwill from being honest about the delay.

If this keeps happening with the same job type

A single late finish is a normal field-service night. The same job type running past its slot repeatedly is a scheduling signal, not a run of bad luck.

• Check the booked duration against the actual finish times for that job type over the past several instances. If it consistently runs over, the template is wrong, not the techs.
• Adjust the default slot length for that job type going forward, and consider whether it needs to stop being scheduled as the literal last appointment of the day if it has a real chance of extending indefinitely.

The recap

1. Get an honest, specific read on how much work genuinely remains before deciding anything.
2. If the job is nearly done, push through, notify as needed, and log the finish time.
3. If real work remains, weigh push-through against return-tomorrow on safety, customer urgency, and fatigue, not on sunk cost.
4. If stopping, leave the site safe, explain the decision honestly, and prioritize the return visit first thing.
5. If a job type chronically runs past its slot, fix the booked duration rather than repeating the same late night.

References

• See related: The Real Cost of an Inefficient Schedule
• See related: Communicating ETA Changes Without Annoying the Customer
• OSHA, guidance on worker fatigue and extended-hours safety risk
• Trade-standard practice for field-service end-of-day scheduling decisions