Ask three people whether travertine belongs in a shower and you get three confident answers. The stone yard says yes, the contractor says it stains, the supplier says it depends which travertine. All three are describing something real, because none of them is talking about the same thing.

What decides the outcome is not the stone. It is five items sitting between the block and the finished room: cut and fill, sealer chemistry, fall, cleaning protocol, and whether anyone agreed in writing who maintains it. Five right, and travertine is among the most durable wet-area materials available. Two wrong, and the dark patches arrive before the defects period ends. The rest of this covers the shower environment specifically — floor, walls, thresholds, niches and monolithic basins — written for specification rather than for decorating.

Can you use travertine in a shower?

Yes. Travertine is a freshwater limestone laid down in hot springs, used in bathing architecture for some two thousand years — most of that without a chemical sealer near it. What it is, is open: the voids are the reason anyone specifies it, and those voids are continuous pathways for water and soap. Every problem below follows from that fact, and every solution routes around it rather than through it.

Four conditions make it work, all four belonging on a drawing rather than in a site meeting:

1. Filled pore structure, honed finish for anything walked on. Unfilled travertine collects soap solids in its voids that cannot be scrubbed out.
2. Penetrating sealer, onto a genuinely dry slab before installation and again after grouting has cured.
3. A formed slope — machined into the top face — of at least 1.8°, or about 1:32.
4. A written maintenance schedule with a reseal test method and a named responsible party. "Reseal periodically" is not a requirement.

Drop any one and the answer shifts from yes to "yes, but it will be somebody's problem within about eighteen months".

How travertine fails in a wet area

Failures arrive in a recognisable order, and where they appear tells you why. This timeline is field observation rather than laboratory result Estimate — verify, but anyone who has walked onto enough of these rooms will recognise it.

First three months. A faint patch, usually near the drain or in the corner nearest the shower head. Water that should have left is standing, carrying minerals and soap solids into pores that are still open.

Three to twelve months. The edges darken. Known as picture framing, this is moisture travelling along the back face and migrating toward the perimeter where evaporation is fastest. It reads as a cleaning problem and is not one; cleaning harder accelerates it.

Twelve months onward. Soap film colonises the same damp spots and goes grey, then pink. That is biofilm inside the pore structure. It will not scrub out, so it gets chemically removed — which on a calcareous stone usually means more damage.

Three years. Either nothing changed because somebody maintained it, or the room looks twenty years older than the rest of the house.

Every stage shares one cause: water did not leave, or it left by evaporating through the stone rather than running off it. Both are solved upstream of whatever sealer gets applied.

Root cause: the pores are the material

Travertine precipitates out of mineral water, and gas and plant matter leave voids in it as it forms. Those voids are the aesthetic argument for the stone, so you cannot solve wet-area performance by wishing porosity away — and attempts to do exactly that cause more failures than the porosity does.

The water arrives from two sides

Everyone seals the face. Very few people think about the back. A slab bedded onto mortar over a screed takes in moisture continuously through its reverse face, and a hotel shower runs a dozen or more cycles a week. Seal one face and you have dealt with roughly half the load — the half nobody notices.

A film-forming sealer turns that half into a trap

Acrylics and urethanes sit on top as a continuous layer. Vapour driven out of the substrate reaches the interface, cannot pass, condenses, and the coating clouds, blushes, then lifts. It presents as patchy loss of gloss, gets read as "the sealer wore off", and gets another coat. That buys about six weeks.

A sealer is not waterproofing

The most expensive single misunderstanding in stone specification. A penetrating sealer changes the rate of uptake; it buys time for water to drain or evaporate. It does not stop water. The barrier is the membrane behind the stone — its own line item, with its own compatibility chain: membrane to adhesive, adhesive to stone, stone to sealer.

Is travertine slippery when wet?

Maybe. That is not a dodge — it depends which finish the project selected, and on a shower floor polished is the wrong answer. The only defensible way to write this is to name the test. In North America, the working figure for an interior level floor walked on when wet is a wet dynamic coefficient of friction of at least 0.42 measured to ANSI A326.3 Standard. Three things matter: it applies to level floors, it is measured on the finish being supplied, and honed and polished versions of the same block will not produce the same value.

