An ingress level is not a feature you add to a product at the end of the specification. It is a set of design decisions taken at the start, and each one of them — a seal line, a port cover, a button treatment, an acoustic mesh — is a tooling item, an inspection item and a validation item. That is why two products with identical electronics can carry different codes, and why a quotation that is lower on unit price can be higher on the whole programme.
So the conclusion, stated first, in the form we would put it in a sourcing note.
Choose the level from the exposure in the channel, in writing, before the model is chosen. Write down where the product will be, for how long, and how often. That single paragraph decides the digit pair, the engineering, the schedule and the wording your listing is allowed to use.
If the exposure in your channel is splash and sweat — rain on a street, sweat on the temples, a workshop floor — the second digit 4 addresses it, and nothing above it does. Buying a higher level because a competitor has one is the most common way a smart glasses programme pays for a claim it never uses.
If your product has a sound port, a microphone port, a charging port or a charging contact, the achievable level is bounded by the acoustics and the openings, not only by the seal. An open-ear audio product cannot be sealed like a camera product, and a supplier who quotes a high level for an open-ear design is quoting a level for a different assembly.
Do not publish a protection claim without a document behind it. If the evidence in your hand is a catalogue annotation rather than a report number, the honest wording is narrower, and a narrower claim is a smaller commercial risk than a wider one that a customer tests in the rain.
IP rating: two digits, and the wrong reading
The code is defined by IEC 60529 as degrees of protection provided by enclosures, and it has two independent parts. The first digit rates protection against solid foreign objects, from 0 to 6. The second digit rates protection against water, from 0 to 9. A letter X in either position means that part was not evaluated at all.
The X is worth dwelling on, because it is an absence rather than a poor score and it is read wrong constantly. An IPX7 product carries no dust opinion whatsoever. Reading it as "seven for everything" flatters the product and misleases the customer, because nobody has an opinion about the dust at all.
Three readings cause most of the trouble in the field, and all three are supplier-side rather than buyer-side.
Treating the digits as a linear scale. They are not, and the jump between 6 and 7 is far larger than the jump between 4 and 5. A second digit of 4 is protection against splashing water from any direction. A 5 is water under low pressure. A 6 is powerful water jets. A 7 is temporary immersion to a stated depth for a stated duration, and it is the first level that addresses a product in a swimming pool — and even it does not address the chemistry of a chlorinated pool. A comparison written as "one digit better" is not a proportional comparison, and a budget built on it is mis-sized.
Reading a 4 as waterproof. It is not immersion, not a shower, not a hose and not sustained rain. A catalogue that says IP54 and a product page that says waterproof are making two different claims, and the second is not supported by the first. In a business-to-business channel this matters less than it does in retail, because your channel buyer will read the claim and compare it against a competitor's claim, and the mismatch shows up as a returns problem rather than as a compliance one.
Treating the code as the whole rating. Underneath the two digits, four rows do the real work and almost none of them appear in supplier material: the tested configuration, the model list on the document, the condition of the unit that actually arrives, and the service life of the seal itself. A rating attaches to an assembly, and an assembly is a much larger set of things than a product name.
The service-life row is the one buyers forget, and it is the one that produces a slow problem. Every seal ages. The rating describes the product as it left the factory, not the product in year three. If the channel is a multi-year programme, the seal's expected life and the warranty period have to be set against each other deliberately rather than inherited from a calendar assumption.

The exposure in your channel obliges you to specify, it does not merely suggest
The most useful thing a buyer can do before meeting a supplier is to write the exposure down in language that a factory can quote against. The trap is describing the environment as a category rather than as a situation, because the level that suits one category is frequently wrong for the next, and it is the difference between the level being a specification and the level being a hope.
The table at the top of this article is the form we would use: one line per channel, one code, one statement of what the code permits, and one wording the listing is allowed to use. Worked examples of the line a channel produces:
- Retail, outdoor clothing, European market. Rain on a street and sweat on the temples, occasional, seasonal. Second digit 4. Wording: protected against splashing water from any direction. Not immersion, not a shower.
- Retail, sports and cycling, three markets. The same exposure plus a dusty trail and a wash-down at the end of a season. Second digit 4 and first digit 5, and the dust requirement has to be stated separately because a 4 in the second position says nothing about solids.
