
Most service companies measure this wrong and get a flattering number. Count return visits against original jobs rather than jobs closed on 1 visit, record the reason for every return, and 4 causes will account for nearly all of them. Only 2 are quick to fix.
A second visit is the most expensive event in a service business. It consumes a full slot, another round trip, and a customer's patience, and it produces no additional revenue on a fixed quote.
Yet almost no service company can say what its first-time fix rate is, and the ones that can usually measured it in a way that made the number look better than it is.
Getting the measurement honest is most of the work. Once the reasons are recorded, the improvements are obvious and largely mechanical.
The definition decides whether the number means anything.
The flattering version counts jobs marked complete on the first visit. It looks good because a technician who plans to return frequently closes the job anyway and a new one gets raised.
The honest version counts return visits to the same piece of equipment for the same fault within a window, usually thirty days, as a proportion of original jobs. That captures what the customer experienced, which is somebody coming back.
Which requires two things: jobs attached to equipment rather than addresses, and return visits linked to the original job rather than created as unrelated work. Roopairs holds equipment records with service history that jobs attach to, which is what makes the second visit visible as a second visit.
| Cause | What it looks like | How fast it moves |
|---|---|---|
| The part was not on the van | Diagnosed correctly, returned to fit it | Quickly, with van stock and history |
| The equipment was not what was expected | Wrong model, wrong voltage, wrong configuration | Quickly, with equipment records |
| The diagnosis was incomplete | Fixed a symptom, the cause returned | Slowly. This is training and time on site |
| Access or authorization failed | Could not reach it, or could not spend | Medium. Mostly a records and process fix |
The first two are the ones worth attacking first, because both are information problems rather than skill problems, and information problems are solvable in a quarter.
Both come from a technician arriving without knowing the equipment.
A job dispatched against an address tells the technician where to go. A job dispatched against a piece of equipment tells them what they will find: manufacturer, model, serial number, what was done to it last time, and what failed before.
That changes what goes on the van. A technician attending a specific combi oven with a known fault history carries different parts from one attending "oven fault, site 12".
It also changes the diagnosis. Roughly a third of the useful information in a repair is what happened last time, and it is free if somebody recorded it.
Which is why first-time fix and equipment records are the same project. A service company trying to improve this without equipment-level history is guessing at what to carry.
Knowing what the equipment is only helps if the part is reachable.
The parts to carry are not the ones a catalog suggests. They are the ones your own history shows you fitting, on the models your own customers actually have.
Pull last year's parts consumption, sorted by frequency and by model. The top of that list is your van stock, and most service companies find it shorter and more specific than what is currently in the vehicles.
Then review it quarterly. Equipment ages, customers change, and a van stocked from a list built two years ago is carrying weight for equipment nobody has any more.
Slower, and worth doing after the first two.
Incomplete diagnosis is the one that requires time rather than data. What helps: a rule that a repeat fault triggers a longer slot, and a habit of recording the cause rather than only the fix. A piece of equipment with the same fault three times has a diagnosis nobody finished, and that is visible only where faults are recorded as categories.
Access and authorization failures are process. A technician who cannot reach the unit, cannot get a key, or cannot exceed a spending limit has produced a return visit for reasons nobody in the field controls. Site access details on the record and an agreed out-of-hours authorization route remove most of them.
Do the arithmetic once and the priority becomes obvious.
Take your return visit count for a quarter. Multiply by the loaded cost of a visit including travel both ways. That figure is what the problem costs you directly, and it is usually larger than expected because the visits are spread across weeks and nobody adds them up.
Then add what is harder to price: the slots those visits consumed that could have carried new work, and the customers who now believe your first visit did not fix it.
On portal and chain accounts there is a third cost. Return visits appear on a scorecard, and a vendor with a visible pattern of them is a vendor being compared with alternatives.
Four steps, and the first is the only one that requires nothing new.
Step one alone changes the conversation. Most service companies assume their returns are diagnosis failures and discover they are parts and information failures, which are considerably cheaper to fix.
Two traps once you start reporting it.
Technicians closing jobs they intend to return to, which is the metric shaping behavior rather than behavior improving. Measuring returns rather than closures removes the incentive.
And treating every return as a failure. Some are unavoidable: a fault that only appears under load, a part nobody could have carried, a customer who declined the work on the first visit. Recording the reason is what separates those from the ones you can act on.
Roopairs runs a product tour where you can see equipment records, service history and parts inventory behind the dispatch board.
“Estimates for New Equipment Bids: The ability to create detailed estimates for new equipment bids is a fantastic addition that saves us time and enhances professionalism.” — Brady Hale
Count return visits to the same piece of equipment for the same fault within thirty days, as a proportion of original jobs. Counting jobs closed on the first visit produces a flattering number, because a technician planning to return often closes the job and a new one gets raised.
Four things: the part was not on the van, the equipment was not what was expected, the diagnosis was incomplete, or access and authorization failed. The first two are information problems and move quickly.
Because a job dispatched against an address tells a technician where to go, and one dispatched against a piece of equipment tells them what they will find. That changes what goes on the van and what the diagnosis starts from.
From your own parts consumption history, sorted by frequency and by model, rather than from a catalog list. Most service companies find the result shorter and more specific than what is currently in the vehicles.
The loaded cost of a visit including travel both ways, multiplied by your quarterly return count, plus the slots those visits consumed. On chain accounts add the scorecard, since a visible pattern of returns invites comparison with other vendors.
Record the reason on every return visit from a short list. That costs nothing, and after a quarter most service companies find their returns are parts and information failures rather than the diagnosis failures they assumed.
The all-in-one platform built for commercial kitchen service companies. Dispatch, invoicing, parts ordering — all in one place.
Book a product tour →