Emergency vs Scheduled Jobs: How to Triage Without Wrecking the Day's Schedule

TL;DR

The 9:40 a.m. call that sounds like an emergency is the most expensive two minutes in a service business's day. Say yes carelessly and you bump four booked customers, run into overtime, and pick up a one-star review from someone who did nothing wrong. Say no carelessly and you hand a high-value job to the competitor who answered second.

As of August 2026 the fix is not better instincts, it is a written triage rule that a dispatcher can apply in under two minutes: a defined test for what counts as an emergency, a defined amount of held-open capacity, a defined premium that is quoted before dispatch, and a defined recovery script for anyone you bump. This guide covers all four, plus the specific ways scheduling and dispatch software change the calculation.

Why triage fails in most shops

Almost nobody triages badly on purpose. It fails for three structural reasons.

The decision has no owner. The person on the phone does not feel authorised to say no, so they say "let me check and call you back", which is the worst outcome: the customer is not helped, and they call the next company while waiting.

The decision has no rule. Without a written test, urgency is judged by how upset the caller sounds. That systematically rewards the most demanding customer and penalises the polite one who booked properly last Tuesday, which is exactly backwards from a retention point of view.

The cost of a bump is invisible. Taking the emergency feels like adding revenue, because the emergency's value is a number you can see. The cost sits in places nobody totals: the bumped customer's lifetime value, the tech's overtime, the second bump that cascades from the first, the review you will not get.

A triage system exists to make those three things explicit before the phone rings, so the decision takes two minutes instead of twenty.

Step 1: write the test for what a real emergency is

Every trade's list differs, but the structure is the same. A real emergency has at least one of these: active damage that gets worse by the hour, a safety or security risk to people or property, or total loss of an essential service such as heat, water, power, or access to a home or vehicle.

That structure does real work. A burst supply line is active damage. A water heater that is leaking slowly into a pan is not, however unhappy the caller is. A dead furnace in a hard freeze is loss of an essential service. A furnace that is running but noisy is not. Someone locked out of their car in a car park at night is a safety and access issue. A spare key they have been meaning to order for six months is not.

Write your list on one page, with examples in your customers' words rather than trade language, and put it where whoever answers the phone can see it. Then add the two modifiers that change the answer:

Vulnerability. A frail or elderly customer, a household with an infant, or a commercial site where the outage stops the business trading legitimately moves up the queue even when the technical severity is moderate.

Consequence of waiting. Ask directly: what happens if we come at 8 a.m. tomorrow? Most callers will tell you honestly, and a surprising number will say that tomorrow is fine once someone gives them a specific time instead of a maybe.

Step 2: hold capacity open on purpose

The single most effective structural fix is also the one owners resist hardest: deliberately leave part of each technician's day unbooked.

Most shops find one to two hours per technician per day absorbs normal emergency volume without touching anyone's booked appointment. The correct number for your business is not a rule of thumb, though. It is in your own history: look at the last ninety days, count the same-day emergency calls you actually accepted, and size the buffer to the typical day rather than the worst one.

Held-open time feels like lost revenue. It is not, once you price the alternative. Every emergency you take out of a fully booked day costs you a bumped appointment, an unhappy customer, and usually some overtime. The buffer converts those costs into a slot you can fill with a same-day booking by mid-morning if no emergency arrives. The general principle is the same one behind blocking time for lunch, admin and drive time: the calendar has to reflect reality, or reality overwrites the calendar.

Two refinements are worth the effort. Put the buffer where your emergencies actually arrive, which for most trades is mid-morning and late afternoon rather than evenly spread. And in genuine peak season, widen it, because that is when both the emergency volume and the cost of a cascade are highest. That interacts directly with seasonal demand planning.

Step 3: decide with a table, not a feeling

Here is the decision matrix in its simplest usable form. Cross the severity test against your current capacity and the answer is already made.

The value of the table is not its cleverness. It is that a dispatcher can apply it while the caller is still on the line, and that two different dispatchers reach the same answer on the same call.

Step 4: price urgency honestly, before you dispatch

Charging a premium for emergency and after-hours work is fair. Overtime is real, and reordering another customer's day has a cost. What breaks trust is not the surcharge, it is the surprise.

Three rules keep it clean. Publish it, so it is not a secret. Quote it on the call, out loud, as part of the total, before the technician rolls, not as a line item discovered on the invoice. And confirm it in writing by text, so both sides have the same number in front of them. Two-way texting makes that a five-second step rather than a phone call, and we covered the mechanics in the two-way texting playbook.

One useful offer to keep in your pocket: for a caller whose situation is uncomfortable but not dangerous, quote the emergency total and the next-morning total side by side and let them choose. A meaningful share will take the morning slot, which protects your day and leaves the customer feeling in control rather than upsold. The ones who take the premium slot chose it knowingly, which is exactly the customer you want on a premium job. The same logic applies to how you structure after-hours and on-call coverage.