- Institutional, warehouse and workshop, single market. Dust in the air continuously, occasional splashes from a spill, no immersion, no hose. First digit 5 or 6 is what matters here, and the second digit stays at 4.
- Field service and construction. Mud, dust and rain, with the unit in a case that may itself get wet. The configuration question dominates, because the rating attaches to the assembly in the field, which is the case plus the unit.
- Hospitality and front-of-house, shared dock. A charging dock, a shared table, occasional spill. The exposure is mild, and the binding constraint is the dock and the port rather than the weather.
Once those lines exist, the negotiation changes character. A channel served only from environments where the approval already exists is worth a materially better price than one that requires new testing, and that is a decision to make before the quotation rather than after it. The market list becomes a commercial variable.
The second constraint that is specific to this product form factor is the acoustic one. An open-ear platform has a sound port and microphone ports, and they are openings in the enclosure by design. A higher ingress level in an open-ear design is not a bigger gasket; it is an acoustic compromise, because a sealed or meshed port changes the sound the product produces, and often changes it in the direction the product was not tuned for. A camera product with a sealed lens assembly has a different problem — the lens itself is an opening and an optical element, and covering it changes the picture. Both constraints mean the same thing commercially: the achievable level is bounded by the design, so the level a supplier offers tells you about the design they have already chosen, not about what they would build for you.
There is a third cost that is the hardest to see and the most expensive to discover, and it sits behind the variant structure. Every colour, every logo, every private-label derivative and every market-specific label is a configuration question, and the re-validation cost is per configuration rather than per order. If the change touches the enclosure, the seal material, the interface or the frame material, the marking may no longer apply and the configuration has to be assessed again. A four-colour range with a single ingress level is four configurations rather than one, and the cost scales with the range while the unit price scales with the volume. Buyers who treat the colour count as a merchandising decision and discover the re-validation cost afterwards have usually already committed to the artwork.
The practical countermeasure is unglamorous and effective: fix the configuration list before the first purchase order, state which variations are cosmetic and which are structural, and get the structural variations assessed once, up front. A cosmetic variation that changes only the exterior finish can often be covered by a written equivalence from the supplier. A structural one cannot, and asking for the equivalence in advance is the difference between one assessment and four.

From a channel description to a filed document
Verification here is short, and the value is in the order rather than in the effort. Each step assumes the previous one passed, and a failure at any step tells you what the next step would have told you.
Step one: request the report number rather than the report. A number is the only part of a compliance document that a broker, a marketplace or your own compliance team can check later. Ask for the list of numbers with the issuing body and the validity dates, in the body of the email, before the first sample ships. Cost: one email.
Step two: resolve the number in the issuing body's register. Confirm three things: that the model designation on the document matches the model you are ordering, that the document covers the functions you are shipping, and that the validity date has not passed. Compare the model string character by character. Where a family holds several models with similar designations, this is the check that catches a substitution, and a mismatch is a stop condition rather than a query.
Step three: establish the tested configuration. Ask which ports, contacts and interfaces were fitted on the tested sample, and then compare that list against the unit in your hand, physically, before anything else is done with it. Where a charging contact is exposed on the shipped unit and was capped on the tested sample, the marking does not cover the product in your packaging. This is the single most common gap in a private-label programme, and it is found by picking up the sample and looking at it rather than by reading a document.
Step four: run your own test on a sample, on a fixed protocol. The protocol is narrow: the specific failure mode your channel actually has, with the duration, the environment and the pass condition written down before the first reading. A generic spray test is none of these. The protocol also has to name the interface, because the interface is where the failure actually occurs.
Step five: put the criteria into the purchase order. In a category with no universally quoted standard, the contract is the standard. A criterion that lives in an email is not binding; a criterion in the purchase order is an acceptance standard, and a lot that fails it is a lot you can refuse.
The documents to request before a deposit are these, and they are the same list whether the level is low or high:
- The report or declaration number, with the issuing body and the validity dates.
- The standard and its edition, as applied by the issuing body.
- The model list the document names, and whether your model designation is on it.
- The tested configuration: ports, contacts, interfaces, and the frame material.
- The sample identity: the batch, the build and the number of units.
- The seal material, whether it is replaceable, and the expected service life.
- The re-validation cost per configuration, and the trigger that causes it.
- The remaining certificate numbers for the rest of the conformity file.

