What Is Dishwasher Safe: A Claim With A Test Behind It
Dishwasher safe means an item survived a defined test, in a defined machine, for a defined number of cycles.
It is not a property of a material. Ceramic is not dishwasher safe. A glaze is not dishwasher safe. A specific plate, with a specific decoration, passed a specific test.
That difference sounds academic until a container comes back. The buyer wrote “dishwasher safe” on the order. The supplier agreed. Then a customer runs the plates through a hot programme for a year and the rim pattern goes chalky.
Nobody broke the agreement, because “dishwasher safe” was never a number. This article turns it into one.
The Answer In One Line
“Dishwasher safe” is a claim that a test result supports. The test is a dishwashing resistance test, and it is scored, not passed by opinion.
The European reference method is EN 12875-1, Mechanical dishwashing resistance of utensils — Part 1: Reference test method for domestic articles.
The 2005 edition is being replaced. The current draft of the test method is prEN 12875-1:2024, written by CEN/TC 194, the committee for utensils in contact with food.
Three sentences from that draft matter more than the rest. The results “serve for comparison purposes”. They give “an indication of resistance only under the standard conditions of this test”. And the test itself is “time consuming and expensive”.
Read those again. A comparison. An indication. Under stated conditions. That is the whole promise behind the phrase on the box.
What The Words Promise
There are two phrases in the trade and only one of them is defined.
Dishwasher safe is tied to the standard. One seller that publishes its own position puts it plainly: appearance and usability must stay unchanged after many cycles.
A seller that publishes its cycle count claims 125 cycles for one bowl. That is its number for its product, not a rule for the industry.
Dishwasher proof has no such backing. The same page says it is not a defined term at all. Anything that survives a wash can use it.
That gap is where returns live. A supplier can write “dishwasher proof” on a spec sheet, mean “it came out in one piece”, and be telling the truth.
The standard does not rescue a buyer by itself. It only helps if the report is asked for, and read.
The Three Things One Cycle Does
A cycle is not a wash. It is three attacks in sequence, and each one picks a different target.
Heat. Hot water and a hot drying phase expand the body. A glaze, a print or a coating expands at a different rate.
The difference is small per cycle. Repeated a few hundred times, the layer at the interface starts to craze and lift.
Alkali. Dishwasher detergent is built to attack food residue, and it does not stop at the boundary of the item.
Alkali strips organic binders. It dulls soft metals. It works under a decoration that was never fired into the glaze.
Force. Water leaves a spray arm at speed and carries grit from the load with it. A loose grain of rice becomes a small abrasive.
That is why a print on the inside of a bowl fails before the print on the outside. More jets, more food contact, more cycles.
The three attacks explain the hand-wash list a cleaning body publishes. Aluminium, cast iron, hand-painted china, crystal. Each one loses something different.
The hand-wash list from a cleaning body adds one detail worth copying into a spec. Silver and stainless steel in the same basket can damage the silver. The two metals touching set up a small current.
That is a design rule, not a care tip. A cutlery set that mixes both metals in one basket will fail in a way no coating test catches.
What The Test Actually Does
The reference method is a real dishwasher, run again and again, on a set of samples, with the parameters held.
The draft lists what it controls. The water. The cleaning and rinse agents. The features of the test machine. The specimens. The loading pattern. The cycle. The number of cycles.
Then it controls how the result is read. Minor deposits are set aside first, so a hard-water film is not scored as damage. Inspection conditions are fixed. So are the criteria.
None of that is a marketing detail. Every one of those parameters changes the outcome, and most of them are chosen by whoever pays for the test.
| Parameter | What the method pins down | What a supplier can still choose |
|---|---|---|
| Water | The grade used for the run | Which lab runs it, and where |
| Detergent | Type and dose, written into the report | A milder detergent than the market uses |
| Rinse aid | Presence and dose | Dose high enough to hide spotting |
| Machine | The features the test machine must have | An older machine with softer jets |
| Loading | Where the samples sit | Top rack only, away from the spray arm |
| Cycles | The reference number for the method | The count written on the marketing claim |
| Inspection | Conditions, and who may inspect | One inspector instead of two |
| Pass rule | The rating scale | Which rating counts as a pass |
The right-hand column is not a loophole in the standard. It is the space between a test result and a claim, and it is where a buyer has to stand.
