Ask any field-service owner where the money goes and you will hear about diesel, salaries and customers who pay in ninety days. Almost nobody says the real answer, which is the second visit.
A job that gets closed on the first trip is a business. A job that needs a second trip is a hobby with a van. You pay a technician twice, you burn fuel twice, you lose a slot that another customer would have paid for, and you bill exactly once — because no customer in India has ever accepted a charge for you coming back with the right part.
The uncomfortable part is that repeat visits are not random. They cluster. The same three causes produce almost all of them, and all three are fixable without hiring anyone.
What a repeat visit actually costs you
Do this arithmetic once with your own figures and you will never look at a reopened job the same way. Take a technician's daily cost — salary, incentives, phone, whatever you actually spend to have him on the road. Add fuel and vehicle cost for the trip. Divide by the number of jobs he can genuinely do in a day, allowing for traffic and the customer who is at lunch.
That figure is what one visit costs. A second visit costs the same again, and it is pure loss, because the revenue for that job was already booked on the first trip. Worse, it is not one lost job. It is one lost job plus the paying job that could have filled the slot.
- The direct cost: the technician's time and the vehicle, spent again for zero additional revenue.
- The opportunity cost: the billable job that did not fit into the day because a fixed one came back.
- The scheduling cost: a return trip is squeezed in as an exception, which pushes something else late.
- The reputation cost: the customer now describes you as the firm that comes twice, not the firm that fixes things.
If ten of every hundred jobs need a return, you are running a tenth of your fleet for free, permanently. That is normally larger than the margin anyone is fighting over in a pricing meeting.
The three causes: wrong information, wrong part, wrong technician
Every second visit traces back to one of three things, and it is worth being blunt about each.
Wrong information. The complaint on the job says "AC not cooling". Whoever took the call did not ask for the make, the model, the tonnage, when it was last serviced, or whether it had the same fault four months ago. The technician arrives to discover a machine he has no context for.
Wrong part. Everybody's favourite. The fault is diagnosed in eleven minutes and the fix requires a component that is in the store, or worse, in another technician's van, or worse still, nobody is quite sure where it is.
Wrong technician. The job needed someone who has actually worked on that equipment, and it was allocated on the basis of who was free. The technician spends an hour learning the machine and then declines to take a risk with it, which is the right professional call and an expensive operational one.
First-time fix is not a technician skill problem. It is an information problem that shows up wearing a technician's uniform.
Asset history as the briefing your technician never gets
The single biggest upgrade to first-time fix is deciding that jobs hang off an asset, not off a customer. Not "Sharma Textiles, Unit 2, machine not working" but a specific unit at a specific site, with a serial number, a make, a model, an install date, a warranty status, and every visit anyone has ever made to it.
Once that exists, the job card stops being a complaint and starts being a briefing. Before the technician leaves he can see that this machine had a similar fault in February, that a particular part was replaced then, that it is out of warranty, and that access requires a gate pass which takes twenty minutes to arrange. He carries the right part and the right expectation.
- Serial, make, model and install date — so the correct spare is identified before the van moves, not after.
- Warranty status — the difference between a chargeable job and a manufacturer claim, decided in advance.
- Full service history on the unit — repeat faults become visible instead of being rediscovered every time.
- Site notes: access, gate pass, who holds the key, which hours are permitted, where to park.
- Which technician was there last, so the same person goes back where continuity matters.
The site notes look trivial on that list and are not. A wasted hour at a gate is the same hour whether it was lost to a missing part or a missing pass.
Van stock: the inventory that is not in your store
Most field-service firms think they have one inventory. They have two. There is the stock in the store, which somebody counts occasionally, and there is the stock riding around in four vans, which nobody counts at all.
The second one is where money hides. A technician draws parts on a Monday, uses some, keeps the rest in the vehicle as a personal buffer, and by month three there is a small hardware shop under his back seat. He is not stealing. He is compensating for a system that cannot tell him whether a part is available, so he hoards. Meanwhile the store shows those parts as issued, the reorder logic thinks they are consumed, and you buy more of something you already own four of.
You do not fix hoarding with a lecture. You fix it by making availability visible and issue quick, so keeping a private stash stops feeling necessary. Parts with a part number, MRP, cost and live stock, deducted automatically when they go onto a job, with low-stock alerts before the shelf is empty — that is what removes the incentive to hoard.
Parts consumed on site, and the ones that vanish
There is a second, quieter leak here, and it is on the revenue side. A technician replaces a component on site and does not record it. The customer is billed for labour only. The part is gone from stock and gone from the invoice, so you paid for it and gave it away.
It is almost never dishonesty. It is that recording the part meant a phone call to the office, and he was already late for the next job. Any system that makes recording a part harder than not recording it will lose this fight every single day.
- Parts go onto the job as lines while the work happens, not from memory at the end of the week.
- Adding a part to a job deducts it from stock in the same action — one entry, not two.
- The job shows both service and part lines, so the invoice is built from what was done rather than reconstructed.
- Anything not consumed goes back as a return, so van stock and store stock stay honest.
The test is simple. Pick a job from last month, open it, and see whether you can tell what parts were used without phoning anyone. If you cannot, that job was probably under-billed, and so were the forty jobs like it.
Closing a job properly: readings, photos, customer sign-off
A job is not finished when the machine starts working. It is finished when the record of what happened is complete enough that nobody has to ask a question about it again.
That means the job moves through its states honestly — open, assigned, in progress, completed — rather than sitting on "in progress" for a fortnight because closing it required paperwork. It means the readings taken, the fault found, and the work done are written down at the site. And it means the customer acknowledges it, because a customer who has signed off does not dispute the visit at renewal time.
The billing follows from the same discipline. A properly closed job already carries its service and part lines, so a GST invoice with the CGST/SGST split is a conversion of an existing record rather than a fresh act of data entry three days later. And a pay-link sent while the customer still remembers the fix being made collects far faster than one sent at month-end, when the machine has been running fine for a fortnight and the invoice looks like an interruption.
Measuring first-time fix without hiring a consultant
First-time fix rate sounds like something from a training deck. In practice it is one fraction: jobs closed on the first visit, divided by jobs closed. You need two things to compute it — a job record that knows how many visits it took, and the discipline to reopen the original job instead of creating a shiny new one when the technician goes back.
That second habit is where most firms lose the measurement. A return trip logged as a fresh job makes the numbers look wonderful and tells you nothing. One job, two visits, is the truth. Track it monthly and then cut it three ways, because the aggregate number is far less useful than its parts.
- By technician — reveals who needs training and who should be mentoring, which is often not who you assumed.
- By equipment type — reveals the machine your team is under-equipped for.
- By reason — no part, no information, no skill, customer unavailable. Four buckets, filled in at closing, in five seconds.
- By customer — one client with a chronic repeat rate usually has a site condition nobody has escalated.
Reordering before a job stalls waiting for a part
The final piece is boring and decisive: buy the part before you need it. Most stock-outs in this trade are not cash problems, they are visibility problems. Nobody knew the last two were gone until a technician was standing in front of a dead machine.
Set a reorder level on the parts that actually move — you will find a short list does most of the damage — and let a low-stock alert fire while there is still time to place an order. Record purchases against the supplier line by line so cost and stock stay accurate, because a reorder level built on wrong stock figures is just a slower way of being surprised.
None of this is a transformation programme. Asset history so the technician is briefed, live parts stock so he carries the right component, honest job closure so the work becomes an invoice, and a reorder level so the shelf is not empty. Four habits, and the second visit stops being a fact of life and starts being an exception you can name.