Where an ingress programme goes wrong, and the early warning for each
Five failure patterns show up repeatedly in this category, and each of them is cheap to prevent before the tooling is committed.
The claim says waterproof and the customer tests it in the rain. The symptom is a cluster of returns from one channel with a description that says the product is broken. The root cause is that the marketing instinct is stronger than the marking. The prevention is to write the marking in words, publish the exclusions next to it, and have the wording approved against the document rather than against the intuition of whoever writes the copy.
A certificate arrives naming a product nobody is shipping. The symptom that surfaces is a warranty claim, a customs query, or a listing rejection with no way to close it. The root cause is that a family-level document was treated as coverage. The prevention is to make the exact model designation a condition of the purchase order and to check the string against the units on receipt.
Water is found at the charging interface. The symptom is a unit returned with corrosion from a customer who did nothing wrong. The root cause is that a designed opening is an ingress path and the test protocol ignored it. The prevention is to put the interface first in the protocol, to write a cleaning instruction for it, and to make the interface part of the acceptance criteria rather than a footnote.
A variant ships outside the covered configurations. The symptom is a consistency problem that appears after launch, usually in the colour that sold best. The root cause is that no per-variant configuration check existed, because the change was treated as a colour rather than as a change to a sealed assembly. The prevention is to put the seal, the port, the contact and the frame material on the change-control list, with a notice obligation attached.
The protection is fine in year one and the returns start in year three. The symptom is a slow return-rate drift with no obvious cause and a support team that cannot explain it. The root cause is that the seal aged and nobody tracked the service life. The prevention is to set the warranty against a measured failure criterion rather than against a calendar assumption, and to re-test a retained unit periodically against a fresh one.
The commercial structure follows the same pattern. Negotiable: the number of configurations you commit to, the per-colour minimum for each, the artwork round count, the per-market label scope, the market list itself, the re-validation cost per variant and the trigger that causes it, the delivery split against a documentation milestone, and a retention held against the documentation package, which protects you on the failure that costs most — a product that arrives with a marking nobody can evidence. Not negotiable: certification cost and certification time, which follow the assessment process rather than your order size, and the enclosure design itself, which is not a commercial variable. If a supplier will not change the seal material, the seal geometry or the port design, the ingress performance of that programme is fixed, and the programme should be built around that fact rather than against it.
The boundary of our own ingress documentation
We publish an ingress marking exactly as the catalogue states it, mark everything else as open, and will not describe a splash-protected product as waterproof, rain-capable or immersible.
Concretely, and without softening it: in our catalogue the IP54 marking appears on the V06 and the S7, as an annotation. We publish no third-party test report number behind it, no standard, no test condition, no sample size and no laboratory, because the company has not confirmed them to us. No other model in the range carries a published ingress marking. The Bluetooth audio family carries none at all, which for an open-ear design with a sound port and microphone ports is a structural point rather than only a documentation gap. We publish no model-level certificate numbers, no test reports, no minimum order quantities, no production capacity and no lead times, and those remain open internally.
The other half is worth stating as plainly. No marking on a model is not a statement that the model cannot handle the environment; it is the absence of a claim. And a marking without a report number is a statement rather than a document, and the difference matters the first time somebody asks — a broker, a marketplace, a retailer's technical team or a customs inspection.
Three things to do before the next supplier call.
- Write the exposure down per channel. One line each: where the product is, for how long, how often, and what happens afterwards. If a line says "occasionally, in the rain", you have a splash case. If it says "under a pressure washer", you do not, and the codes are not comparable.
- Send the document request as a list, not as a question. The report number with the issuing body and the standard, the model list, the tested configuration with its ports and contacts, the sample identity, the seal material with its expected service life, and the re-validation cost per variant. A supplier who answers all six has a documentation habit; one who answers two is telling you something you need before the deposit.
- Send us the environment, the channel and the price band. We will say which level that justifies, which of our models can carry it, what the acoustic and interface cost would be, and which rows of documentation you would have to obtain before the claim could be made stronger. The codes are explained under IP ratings and durability, and the models that carry a marking are on the sport smart glasses range page.
Related reading
- The codes and what each one permits — IP ratings explained
- Choosing between two products fairly — how to compare IP ratings
- The evidence a claim needs — certification requirements
- The models that carry a marking — sport smart glasses range
Send the environment, the channel and the price band, and we will tell you which level it justifies.