There is also a faster route, and it carries a trap. Part 4 of the same series is a rapid test for domestic ceramic articles.
A lab report for a decorated mug shows how one of these is run. It used four test pieces plus one control. Each piece was checked at 16 hours, then again at 32 hours.
The same report states the limit in one line. The rapid method “does not define the number of dishwashing cycles that any given product should withstand”.
So a rapid report proves the decoration survived a defined soak. It does not prove the item will survive two years in a household.
The scoring is the part buyers never see. The mug report scores five things: colour, gloss, clouding, resistant deposits and iridescent layers, and other aspects.
Each is rated on three levels. Zero is no visible change. One is the first discernible change. Two is a clearly visible change.
Two inspectors score independently. Their colour judgement has to meet the requirements of EN 12875-2, the part of the series that governs inspection and evaluation.
That is what “dishwasher safe” means in practice. Four samples, two inspectors, five criteria, a stated cycle count, and a number from zero to two.
Now put that next to a line on an order that says “dishwasher safe: yes”. The gap is the whole problem.
Which Materials Survive
No material is safe on its own. What survives is a build: body, surface, decoration and fittings, tested together.
| Build | What fails first | What to ask for |
|---|---|---|
| Ceramic, decoration fired under the glaze | Glaze crazing at the rim | Cycles, then the glaze score |
| Ceramic, decoration fired on top of the glaze | The decoration, by a wide margin | Colour and gloss scores after cycles |
| Glass, clear | Clouding, then etching | The clouding field in the report |
| 18/10 stainless steel | Water spots, then pitting at a weld | Grade of the steel, and the weld finish |
| Anodised aluminium | The anodic layer at edges and threads | Coating test, not a material name |
| Coated cookware | The coating at the rim, where it is thin | Adhesion after cycles, at the rim |
| Polypropylene | Almost nothing; it is the safe plastic | Which polymer, in writing |
| Polycarbonate and polystyrene | Stress cracks, then crazing | An honest “hand wash only” |
| Wood and bamboo | Swelling, then splitting and mould | Do not claim it at all |
| Silver, and gold rims | The thin metal itself | Keep it out of the claim |
Two rows deserve a note. Polypropylene is the one plastic that usually earns the claim, because it takes hot water and alkali without moving much.
Polycarbonate is the opposite. It looks tougher, and it cracks under the stress that moulding and hot water leave in it.
Cast iron also sits on the list, for a different reason. A bare pan loses its seasoning in one hot cycle and starts to rust the same day.
Where a set mixes materials, the weakest part sets the claim. A plate that passes, with a gold rim that does not, is not a dishwasher-safe plate.
That failure is also the easiest to miss at goods-in, because the plate and the rim look like one item until the rim goes dull.
Where Print And Coatings Go First
The order of failure is stable across materials, even when the cycle count is not.
Decoration goes first. Then the glaze or the coating. The body goes last, and often never.
Decoration means anything added on top of a finished surface: a printed rim, a decal, a metallic band, a sprayed logo.
If the decoration was fired into the glaze, it is part of the glass and it behaves like glass. If it was fired on top, it sits in the attack path.
The mug report shows the vocabulary used to record that. Colour fades. Gloss drops. Clouding appears. Resistant deposits and iridescent layers form and will not rinse off.
Each of those is a different defect with a different cause, and the report scores them apart instead of calling the result “faded”.
Coatings fail in their own pattern. An anodised layer goes at a cut edge, a thread or a rivet hole, because those are where the layer is thin or broken.
A non-stick coating fails at the rim first, for the same reason. It is sprayed thin there, and it takes the jet at full force.
Adhesive labels are the quiet one. A glued brand plate can survive the water and let go in the drying phase, and it takes the top coat with it.
That is why a supplier’s promise about the body answers the wrong question. The body was never at risk.
Ask which decoration, applied how, and what the report scored it at. A supplier who can answer that has read its own report.
What To Ask Before You Print The Claim
Six questions separate a supportable claim from a sentence on a box. Each one has a document behind it.
One. Which standard, and which part? EN 12875-1 for the reference method. Part 4 or 5 for a rapid test. A report that names no part is not a test of the claim.
Two. How many cycles? The reference method gives a number. The rapid ceramic method explicitly leaves the number to the buyer. Ask for it in writing.
