Why a hotel hears footsteps from the floor above
Impact sound travels through the structure, not the air. Why a hard stone floor changes what a guest hears, and where the path actually runs.
7 November 2026 · 15 min read · GranTerra Stone, Hui'an
A guest on the third floor calls reception late at night and says the people above will not stop walking. Reception sends somebody to the fourth floor. The fourth floor is quiet. The noise continues. It is coming from the fifth floor, or from the service corridor at the end of the third, or from a plant room two levels down where a pump is running against a pipe clipped to the slab.
The guest is not mistaken. They are reporting the only thing they can locate, which is where the sound arrived, not where it began. That gap between arrival and origin is the whole subject of this page, and it is the reason stone floors get blamed for a problem they are usually only a small part of.
The complaint never names the right floor
In a hotel, a floor complaint is a description of an experience written by somebody who has no access to the building’s drawings. They know a direction and a rough sense of distance. Everything else is inference, and the inference is almost always wrong in the same way: people assume the sound came from directly above, because that is the only place they can point at.
A hard floor makes this worse without causing it. Carpet and rubber absorb a portion of the energy at the point of contact and spread what is left over a longer time, in a lower register. The result is a dull sound with no direction. Stone absorbs almost nothing at the surface, so the energy arrives clean and intact, and a clean sound can be followed. A guest who can follow a sound will also guess where it came from, and a confident guess from the third floor is how the fourth floor gets inspected at midnight.
This is why the first useful move on a complaint is not to look at the floor above. It is to establish the path.
Airborne sound and impact sound fail in opposite directions
Two kinds of sound, and they are so different in mechanism that almost no single measure addresses both.
Airborne sound — a television, a conversation, a door closing — travels through air and then through whatever gap it can find. Under a door, around a socket, through a duct that was never sealed, along a service penetration where the sleeve was left open. It fails at the joints. Seal the joints properly and the level drops, sometimes dramatically. This is well understood in hotels, which is why doors, seals and penetration details usually get attention.
Impact sound — a heel, a chair leg, a suitcase wheel, a dropped glass — never travels through air at all. It enters the structure directly at the point of contact, travels through the screed, the slab, the steel, and leaves the structure somewhere else entirely as noise. It fails at the connections. There is no gap to seal, because the path is not a gap. The path is the material.
The practical consequence is that a hotel can spend a great deal on acoustic doors, sealed risers and solid walls, and still have a floor complaint. And a hotel with thinner walls and hollow doors can have none, because nobody in it wears hard shoes on a hard floor above a light sleeper. The two problems are being solved by different trades, at different stages, from different drawings, and the one that gets the budget is usually the one with a product to sell.
A hard floor does not make it louder, it makes it clearer
Stone does not create impact sound. The heel does. What the stone changes is the shape of what arrives.
Under a soft floor, the energy of a footstep is partly absorbed and the remainder is stretched over a longer time and a lower frequency. It arrives as a thud. The ear reads a thud as weight — something is being put down. It does not read it as a person, and it does not read it as a sequence of events.
Under stone, nothing is absorbed at the surface and the energy leaves the point of contact in a very short time at a higher frequency. It arrives as a tap or a click. The ear reads a tap as an event, and events carry identity: a heel rather than a shoe, a chair being pulled rather than a chair being sat on, a trolley crossing a threshold, a child running, somebody stopping to take a call.
This is the part that surprises people who have only looked at measured levels. The complaint in a stone-floored hotel is rarely that the floor above is too loud. It is that the guest can tell what the person above is doing. Nobody lies awake over a thud at midnight. They lie awake because they can follow somebody’s evening from one side of the room to the other, and following it is what keeps them awake.
Volume responds to mass and to absorption. Clarity responds to decoupling — breaking the path so the impact never reaches the structure as a clean impulse. They are separate purchases, and buying one does not buy the other.
The path is longer than the slab
A floor is drawn as a stack: structure, then whatever sits under the stone, then the stone. That stack controls how much energy gets in. It does not control where the energy goes once it is in.
The routes out of a floor are flanking paths, and in a hotel they are usually the reason a carefully built floor still fails. None of them are on the stone drawing. All of them are on somebody else’s.