Europe answers the question with a pendulum test value — to EN 14231, or in the UK to the BS 7976 series — while German and much of Central European practice uses ramp classification: DIN 51130 for shod traffic under oil, DIN 51097 for wet barefoot areas. A shower floor is a barefoot surface, which is the part people get wrong. An R-rating is not the correct axis for it.

These do not convert. A wet DCOF number and a pendulum value come off different machines on different principles, and any table claiming to translate between them should be treated with suspicion. A tender asking for "0.42 DCOF or equivalent" has effectively said nothing. Name the method applicable to the destination market and require a report for the finish supplied.

One finish note that saves arguments: specify honed for every surface in the wet zone. Polishing leaves a mirror-flat contact that behaves very differently wet and shows every acid etch.

Slope: the cheapest thing you can get wrong

Standing water produces every symptom listed earlier, and the most effective control is the one most often left to whoever happens to be laying the screed that week.

There are two ways to build a fall: form it in the substrate, with a wet-laid mortar bed or a pre-formed tray, and set stone on top; or machine it into the top face. For a monolithic unit, machining wins on every count — no dependence on site workmanship, no extra build-up thickness, and a result verifiable in the factory before anything ships.

The angle conversions are pure arithmetic — angle equals arctan of gradient — so anyone can reproduce them. The fall itself is small enough that a spirit level is no help: at 1:32 over a 900 mm run the drop is 28 mm, obvious in section and undetectable underfoot. Enough gradient to move water to the drain, not enough for anyone to perceive it.

The critical location is never the middle of the pan. It is the last hundred millimetres before the drain flange, and the corners, where the field gets cut and the fall has to survive that cut. Ask for the shop drawing at the drain before approving anything.

Penetrating or film-forming

Penetrating. In a shower it is not really a debate, and the reason is not the one most people give.

Two further points decide more than the product choice does. First, a penetrating sealer must go onto stone that is genuinely dry: installed in week three of the build, the slab still carries mortar moisture, and sealing over it locks that moisture in — which is how you get picture framing without ever using a film product. Second, some penetrating sealers darken the stone, and honed and polished faces on the same piece will not darken equally. Test on an offcut retained from the actual block.

Cleaners that ruin travertine in six months

Travertine is calcium carbonate. Calcium carbonate dissolves in acid. Those two sentences explain most of the damaged stone anyone walks onto.

• Acids of every kind: vinegar and acetic cleaners, plus hydrochloric, phosphoric and sulfamic products — which is what most limescale removers and plenty of daily shower sprays are made of. All etch permanently, worse on polished faces.
• Acetoxy-cure silicone against a stone edge, releasing acetic acid as it cures and etching the very arris it is protecting.
• High-pH strippers and heavy bleach, degrading the sealer long before they shift whatever stain prompted them.

An organisational failure hides in here as well. The quantity surveyor assumes the maintenance contractor checked, who assumes the spec covered it, who assumes the material arrived with instructions. Put two approved cleaners in the O&M file by brand, write "no acidic cleaners, no steam cleaning, no pressure washing" into the handover pack, and specify a squeegee hook. That last one is not a joke: over ten years it does more for the appearance of travertine than any cleaning product on the market.

What maintenance has to happen

Maintenance reduces to a short schedule, and one test that replaces guessing.

Drop test: put a teaspoon of water on the surface, wait ten minutes, wipe. If the stone darkened underneath, reseal now. Ninety seconds of somebody's time, and it turns a schedule into a measurement.

Six to twelve months is the usual interval in a shower used daily Estimate — verify, and the range is wide for a reason. A guest bathroom used twice a week may go years; a hotel bathroom turned round twelve times a day wants attention at six. Usage decides, not the date. Closed rooms are not exempt — a resort reopening after a closed season should be inspected and resealed before guests rather than after complaints.

How to specify travertine for a wet area

Six lines, written as requirements. Each exists because its absence has caused a problem somewhere.

