'They said we never turned up'

Sooner or later, if you subcontract to facilities-management companies or managing agents, you get the call. Your engineer attended, did the work, and drove to the next site. Weeks later the client says there is no record of the visit, the SLA was breached, and they are withholding payment or applying a penalty. Your engineer remembers being there. The client's system says otherwise. Whose account wins?

Neither. The record wins, and if you don't have one, you lose by default. That is the uncomfortable centre of FM subcontracting: without timestamped evidence of attendance, a breach claim isn't a disagreement you argue, it is a dispute you have already lost. Memory, however honest, is not evidence. The engineer who knows he was there has nothing to show for it, and the penalty stands.

The structural problem is that you are rarely the one holding the system of record. The FM company logs the work orders. The managing agent runs the portal. When a query lands, they check their system, and their system reflects what reached it, not what happened on site. If your engineer's attendance never arrived in that portal as a timestamped, evidenced event, then from where they sit it may as well not have occurred. You are arguing against their data with your recollection, in a forum they control, months after the fact. That is not a fair fight, and it was never built to be one.

Records close disputes; memories start them

The flip side is that good records end these fast. The same FM analysis describes an operation that, once it had auditable timestamp evidence for every visit, cut dispute resolution from three or four days down to about two hours. A dispute with evidence is barely a dispute. Someone pulls the record, sees the arrival time and the signature, and the conversation is over. A dispute without evidence is days of email, defensiveness, and a relationship quietly souring.

Picture two versions of the same Tuesday. In the first, the client emails to say the fortnightly flush at a block of flats was missed on the 12th, and they are docking the month. You ring the engineer, who is fairly sure he went but couldn't swear to the date. You dig out the paper sheets, find the right site, and the entry for the 12th is there but undated, because the pad was filled in at the end of the week. Three days of back-and-forth later, you write off the charge to keep the relationship. In the second version you open the visit, read arrival at 08:41, a flush recorded against each outlet, a photo of the running tap and a caretaker's signature, all stamped on the 12th. You forward it, and there is no version three.

This is also why paper is losing its place in the chain. FM clients increasingly won't accept paper-based contractor workflows once you are onboarded, and they want completion evidence captured before the work order is closed, not written up that evening. The reason isn't fashion. A signed paper sheet scanned in three days later carries no reliable time. A record created and timestamped on site does. If you are still trying to win FM subcontracts in the first place, being able to produce that evidence is fast becoming the price of entry.

"Before the work order is closed" is the phrase that deserves a second read. It means the evidence and the completion are the same act, not two steps with a gap between them where the evidence gets forgotten. A work order closed without a photo, a time and a signature is, to a modern FM client, not closed at all. The burden has moved. It now sits on the contractor to prove attendance as a condition of getting paid, rather than on the client to prove absence before deducting.

What a defensible visit record contains

Strip it back and a record that survives a dispute holds a few things, all captured at the time rather than reconstructed later:

• Arrival and departure times, set when they happened, not rounded off in the van.
• The engineer's name, tied to the visit itself rather than to a rota somewhere.
• The per-asset readings and checks that show the work was actually done, not merely that someone stood in the building.
• Photos. The habit worth building is before, during and after, a tank lid off and back on, a gauge reading, a cleaned strainer, because a photo with a timestamp answers questions a line of text can't.
• A signature from someone on site, wherever you can get one.

The thread running through all of it is the timestamp, applied when the thing happened. A visit written up at 9pm from memory can be perfectly honest and still prove nothing, because every time in it is the time of writing, not the time of doing. Evidence is about when the record was made as much as what it says. A record that settles a dispute is also the one that shouldn't be sitting in a pile as proof debt, waiting for month-end to be written up.

Sometimes the client is right

Worth saying plainly: sometimes the visit really was missed. An engineer ran out of daylight, a job got dropped, and the record is absent because the work is absent. Good records protect the client in exactly the same way they protect you. That is the reason FM companies push for them, and it is a fair reason. A system where attendance is provable is one where a genuine miss shows up honestly and a false accusation falls apart just as fast. If your instinct is to resist that, it is worth asking which of the two you are actually protecting.

Auditable records are not a way to win arguments you deserve to lose. They are a way to make sure the argument is decided by what happened, which over a long relationship is the only footing worth being on.

The BlueWave engineer app captures the record on site: the engineer's name, every reading, photos and a signature, each stamped with the time it was taken, and held even when the plant room has no signal, then synced when it does. When an FM client queries a visit from six weeks ago, the answer is a record with a time on it, not a phone call to an engineer asking what he remembers. We wrote about the offline side of that separately, because plant rooms are where field apps quietly fail, and an attendance record that didn't save because there was no signal is no record at all.

One line to keep from all of this: if a record's timestamp is the time someone typed it up, it is not evidence, it is a story with a date on it. The only attendance record worth having is the one made on site, while the engineer was standing in the plant room. Everything written up later is just you, remembering, which is exactly what the client is doing too.