- The skirting. A stone or timber skirting fixed hard against both the floor and the wall is a bridge. It takes vibration out of the floor and puts it into the wall, and that wall is the ceiling of the room below.
- The door threshold. A threshold that runs continuously from a carpeted corridor into a stone bathroom joins two floor build-ups that were designed separately, by two people, for two different purposes.
- The service shaft. Risers and ducts are often built as one continuous enclosure through every floor of the building. Sound uses them the way a chimney uses a fire.
- The bathroom floor. A recessed slab for drainage means the stone sits at a different level on a different build-up, and the two conditions meet at the door — which is exactly where a hard material begins.
- The facade connection. Where a stone floor runs into a curtain wall, the floor and the facade can behave as one surface, and the facade is connected to every other floor above.
Each of these is a detail of a few lines on a drawing. Each one is owned by a different trade. In a dispute after handover, each one is owned by nobody, because the person who drew the floor is not the person who drew the skirting, and no single drawing shows the two of them meeting.
The joint details between stone and other materials are covered in more depth in stone meeting metal, timber and glass, which is the neighbouring problem: what happens where two materials are fixed to each other. This page is about what happens when they transmit to each other.
What the stone layer can actually decide
Given how much of the outcome is settled elsewhere, it is fair to ask what a stone order is really deciding. Three things, and none of them are small.
Whether the stone is bonded or laid. A piece bedded in a full mortar or adhesive bed is in hard contact with the structure across its entire underside. Every square of that contact transfers. A piece laid on a resilient system is in contact only where the system is designed to allow it. The stone is identical in both cases. The floor is not.
Whether the floor is continuous through a doorway. A threshold detailed as a break, with the two sides of the floor separated, closes a flanking path that is otherwise wide open. A threshold detailed as one continuous surface opens it. This is a decision made on a detail drawing, and it costs a joint.
Whether the edge is free. A skirting that sits on a compressible layer rather than hard against the wall does not bridge. A stone architrave that grips the wall on both sides does. The difference is invisible after the wall is painted and expensive to change once it is.
Weight plays a part, and it is often misunderstood. A heavy stone floor has more mass above the structure, and mass resists the movement of a structure that has been struck. That helps with the low, dull component and does nothing at all for the sharp one, because the sharp one is a function of contact, not of weight. Adding mass to a bonded floor makes it heavier and usually makes it sound more solid without making it quieter.
A stone stair is a drum
Stairs concentrate everything that is difficult about a hard floor into a small area.
A stone tread that sits in direct contact with a structural stringer or a concrete flight is a plate in hard contact with a large resonant body, and that body is connected to two floors at once. Every footstep on the flight makes the whole flight respond. In a hotel the stair is also usually the escape stair: a continuous vertical shaft with a hard landing on every level, which is the geometry of a very effective loudspeaker.
The instinct is to specify a thicker tread. A thicker tread lowers the pitch and adds mass, which some people prefer and nobody mistakes for quiet. The change that actually matters is that the tread and the structure stop being one object: a tread separated from the flight by a resilient layer, and detailed so the nosing does not close that separation at the edge.
Setting out and edge profiles on stairs are covered separately in stone stair treads. The acoustic point is narrower: a tread that is isolated at its bed but gripped at its nosing is not isolated.
The corridor, and what the carpet was doing
Guest rooms are tested. Corridors are not, and they should be.
A corridor is the longest continuous floor surface on a guest floor, and it is shared by rooms that are otherwise separated by walls. It is the one element that joins every room on the level. A hard corridor laid continuously through every doorway, finishing hard against every architrave, is a plate running the length of the building, and both ends of it are connected to the stair and the lift lobby.
This is where the stone decision and the acoustic decision usually part company. The corridor is specified by an interior designer, and it is judged at three in the afternoon by how it looks. It is heard at three in the morning by somebody trying to sleep, when a housekeeping trolley goes past, and by then it is nobody’s line item.
Renovations are where this becomes visible. A hotel replaces a carpeted corridor with stone for reasons that are all sound: cleaning, lifespan, wear at the trolley route, and the fact that stone still looks like itself after a decade. What the carpet was doing was invisible and unpaid for. It was absorbing a portion of every impact at the surface, and its underlay was separating the walking surface from the slab at every point. Neither of those functions appears on the finish schedule, so neither is replaced when the carpet goes.