Three. What was the pass rule? A score of zero on every criterion is a different promise from a score of one on colour. Both can be called a pass.
Four. Which detergent and rinse aid? They are part of the method for a reason. A mild dose flatters a decoration.
Five. Who inspected, and how many? Two independent inspectors are part of the inspection rules. One inspector checking his own run is not the same test.
Six. Which surface was tested? The food contact surface is the one that matters for a plate. A print on the outside base is a different question.
Add a seventh for any metal part. Which stainless grade is on the label decides whether a basket of cutlery pits or not, and it is not covered by a glaze test.
If a supplier answers all seven, put the answers in the order. If it answers three, the claim is decoration too.
Checks At Goods-In And In The Lab
Goods-in cannot run a 125-cycle dishwashing test. It can catch the builds that will never pass one.
Decoration by eye, under one lamp. Look for a print that sits on the surface rather than in it. A fingernail catches an edge that a fired-in decoration does not have.
Gold and metallic bands. Check that the claim on the carton excludes them, or that the report covers them. These are the first parts to dull.
Rim and edges on coated ware. Look for thin coverage where the spray ends. A pale edge on a dark coating is a thin edge.
Welds and rivets in cutlery. A rough weld holds water and starts pitting. Run a thumb along it, then look at it under the lamp.
The polymer name. Ask for PP or PC in writing. Two parts that look alike behave in opposite ways in a hot cycle.
Label adhesion. Press a glued plate and look for a lifted edge. If it lifts dry, it will lift wet.
One destructive sample. Run ten hot cycles in a domestic machine, then compare the sample against a control that never left the carton. That is not a standard test, but it finds the obvious failures before a container ships.
Then have somebody outside the factory repeat the list. A check run by someone who does not work for the factory is the only version of this list that a supplier cannot soften.
Two of these checks need a lab. The rest need a lamp, a thumb and a carton of samples. The sampling table behind a lot check decides how many pieces you have to look at.
Seven Lines For The Order Sheet
Seven lines turn the claim into something a factory can be held to. Each one names a document, not an adjective.
One. The standard. EN 12875-1, or Part 4 for the rapid ceramic route, named with its year.
Two. The cycle count. A number, agreed before the run, written next to the method.
Three. The pass rule. Which criteria are scored, and the highest score accepted on each.
Four. The detergent and rinse aid. The products used, with dose, as they appear in the report.
Five. The surfaces covered. Food contact surface, outer surface, decoration, metallic parts, fittings. Say which are in and which are out.
Six. The report. Issued by a named lab, on the exact build, with the cycle count and the scores visible. Not a certificate with no test data behind it.
Seven. The fallback. If the build changes, the claim is withdrawn until a new report exists. This is the line that protects the buyer after the first price cut.
A supplier can sign all seven on one page. The format that makes them stick is the sheet that turns a claim into a clause, not a chat message.
For a food contact surface, the claim also sits inside a wider set of rules. How a food-contact surface is classed decides which paperwork the item needs before it can be sold at all.
Send me the dishwashing report or the spec line behind your claim. I will check which standard it names, what cycle count it ran, how the scores were read, and which surfaces the test actually covered. No cost, no obligation.
Check My ClaimQuestions Buyers Ask
Is porcelain dishwasher safe?
The body usually is. The decoration often is not. Ask for a report on the exact pattern, and read the colour and gloss scores rather than the headline.
What is the difference between dishwasher safe and dishwasher proof?
Safe is tied to a test standard. Proof is not a defined term, so any supplier can use it for anything that comes out in one piece.
How many cycles should a claim cover?
There is no universal number. One seller claims 125 for its own bowl. What matters is that the number is written down and tested on your build.
Does a rapid test report prove the claim?
Partly. It proves the item survived a defined soak, which is why rapid reports are accepted. It does not set a cycle count, so the promise stays open.
Why does the pattern fade but the plate stays white?
Because the decoration is the thin layer in the attack path. The body is thick, fired and stable. The print is a surface film taking the jet and the alkali.
Can I put stainless and silver in the same basket?
Not if you want the silver to last. Two different metals touching in a wet, salty machine set up a small current that damages the softer one.
Questions & Comments
We read every one and reply within 24 hoursIf a supplier’s dishwashing report is on your desk, send it over. I will read the cycle count and the scores before you print the claim on a carton.