What insurers cannot see, they price as risk.

A large construction site now keeps a continuous record of itself. Fixed cameras, walk-throughs, moisture and weather sensors, progress against the programme - bought by the contractor or the owner, for the site's own reasons, and running whether anyone on the insurance side looks or not. Almost none of it reaches the people carrying the risk.

The usual explanation is that nobody asked. That is half the story. The other half is that the people who own the record have reasons not to share it, and most of those reasons are good ones. Anyone who has asked a contractor for access to a camera feed has heard some version of what follows. This piece takes the reasons in turn, because they deserve better than to be waved away.

"It will be used against us"

The loudest objection, and the one everything else hides behind. Share a continuous record of the site and, when a claim comes, that record is the first thing the other side reaches for.

Two things are true at once here. Evidence cuts both ways; a record that shows what happened will show it whichever way it went. But an account that exists cannot be reconstructed against you from fragments and recollection, and an account that does not exist will be. That is the position today. What a day is worth set out the reasoning: a delay claim is an argument about sequence, and that argument is shorter when the sequence was observed.

There is also a question of who is reading. The objection imagines the adjuster, after the loss. The reader Pikt builds for is the risk engineer, before one - the function whose job is to see a problem developing and help get it fixed while it is still only a problem. That is a different relationship with the same record.

"It shows the owner we are behind"

This one is quieter and, for a contractor, often the real one. Progress data shows slippage, and it shows it to the owner and the funders before the contractor has had a chance to recover it or explain it.

It is the hardest objection to answer, so it is worth being honest about what is and is not in it for the contractor. What reaches the risk engineer is the condition of the site and its progress against the programme, read for risk. It is not the commercial position, and the risk engineer is not the owner's quantity surveyor. Progress against the programme is already in the engineer's report today; what changes is when it is seen, not whether. Access is granted per source, to the risk function. And slippage that is visible early is slippage that can be managed; the alternative is not that it stays hidden, but that it is discovered later, by the same people, when it costs more to explain.

"It is surveillance"

Cameras on a site raise a reasonable question about who is watching the workforce, and it is a question a site team is right to ask.

The answer is that nothing new is installed. The cameras and sensors are already on site for the project's own purposes; the question is who else has access to what they already produce, and for what. What the risk engineer is looking at is the works - the slab, the scaffold, the water where water should not be. People appear in the frame, as they do on any site camera, but nobody is watching the workforce: the engineer is not identifying individuals or building a case against anyone. Pikt records what the site's systems already capture; it adds no camera and no new kind of watching. The engineer is not there to catch breaches for penalty. The output of a risk engineer's screen is a recommendation and a conversation, not a sanction. This is not the insurer's CCTV.

"Who else gets it?"

In a market where a single project's risk is shared across many carriers, the question of onward sharing is not paranoia. Grant access to one party and it is reasonable to ask how many others are behind them.

The site's data stays where it is, in the systems the project already runs, under whatever terms the project bought them on. Pikt reads it with the consent of whoever holds it, keeps a time-stamped account of what it read and when, and does not sell or trade any of it. It records what the site's own systems say, with the source and the time, rather than judging it. Pikt is an independent technology company. What flows onward is what already flows today: the risk engineer's account, to the parties already entitled to it. The difference is what stands behind that account.

"What is in it for us?"

The honest answer starts with what is not in it. A project-specific policy is bound once. There is no renewal at which good behaviour is repaid, and no promise from anyone that sharing site data moves a premium or a deductible on this project. Pikt does not make that promise and is not in a position to.

What can be said is this. The market has already conceded the point on single perils: in the United States, Builders Risk insurers have offered deductible and premium value in exchange for continuous site data on water. The mechanism exists; what remains is whether it extends to the rest of the site, and that is the market's decision. More concretely and more immediately, where progress updates are already owed to insurers, they can be assembled from the site's own data rather than written by hand - fewer reports, more to show. And the site gets the risk engineer's attention directed to where it has changed, rather than spread evenly across a site that does not wait. The expertise was never the constraint. The time to apply it was.

"Once they can see, they will hold us to everything"

An observed site is a site where every deviation is visible: the hot-works permit that lapsed, the fire door propped open, the materials stored where they should not be. The objection is that visibility raises the standard.

It does, and there is no point pretending otherwise. What changes is when the deviation is found and by whom. Found by the risk engineer this week, it is a conversation and a fix. Found by an adjuster after the fire, it is a very different conversation, and the same camera would have shown it either way. The engineer is not there to catch the site out. The engineer is there so the site does not get caught out.

The common thread

The objections that carry the most weight assume the reader is someone who prices or settles - someone whose interest in the record begins when something has gone wrong. That is the reader the market has trained contractors to expect, and the reluctance is a rational response to it.

The reader Pikt builds for is the one who looks before anything has gone wrong, and whose fee already pays for looking. Put the record in front of that function and the objections do not disappear, but they change shape: from what could be used against us to what could be caught in time.

What insurers cannot see, they price as risk. What can be shown, they can weigh.

See every site, every day. Survey the ones that need it.