Signed, accepted, handed over: not the same thing
Four signatures stand between a crate and a finished floor, and each proves something different. What a delivery note and a handover really record.
9 November 2026 · 15 min read · GranTerra Stone, Hui'an
At the end of every stone job there is a moment when somebody takes a pen out and writes their name on a piece of paper. It happens at the kerb, on a clipboard, in a corridor with a delivery still half unwrapped. It is treated as an administrative step, and it is the single most consequential act in the whole project, because after it happens the argument about who is responsible changes shape permanently.
Most of the trouble around acceptance comes from one mistake: treating all signatures as the same act. They are not. A signature can record that something arrived, that something was seen, or that something was approved, and those are three different claims carrying three different weights. When a job goes wrong at the end, it is usually because one of those was used as if it were another.
The useful thing to understand is not how to pass an inspection. It is what each sheet of paper in the sequence is actually evidence of, so that nobody signs away a right they did not know they had.
A signature can prove three different things
The first is receipt. Something was handed over and counted. The number of crates matches the number on the manifest, the driver has a signature, the vehicle leaves. This claim is about quantity and nothing else. It says nothing about whether the contents are correct, intact or fit for purpose, and the person making it is often not qualified to say.
The second is observation. Somebody looked. They walked the load, opened a crate, turned a piece over, and wrote down what they saw. This is a stronger claim, but it is bounded by how much was actually examined, and by whether the examination happened in conditions where a defect could be seen at all.
The third is approval. The work is accepted, the standard has been met, and the responsibility for what happens next has moved. This is the claim that carries money with it, and it is the one that gets made by accident most often.
The failure mode is consistent: an observer signs something that reads like an approver, and the distinction collapses. A site engineer writes “received” on a delivery note because that is what the form asks for, and three weeks later the form is produced in a meeting as evidence that the material was accepted as satisfactory. Nobody lied. The document simply recorded more than the person intended to say.
Four sheets of paper, four different jobs
A stone package normally produces four documents, and they arrive in this order: the delivery note at the kerb, the concealed work record before the stone is closed in, the staged or partial handover as areas finish, and the final handover pack with its list of outstanding items. Each of them has one job, and none of them is a substitute for another.
The delivery note exists to move goods and settle a freight account. It is issued by a transport company and designed around their needs: how many, how heavy, where to, who took it. It has no field for condition and no vocabulary for it. Expecting it to carry quality information is like expecting a taxi receipt to describe the passenger’s health.
The concealed work record is the one that matters most and gets made least often. It covers everything that will be invisible once the next trade arrives: the substrate, the waterproofing, the fixing, the bedding, the reinforcement around an opening. Stone work has an unusual amount of this. A floor is one layer of stone over several layers that nobody will see again.
The staged handover exists because buildings are finished in pieces. A hotel opens a wing while the lobby is still a building site, and the stone in the finished wing needs to be signed off while the rest of the job continues.
The final pack is a different kind of object altogether. It is not a test. It is a transfer of knowledge, and the note on where supply ends and fixing begins sets out what belongs in it, because the boundary between those two things decides who compiles it.
What no pack can do is keep the work in the condition it was accepted in. That is a separate problem with its own timetable, and the first months of a stone floor in use is where the sequence carries on.
The delivery note records a count, not a condition
This is worth dwelling on, because it is where most claims are quietly lost. The document that arrives with a container is a consignment note. Its purpose is to establish that a named quantity of goods was transferred from one party to another at a particular time. Almost every line on it is a number, and the only free-text space is usually reserved for exceptions.
What the receiving clerk can realistically verify in the twenty minutes a driver is willing to wait is whether the crate count matches. Opening every crate, unwrapping every bundle and inspecting every piece is a half-day job for two people with somewhere to lay the material out. That is not what is happening at the gate.
So the signature on that sheet means: the crates are here, and there are the right number of them. If the crates are later found to contain the wrong pieces, or pieces damaged in transit, that signature does not destroy the claim. What destroys the claim is a signature on a form that says the goods were received in good order, because that phrase is an assertion about condition dressed up as a receipt.
The practical difference between those two documents is a printed line. One says received. The other says received in good condition. A buyer who reads the form before signing, and writes “crates counted, contents not examined” in the exceptions box, has preserved everything. It takes four seconds and it costs nothing, and it is the cheapest insurance in the whole project.
