The service work order workflow — from first call to invoice

A service work order passes through six stages before it becomes an invoice: intake, triage, assignment, the visit, the sign-off, billing. The problem is rarely the stages — it is the handovers between them.

Every handover that changes the medium loses something. Email to phone call, phone call to paper, paper to somebody's memory. You see the effect two weeks later, at invoicing.

Three places where the trail goes cold#

In the companies we talk to, the chain breaks at the same points.

Assignment by phone. The coordinator calls, the engineer agrees, nobody writes anything down. The state of the job now lives in two heads.

A sign-off on paper. Filled in on site, signed by the customer, back in the office on Friday. Or not.

Billing from memory. Someone rings a week later to ask how many hours it took, and gets a number rounded down.

Take one job closed last month and reconstruct it from the records alone. Who logged it, why it was warranty work, who went and when, where the invoice figure came from. What you cannot read off, the company does not hold.

Intake and triage: the two stages that set up the bill#

Intake takes moments, which is why it gets least attention.

A description, not a diagnosis. The commonest mistake is writing your own guess instead of the customer's words — "pump needs replacing" instead of "running loudly and tripping the breaker since Tuesday". The guess travels the whole chain: the engineer takes the wrong part and the second visit costs as much as the first.

An asset identifier, not just an address. The address says where to drive, the gate code and opening hours how to get in, the unit number what was done there before. Without a link to a specific asset, no service history is ever built.

Triage settles what the job actually is — urgent or not, warranty or chargeable, one person or two, part in stock or on order — and what the visit requires in people, parts and certifications.

The most consequential is the billing basis, and it is the one most often left until the end. A month later nobody can reconstruct whether the fault fell under the service contract or came from user error. Recorded at triage it is one sentence; recovered later it is an argument.

Assignment: the most expensive phone call in the chain#

The phone is convenient: it takes moments and answers at once. It costs what it leaves behind — nothing.

The system shows the job as unassigned while somebody is already driving to it, and whoever takes the customer's call cannot give a time slot. At eight crews and three arrangements each a day, twenty-four operational decisions live in one head — and a week of the coordinator's leave stops the company.

What has to be recorded: who is going, when, with what scope, and, separately, that they accepted it. Assigned versus accepted is the one status distinction worth fighting for: the first is what the coordinator wants, the second what will happen.

The visit and the sign-off: the sheet that never comes back#

The visit is the only stage that creates the truth about what happened; everything after it is a copy.

A paper sign-off has three faults, however carefully filled in. It comes back late — between the site and the office there is a day, a week or a van. It cannot be searched — "did we warn the customer the compressor needed replacing" has no answer six months on. It is the only copy — lose it and the visit goes with it.

One question defines what it must contain: what does the invoice need. Time on site and travel, parts used with quantities, work carried out, the customer's confirmation, anything left unfinished. Whatever is missing gets filled in from memory.

A separate field for "needs a return visit". That is where the next jobs come from. Without it the information stays in a conversation on site — sales not lost so much as unnoticed.

Billing: the stage that pays for all the others#

Billing creates nothing: it turns earlier records into an amount. If it takes even one phone call, an earlier stage failed to record something — and that stage needs fixing, not accounting.

Time and small items. Travel, half an hour waiting on site, a seal from the van. Half an hour of undocumented time across ten jobs a day is five hours that appear on no invoice.

The billing basis. Chargeable work invoiced as warranty, because the triage decision was never written down.

One measure covers the health of the chain: how many phone calls it takes to invoice a typical job. The target is zero.

Statuses: fewer is better#

A status answers one question — whose move is it now. Not what exactly is happening to the job; that is what fields and history are for.

The list should mirror handovers of responsibility, not shades of situation: new, triaged, assigned, accepted, in progress, ready to bill, closed, rejected. Anything beyond that has to justify itself.

They stop being trustworthy. When nobody is sure how on hold differs from awaiting parts, people pick whichever comes first, and a report built on that is worse than none: it looks credible.

Every status is an action. Somebody switches it, usually in the field after eight hours of work. A status nobody reads is a cost with no beneficiary.

People go back to the phone. If a status cannot be trusted, calling is faster, and the chain is back where it started.

A working rule: a state that does not change whose move it is, is a flag, not a status. Awaiting parts is a flag — the job is still in progress under the same owner.

Summary#

The difference between a chain that holds and one held together by phone calls sits in the handovers: whether each stage leaves what the next one needs, where it will look.

The rule fits in one sentence: no stage has the right to begin with a phone call. If it does, look one stage earlier, not where it hurts. Keep as many statuses as you have handovers of responsibility — the rest are flags.

The wider context of organising work for crews outside the office is in the guide on managing field teams.