Step 5: have a recovery script for the customer you bump

If you have to bump someone, the bump is not what loses them. Silence is.

A recovery that works has four parts, and it must happen within minutes, not at the end of the day. A human contacts them, by phone if possible, text if not, and does not hide behind an automated message. They get a reason, briefly and truthfully, that an emergency came in, without dramatising it. They get a specific new time, offered as a choice of two concrete slots, never a vague promise to call back. And they get something, whether that is priority placement tomorrow, a discount, or a waived callout, sized to the inconvenience.

Then choose who you bump deliberately, not by whoever is last on the board. Bump the flexible, not the fragile: a routine maintenance visit rather than a customer who took a day off work, an established repeat customer with a good relationship rather than a first-time customer forming their entire impression of you today, a job with slack rather than one with a hard deadline. First-time customers should be near-untouchable, because a bump is the only experience they have of your company.

Automated reschedule notifications help, but they do not replace the call on a genuine bump. The reduce no-shows guide covers the notification mechanics, and the honest version of this is that a personal contact plus a specific replacement time retains almost everyone.

Where software actually changes the outcome

Three specific capabilities move the needle, and it is worth being precise about which.

Knowing who is genuinely closest and free. Triage decisions are usually made on a guess about where trucks are. GPS fleet tracking replaces the guess, and the difference shows up immediately: the emergency often goes to a tech twenty minutes away who was between jobs, and nobody gets bumped at all. That is the cheapest possible outcome, and it is unavailable to a dispatcher working from memory. More on the operational case in the GPS tracking guide and the dispatch software comparison.

Rescheduling and notifying in one action. The recovery script fails when it is manual, because it competes with the emergency for the dispatcher's attention in the same five minutes. Moving the appointment and notifying the customer should be a single step in the dispatch board, not a move plus a phone call plus a note to self.

Capturing the call at all. A significant share of emergency calls arrive while your team is already on another line or under a truck. Those go to voicemail and then to a competitor. An AI receptionist answers, runs the intake, applies the urgency questions, and books or escalates. It is the difference between triaging your emergencies and never hearing about them, and the case against a human answering service is laid out in AI receptionist vs human receptionist.

All three sit on the Pro plan at $199/mo for up to five staff, alongside payments and review routing; Starter is $79/mo for a single operator and Agency is $499/mo for unlimited staff, with two months free on annual billing. The full breakdown is at pricing.

Putting it together

Write the emergency test this week and pin it where the phone is. Look at ninety days of history and set a buffer. Publish the premium and quote it on the call. Write the four-line recovery script and make the bump decision a deliberate one. Then let the dispatcher decide, in two minutes, without you.

None of that is software. What software does is make the good decision cheap: it tells you who is actually free, it moves the bumped customer and tells them in one action, and it answers the phone when nobody can. That is usually the difference between a shop that dreads emergency calls and one that quietly makes money on them.

Frequently asked questions

How do I decide whether a call is a real emergency?

Use a written test rather than a gut feeling: a real emergency involves active damage, a safety or security risk, or a total loss of an essential service like heat, water, power or access. Everything else is urgent-feeling but schedulable, and should be offered the next genuine opening. Writing the test down is what stops the loudest caller from outranking the customer who booked a week ago.

Should I bump a scheduled customer for an emergency job?

Only when the emergency meets your written test and no other technician can take it, and even then the bumped customer should hear from a human within minutes, be offered a specific new time rather than a vague callback, and receive something for the inconvenience. A bump handled well loses almost nobody. A bump handled with silence loses the customer and often the review.

How much capacity should I hold open for emergency calls?

Most field-service shops find that deliberately leaving one to two hours unbooked per technician per day absorbs normal emergency volume without bumping anyone, and the right number is whatever matches your own last ninety days of same-day call history. Holding open time feels like lost revenue until you price in the bumped appointments, the overtime and the churn it prevents.

Is it fair to charge more for emergency and after-hours work?

Yes, provided the premium is published, quoted before you dispatch, and reflects a real cost such as overtime, callout, or reordering another customer's day. What damages trust is not the surcharge itself but discovering it on the invoice, so the number should be said out loud on the call and confirmed in writing by text before the technician rolls.

Who should make the triage decision, the dispatcher or the owner?

The dispatcher should make it, using a rule the owner wrote, because triage only works if the decision happens in the two minutes the caller is on the phone rather than waiting for the owner to be free. Give the dispatcher a clear test, an escalation threshold above which they call you, and the authority to say no.

Can software actually help with emergency triage?

Yes, in three specific ways: seeing which technician is genuinely closest and free right now rather than guessing, rescheduling a bumped appointment and notifying that customer in one action, and capturing the emergency call at all when your team is already on the phone. GetTimePad includes GPS tracking, two-way texting and an AI receptionist on the Pro plan at $199/mo, and there is a live demo at pricing.