The one inspection that cannot be repeated
Most defects in stone work can be looked at again next week. A chip in a tread will still be there. A colour that reads wrong under the lobby lights will read wrong tomorrow. The exception is everything that gets covered, and stone covers a great deal.
There is a window in every wet area, every floor, every cladding panel, every staircase, when the work is complete but not yet closed. The waterproofing is down and the stone is laid over it, and the joint has not been pointed. The substrate is visible at the edges. The fixing is exposed behind the panel. Nothing has been sealed, so nothing has been decided.
That window closes on a schedule that has nothing to do with inspection. The next trade arrives because they are booked, not because the stone is ready to be examined. Once the joint is filled and the skirting goes on, whatever happened underneath is permanent, and the only remaining remedy is demolition.
This is why the concealed record is worth more than its paperwork suggests. It is not a formality that satisfies a building control process. It is the only chance anyone will ever have to see the layers that carry the stone, and the cost of missing it is not a defect. It is the impossibility of knowing whether there is a defect until something fails.
Photographs are the practical answer. Not a wide shot of a finished room, but close, with something in frame for scale, taken while the layer is still open. The waterproofing note at the interface under the stone goes into what to look for. What belongs here is the timing: those photographs have to be taken before the person who took them goes home, because the window does not reopen.
Acceptance is judged by the surface, not by the piece
Here is a distinction that causes a surprising amount of argument. Every individual piece in a stone floor can be perfectly made, perfectly square, correctly finished, free of any flaw that a reasonable person would object to, and the floor can still fail acceptance. What is being judged is not the pieces. It is the face they make together.
The things that fail at that scale are not visible on a single unit. A slight difference in the direction of the grain. A joint that opens by a millimetre more in one run than another. A colour that is consistent within each crate but drifts across the room. A polish that catches the light differently on pieces finished in different sessions.
None of those can be assessed at the bench, and none of them can be fixed by replacing one piece. A single pale tile in a dark field is a repair. A field that reads unevenly because the pieces came from two bundles is not, and the reason is that there is nothing to remove: every piece is doing what it was asked to do.
This is why acceptance has to be done on the installed surface, at the distance people will actually look at it, in the light it will actually be seen in. The note on why the same stone looks different in the same room deals with the light half of that problem. The other half is arithmetic: a sample is one piece, and a room is the relationship between hundreds.
It also explains why a snagging walk that happens from a standing position, at midday, with the lights off, finds almost nothing. The surface that gets judged is the one seen from the doorway at night with the downlights on, and it is a different surface.
The punch list is a document about money
Snagging lists get described as quality tools. They are not. A list of outstanding items is the mechanism by which a payment is released or held, and everything about how it is written follows from that.
This is why the list has a shape. Items are numbered, each one is attributed to a location, and each is closed individually. A list that says “several items in the lobby” cannot be closed, because there is no way to demonstrate that the thing described has been done. A list with twelve numbered items can be closed eleven times and argued about once, which is exactly what a payment schedule needs.
It also explains why the same list is fought over from both sides. The party holding the pen has an interest in a long list, because every open item is leverage. The party being inspected has an interest in a short one, because the same items are the difference between being paid and waiting. Neither of those interests is dishonest, and both of them distort what gets written down.
The way out is not to argue about the list. It is to have agreed, before the work started, what would be on it. A standard that was written down at the order stage produces a list of items that fail it. A standard that is invented at handover produces a list of things somebody dislikes. Those two documents look identical and behave nothing alike, and the problem of judging natural stone is the reason the first one has to be explicit rather than assumed.
Signing early and signing late cost opposite things
Both mistakes are made, and they are not symmetrical. Signing too early gives away a claim that cannot be recovered. Signing too late holds up a payment, a programme and a set of other people’s trades, and it eventually forces a signature anyway, under worse conditions.
Signing early is the common one, and it usually happens for a human reason. The driver wants to leave. The site manager is being pressured to clear the gate. The person with authority is not on site and the person who is wants to be helpful. In each case a signature goes on a sheet, and the sheet says something broader than the situation warranted.
