If you arrived here by typing something close to tilt and turn aluminum windows factory with ISO certification, you typed a phrase that ISO itself asks people to stop using in that shape. On ISO's own page for ISO 9001 the guidance is explicit: in any statement about certification, refer to the standard with the full reference — certified to ISO 9001:2015, not simply certified to ISO 9001. The year is not decoration. It is the half of the reference that tells you which edition was audited, and editions change: ISO has published that a revised edition was expected during 2026 and that organizations holding certificates to the 2015 edition would be given a transition period. Confirm which edition is current on ISO's own site rather than from this article.
The second thing ISO says plainly, on the same page, is that it does not certify organizations. Certification is performed by independent certification bodies, which may themselves be accredited by national accreditation bodies. The joint ISO and IAF communiqué on the expected outcomes of accredited ISO 9001 certification says the same thing in writing: ISO is the developer and publisher of ISO 9001 but does not itself carry out auditing and certification, and it does not control the bodies that do.
So we are not going to tell you that certification is worthless. It is not. We are going to tell you two things that are more useful than either the yes or the no.
That second point is the one this article exists to make, because almost nobody writes it, and because it is the point where a certificate about paperwork meets a window that has to stay square for decades.
When a supplier email says we are ISO certified, the number is usually missing. That matters, because the ISO family contains at least six numbers that appear in window and door marketing, and they are not variations of the same claim. Three of them certify an organization's management system. Three of them accredit the competence of somebody who does assessments. Only one of the six is even capable of being about a product, and it is almost never the one being offered.
| Number | What it governs | Who issues or grants it | What it means for one tilt-and-turn window | The usual misreading |
|---|---|---|---|---|
| ISO 9001 | Quality management system of an organization | A certification body (CB), often accredited | The factory has a documented, audited way of working. Nothing about the sash. | "Certified, so the windows must be good." |
| ISO 14001 | Environmental management system | A CB | The factory manages its environmental aspects under a system. Not a claim about emissions of any product. | "Certified, so the materials are green." |
| ISO 45001 | Occupational health and safety management system | A CB | The factory manages worker safety under a system. It says nothing about your site. | "Certified, so the install will be safe." |
| ISO/IEC 17021-1 | Competence of bodies that audit and certify management systems | An accreditation body (AB) accredits the CB | This is the one that decides whether the body issuing the 9001 certificate was itself competent. | "It says 17021 on the certificate, so the product is certified." |
| ISO/IEC 17065 | Bodies certifying products, processes and services | An AB accredits the product certification body | The only route here that can ever be about a window. A product certificate, if one exists for your market, comes through this door. | "ISO 9001 is our product certificate." |
| ISO/IEC 17025 | Competence of testing and calibration laboratories | An AB accredits the laboratory | Your air, water and wind test reports should come from a laboratory accredited to this, and to the specific test. | "Any test report is as good as any other." |
| ISO/IEC 17011 | Accreditation bodies themselves | Peer evaluation under the IAF MLA | This is the floor under the other six rows. If it is missing, the rest is a logo. | "There is an accreditation mark, that is enough." |
One more line from ISO's own page is worth carrying into every supplier conversation: individuals are not certified to ISO 9001. A person can complete auditor training or a lead auditor qualification. An engineer cannot be ISO 9001 certified. If a sales email says their engineers are, you have learned something about the email.
Every management system certificate carries a scope statement, and that statement is written by the certified organization. This is not a loophole we are exposing; it is what the standard asks for. Clause 4.3 says the organization shall determine the boundaries and applicability of the quality management system to establish its scope. A narrow scope, honestly declared, is not a fault. It is the system working.
Which is exactly why you have to read it. Three fields on that page decide whether it has anything to do with your order, and they are the three that nobody reads:
A scope sentence that reads manufacture of aluminium windows and doors can sit on a factory that has never built a tilt-and-turn, and it can sit on one that also builds a barn entry door. The sentence is a statement about a management system. It is not a claim about either product, and the auditor was not asked to have an opinion on either product.
Here is the part that almost no buyer asks about, and that decides more about a tilt-and-turn window than any other field on the certificate.
The earlier edition of ISO 9001 specifically permitted the exclusion of certain clauses, including design, where they were not related to the organization's system. UKAS's technical bulletin on auditing design activities makes the point that ISO 9001:2015 does not include that provision. It replaced exclusion with applicability: the starting point is that all requirements apply, and the organization determines the boundaries under clause 4.3. The ISO/TC 176 Auditing Practices Group guidance says the same thing and adds the operative rule — if the organization has responsibility for design and outsources the design process, the scope statement for certification shall include wording such as Design of, Development of or Design management of. And if that cannot be demonstrated, design and development should not be stated in the certification scope at all.