The finished corridor is then better looking, easier to clean, and audibly different, and nobody can point to the decision that caused it, because the decision was to remove something that had never been asked to justify itself.
A mock-up will not tell you
A sample room is the right way to approve stone. It is the wrong way to approve a floor.
Performance between two rooms is a property of the pair: the floor, the ceiling below, the walls between, and the connections at all four edges. A single sample room has no pair. You can stand in it, listen, and hear nothing at all, because there is nothing above it to hear.
What a mock-up can do is register the details that will be repeated. If the sample room is built with the threshold detail, the skirting detail and the bathroom step that the rest of the building will get, then the mock-up has tested drawings. If it was built as a one-off to get finishes approved, it has tested finishes and nothing else.
The question worth asking of any mock-up is not whether it looks right. It is which of the details inside it are identical to the ones that will be built again several hundred times.
If the contract has no number, nobody can fail it
Impact sound between rooms is measured with a source in one room and a receiver in the one below, and the measurement needs three things agreed in advance: a defined pair of rooms, a defined state of those rooms, and a method.
If the specification carries a performance figure, four things should be written beside it:
- which rooms form the test pair, and whether they are furnished or empty when tested
- whether the test happens before or after the carpet, curtains and bed are installed
- who carries it out, and who pays for a repeat if it fails
- what happens if it fails — a remedy, a credit, or nothing at all
If the specification carries no figure, the floor cannot fail and it also cannot pass. It will be judged on the only thing left, which is whether anybody complains. A complaint is not a test. It is a sample of the people who happened to be awake, in the rooms that happened to be occupied, on the nights that happened to be quiet enough to notice.
This asymmetry runs through most stone disputes. The material arrives measurable and leaves unmeasured, and everything between those two points is opinion. The layer under the stone has the same problem in a different form, which is why the waterproofing under the stone is argued about in exactly the same way — and why the argument there is settled by water appearing, which at least is unambiguous.
What a stone yard can and cannot do
Being clear about the boundary is worth more than a claim.
We cut stone. We do not design floor build-ups, we do not install resilient layers, and we cannot make a hard floor quiet. What we can do is cut the stone so that it does not close a door that somebody else is trying to keep open.
- We cut to your drawing, including the breaks. If the drawing shows a threshold that separates two floor areas, the stone arrives as two pieces with the joint where you put it, not as one continuous run that has to be cut on site and then re-edged.
- We can supply pieces for a system rather than for a bed. If the stone is going onto a raised or isolated build-up, tell us what the system needs at the edges and we will finish the edges accordingly rather than assuming they will be buried.
- We keep the block record with the order. A hotel is never finished, it is only opened: there is always a phase two, a refurbishment, a floor that gets redone after a leak. Matching the second batch to the first is a stone problem, and it is ours.
- We do not quote an acoustic figure. The stone does not have one, and a number on our quotation would be a number about somebody else’s work. Any supplier who offers one is describing the system they are selling, not the slab they are shipping.
If a floor is being specified and the acoustic question is still open, the useful thing to send is the drawing and the build-up, not a target figure. We will cut what the drawing describes, and we will say so if a detail cannot be made to work as drawn — which in our experience is a more useful reply than a promise.
The part that is decided before the stone arrives
Almost everything that determines whether a guest hears footsteps is settled before a single slab is cut: the structure, the resilient layer, the ceiling below, the threshold detail, the skirting detail, and whether anybody wrote a number into the specification.
Stone arrives last and is blamed first. That is not unfair so much as structural — it is the only part of the assembly a guest can see, and the only part with a name they know. The floor that is heard is never the floor that is blamed.
What can be done about it is narrow and specific. Detail the breaks. Keep the stone out of hard contact wherever the drawing allows it. Test a pair of rooms rather than one. Write the number down, and write down what happens if it is missed.
If you are working on a project where the floor is stone and the acoustic question is unresolved, send the drawing. We will tell you what the stone can be cut to do, and what it will not do no matter how it is cut.
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