What is lost is subtle. It is not the right to complain, which usually survives. It is the right to have the complaint treated as the supplier’s problem rather than as a negotiation. Once a document exists saying the goods were accepted, every later claim starts from a weaker position and has to argue its way back.
Signing late fails differently. A project that will not accept anything until everything is perfect ends up accepting everything at once, in a hurry, with the opening date already visible in the calendar. The final walk happens in an afternoon instead of over a fortnight, the list is written from memory, and the items that needed a decision get waved through because there is no longer time to decide them.
The two errors have the same root. Both come from treating acceptance as a single event at the end rather than a sequence of records made while the work is still open. A project that signs small and often signs well.
Who is actually allowed to sign
Authority is the quiet failure in this whole sequence. The person present at the moment a decision is needed is frequently not the person entitled to make it, and the gap gets filled by whoever is closest to the clipboard.
Four roles turn up on a stone job. The receiver takes delivery, and their authority extends to the count. The site engineer or foreman accepts workmanship, and their authority extends to whether the installation matches the drawing. The designer or client representative accepts appearance, and only they can say whether a surface is acceptable to look at. The commercial party accepts the money consequences, and only they can close a list.
A signature from one of those four, on a document belonging to another, produces the argument that ends projects badly. A foreman signing an appearance approval is not being careless. They are being asked a question outside their role, by someone who needs an answer, at a moment when the correct answer is that they are not the right person.
The fix is procedural and cheap. Name the four roles at the start of the job, say which document each one signs, and print the role next to the signature line. A form that asks for a name and nothing else invites anyone to sign anything.
What we can give you before it leaves
Nothing on this page is an argument that acceptance should be softer. It is an argument that most of the evidence should exist before the crate is opened, and that is the part we can control from here.
Each consignment leaves with a sheet matching every marked piece to a place on the drawing, so that whoever opens the crate can settle within a minute whether the contents are right, rather than finding out two days later with the material already spread across three floors. Which piece carries which mark is dealt with in how the number gets onto the crate, and the reason it matters at acceptance is that it turns a pile of stone into a set with a known shape.
We photograph the packed crates as they close, which is the only durable record of what condition the material was in when it left. We state the batch the pieces came from, which is what makes a later colour argument possible to settle rather than endless. We hold a reference piece from the run, because a physical object ends certain conversations faster than any document.
What we cannot do is inspect the installed surface. That happens in a building we will never visit, under lighting we did not choose, at a distance we cannot see. The boundary is the same one described in protecting stone once it is on site, and it is worth being blunt about it rather than letting the handover pack imply a warranty on workmanship that nobody here performed.
Three questions to settle before the first crate lands
The first is what the acceptance standard is, stated as something a person could check. Not “good quality”, which is a feeling, but the thing that will be looked at, from where, in what light, against what reference. If that cannot be written in three lines, it will be invented at handover by whoever is holding the pen.
The second is who signs each document and what their signature is evidence of. This is a two-minute conversation at the start and a two-month argument at the end. It belongs in the same pack as the drawings, and it belongs there before anyone has an interest in the answer.
The third is what happens to the material that fails. A rejected piece needs somewhere to go, somebody to authorise its replacement, and a decision about whether the replacement has to come from the same batch. Those three questions have very different answers depending on when they are asked, and the notes on what can be put back and on holding spare material exist because the answers get expensive quickly.
None of this is difficult to arrange, and almost none of it happens by default. What happens by default is a pen, a clipboard, and a form designed for a freight company. A drawing and a short message to the sales desk are enough to start, and naming then what will be checked on arrival is worth more than it looks, because a supplier who knows the test can cut for it.
A signature is a small thing to give away and a hard thing to get back. The four sheets of paper are not bureaucracy. They are the only record of what anyone actually saw, and a project that treats them as paperwork will spend the last month arguing about things that were visible in the first week and nobody wrote down.
Every figure on this page comes from our own order sheets. Where something is a trade norm rather than a number we can hold, we say so — send a spec and we will confirm it against a real block before you commit.
More from the yard
Send the drawing. We do the rest.
Reply within one working day, in English or Chinese. Send a spec, a drawing or just a photo of the sample you are matching — we will tell you honestly whether we can hold it.
MOQ 1 × 20FT · FOB XIAMEN · SAMPLES IN 7 DAYS