In other words, whether the word design appears in that sentence is an audited distinction. It is not marketing copy. Two factories can hold genuine, accredited ISO 9001 certificates from the same certification body, and one of them has had its design activity audited while the other has declared that design is not within its system.
Now put that next to our own product. Here are five things about a tilt-and-turn window that are settled on a drawing, long before anything reaches an assembly bench:
| Design decision | What it decides on site | Inside the audited system only if design is in scope |
|---|---|---|
| Whether a mishandling device is fitted, and which type | Whether a wrongly turned handle is a five-minute reset or a two-person lift | Yes |
| Which mass grade and durability grade the hardware set is declared to | Whether the sash sags onto its keeps in five years or twenty | Yes |
| Where water goes — drainage path, chamber equalisation, weep positions | Whether water leaves or waits | Yes |
| Corner joint method and corner drive engagement depth | Whether the sash stays square and the gear stays in sync | Yes |
| Glass rebate depth, packers, and how sash weight reaches the hinge | Where the load path runs, and what the hinge is asked to carry | Yes |
If design is not in that sentence, none of those five were inside the audited system. The certificate is still real. It just does not reach them, and those five are where tilt-and-turn failures actually live.
A tilt-and-turn handle turns through one hundred and eighty degrees and has three states ninety degrees apart: down is locked, horizontal is turn, up is tilt. Running around all four edges of the sash, hidden in the groove, is a continuous run of steel bar with teeth, cams and corner pieces. The handle drives one gearbox in the middle of one edge, and that gearbox shifts the whole track.
A Scottish repair specialist who writes publicly about these windows put the central fact better than any manufacturer has: the gearbox has no idea whether the window is open or shut. It only knows where the handle is pointing.
Everything that follows from that sentence is a design consequence. If the sash is open in turn mode and someone turns the handle up to tilt, the track shifts anyway. The bottom pivots engage while the sash still hangs on its side hinge and the top arm releases. The sash ends up held at one corner with the top swung into the room. It looks like the window has fallen off. Nothing is broken.
Nearly every tilt-and-turn made in recent decades carries a mishandling device: a small spring-loaded block that senses the sash is off the frame and blocks the handle from turning. Public repair guidance is honest about the limits — they work well, they wear, they can be beaten by someone turning hard enough, and older windows may not have one at all. Public repair guidance also says the single most common cause of a mechanism going out of sync is the person operating it, not the factory.
We are quoting that against our own interest, because it is true and because it points at the real question. Whether that block is on the sash at all is a decision made at the drawing stage, years before an auditor arrives. It is a clause 8.3 output, not an inspection result. And that is the whole argument of this section: the failure mode everyone has heard of is prevented, when it is prevented, by a design decision — so a certificate whose scope excludes design has structurally nothing to say about it.
If you want a number that bears on whether a tilt-and-turn sash will still be hanging straight in ten years, it is not on the ISO certificate. It is on a code marked on the hardware. EN 13126-8 is the European standard for tilt-and-turn, tilt-first and turn-only hardware, and the 2017 edition classifies a hardware set in four boxes.
| Box | What it grades | Published values | Why a buyer should care |
|---|---|---|---|
| 1 | Durability cycles | H1 five thousand, H2 ten thousand, H3 twenty thousand | These are bench cycles on a rigged specimen, not a service life. Tilt and turn operations are combined in the count. |
| 2 | Sash mass in kilograms | Ten to fifty in five-kilogram steps, then sixty and above in ten-kilogram steps | This is the box that has to match the sash you were actually sold. |
| 3 | Corrosion resistance per EN 1670 | Grade three and above; grades zero to two are excluded | Coastal and pool-side projects should ask for the grade, not the word resistant. |
| 4 | Test size | Published listings give one thousand three hundred by one thousand two hundred, and larger sizes for heavier hardware | A small test size does not cover a large sash. Ask which one was used. |
A complete code looks roughly like H2-080-3-1,300x1,200: durability grade, mass grade, corrosion grade, test size. Note what that is. It is the hardware's number, not the window's. It travels with the hardware set, and the only way it becomes true of your window is if somebody matched it to your sash.
Everything below is our own arithmetic, with the inputs shown so you can redo it.
Now look at the ladder again. The mass box is graded in five-kilogram steps up to fifty, and in ten-kilogram steps above sixty. Forty-seven kilograms of glass, before a single gram of aluminium, hardware, sealant or glazing tape, sits two steps from the end of the fine part of that ladder. Every kilogram added after that moves into the coarse part.