1. Name the stone, its cut and its fill. "Cross-cut travertine, honed and resin-filled", not "travertine". Prohibit cement-filled material on shower floors — the fill absorbs faster than the stone and telegraphs every drop.
2. Require test reports with the sample, not with the brochure. Absorption to EN 13755 or ASTM C97, plus the wet slip test named for your market run on the finish supplied.
3. Name the membrane and adhesive by product and state the compatibility chain: membrane, adhesive, stone, sealer. Four products, three interfaces, three places for nobody to be responsible.
4. Declare the fall in the units used on the drawing, give the ratio and resulting drop over the actual run, and dimension the drain. Stated in one unit only, it invites an argument nobody wins.
5. Specify the sealing protocol. Penetrating only, onto a dry slab and again after grouting has cured, with product name, two coats, coverage rate and cure time. Require a retained sealed sample — "sealed on completion" is unenforceable.
6. Write the handover pack. Two approved cleaners by brand, acid explicitly prohibited, the reseal test described, and the responsible party named by role rather than implied.

Two cheaper additions worth having: epoxy grout wherever joints sit in standing water, because cement grout dust packs into open pores and never fully buffs out of honed stone; and neutral-cure silicone at every joint touching stone, with bond breaker or backer rod wherever movement is expected. Neither costs anything at specification stage. Both cost a great deal at rectification.

How Bergoto details monolithic travertine

We fabricate travertine vanity tops and monolithic basins, so the wet-area problem gets solved at the machine rather than handed to whoever is on site that week.

• The fall is cut, not screeded. Our monolithic basins carry a machined formed slope of 1.8° (about 1:32, or 3.12%) to the drain, dimensioned on the shop drawing and checked before the piece leaves the factory, so drainage does not depend on how the substrate was laid. Bergoto detail — verify
• The sealer is named on the drawing. We seal before machining is complete, ship a sealed offcut retained from the same block with each order, and put the product name into the documentation — so the maintenance contractor in year two is not guessing from a photograph.
• Cabinet, top and basin are fabricated together. One continuous fall line across the run, instead of two packages meeting at a joint where nobody dimensioned which one is higher.

Bergoto is an OEM and ODM manufacturer of bathroom and kitchen products, operating to ISO 9001 quality management and BSCI social compliance, with roughly thirty years in the industry, ten production lines and annual capacity of about 500,000 pieces. Fifteen designers work on product development and project documentation in-house. Project supply is what the operation is built for: drawings, samples and contract documentation before you commit, repeatability at unit level across a few hundred rooms.

Weighing travertine against marble and engineered stone for a specific wet area? Send the room schedule and drain positions. We will say which of the three suits that application — engineered stone wins on some briefs, and we would rather lose a line item than inherit a failure.

Frequently asked questions

About these figures

Material numbers get copied into specifications and defended in meetings, so here is what each rests on.

• Standard — the 0.42 wet DCOF figure refers to level interior floors expected to be walked upon when wet, measured to ANSI A326.3. Slip requirements referencing BS 7976, EN 14231, DIN 51130 and DIN 51097, and absorption testing to EN 13755 or ASTM C97, must be checked against the edition current in your market and read from the standard text, not a summary. Other floor conditions use other criteria.
• Standard — fall ratios such as 1/4 inch per foot appear widely in North American tile standards for shower receptors, but which minimum applies depends on edition and local code.
• Arithmetic, not estimate. Table 2 conversions follow from angle equals arctan of gradient.
• Bergoto detail — verify — the 1.8° formed slope is our standard machining detail; confirm against the cut sheet for a specific unit before quoting it in a contract document.
• Estimate — verify — failure timelines, the 6–12 month reseal range, the ten-minute drop test threshold and the one to three year re-coat cycle are field indications, varying with pore structure, block selection, sealer product and how hard the room is used. Measure your own.

None of this replaces a project-specific waterproofing design. Have the membrane build-up, movement joints and drainage falls confirmed by the responsible designer before fabrication begins — which is another way of saying before it becomes somebody else's problem.