That is worth saying slowly: for a sash of the size the standard itself tests, the number that governs hardware selection is one the standard only resolves in ten-kilogram steps. A wrong guess up there is twice as far wrong, and nothing in an ISO 9001 certificate is structured to notice. The certificate audits whether product characteristics have been specified. It does not audit whether this one was specified correctly for this glass build.
The same point can be made from the other end of the building. A tilt-and-turn sash and a solid core entry door are, to a hardware standard, the same kind of object: a mass hung on a mechanism, graded by how heavy it is and how many cycles it must survive. They sit in different product families and different standard boxes, and neither grade travels with the factory's ISO certificate. If you change glass specification after quotation, that mass number moves, and the hardware grade that was right at quotation may not be the hardware grade that is right at production.
The same published summaries of EN 13126-8 describe a one thousand newton additional load applied for five minutes, and a reveal test in which a ten kilogram mass is dropped three times.
Our arithmetic: one thousand newtons divided by 9.81 is about one hundred and two kilograms-force. So the hardware standard asks, in effect, what happens if a person's weight ends up on the sash, and what happens if something heavy falls against it. Your ISO certificate was not in the room for either test. That is not a criticism of the certificate. It is a description of what kind of document it is.
Published tables of contents for the two editions tell a quiet story. The 2006 edition classified hardware across nine boxes — category of use, durability, mass, fire resistance, safety in use, corrosion resistance, security, applicable part and test sizes. The 2017 edition classifies across four. Both get called an EN 13126-8 classification. A nine-box code and a four-box code are not competing answers to the same question; they are answers to different editions of the question.
This is the same trap as the missing year on an ISO certificate, one level down. When a supplier quotes you a hardware code, ask which edition it was written to.
EN 13126-8 contains a clause on admissible tolerances, covering sash operation, handle operation and locking point variability. Read that as a buyer and it says something important: the hardware tolerates a defined amount of misalignment by design. It is not asking for a perfect sash. It is telling you how much imperfection it absorbs, and therefore telling you that beyond that figure you are asking the mechanism to do a job it was not rated for.
So the useful question to a factory is not are your windows square. It is: what sash diagonal difference do you hold, what do you do with a sash that is outside it, and can I see the record. Those are three questions a quality system should be able to answer in writing, and a scope statement cannot answer at all.
The joint ISO and IAF communiqué is not a sceptical document. It lists, at some length, what accredited certification is supposed to deliver. Condensed, an organization with a certified quality management system is expected to have a system suitable for its products and processes and appropriate to its scope; to analyse customer and statutory requirements; to ensure that product characteristics have been specified; to determine and manage the processes needed; to resource them; to monitor and control the defined product characteristics; to prevent nonconformities and, where they occur, to correct them including product nonconformities detected after delivery, analyse their cause, take corrective action and address customer complaints; and to run internal audit and management review.
That is a real and useful list, and every line of it is something you can hold a supplier to after something goes wrong. Note the shape of it: it is a list of obligations about process, and the word product appears only as the object of those processes.
Then the communiqué does something unusual. It has a section titled What accredited certification to ISO 9001 does not mean, and it contains two numbered statements. We reproduce their substance here rather than paraphrasing, because paraphrase is how a caveat turns into a selling point:
One. ISO 9001 defines the requirements for an organization's quality management system, not for its products. Accredited certification should provide confidence in the organization's ability to consistently provide product that meets customer and applicable statutory and regulatory requirements. It does not necessarily ensure that the organization will always achieve one hundred percent product conformity, though that should be a permanent goal.
Two. Accredited certification to ISO 9001 does not imply that the organization is providing a superior product, or that the product itself is certified as meeting the requirements of an ISO or any other standard or specification.
There is one more line in that document that buyers almost never use, and it is the most practical sentence in it. In the notes to the expected outcome, customer requirements for the product are described as possibly stated — in a contract or an agreed specification — or generally implied, with the example given being the organization's own promotional material.
Read that again. Their brochure is part of the requirement set their own system is audited against. If a catalogue says German hardware and the crate contains something else, you are not only holding a commercial complaint. You are holding a nonconformity against the system the certificate describes. Ask for the specification in writing, on the order, and keep the page you were sold from.
We would rather give you an inspection you can run than a number you have to believe. Public guidance from a national accreditation body on validating management system certificates sets out three steps, and they take minutes.
| Step | What to check | The part everyone skips |
|---|---|---|
| 1 | That the issuing certification body is accredited by an IAF MLA signatory accreditation body. Anyone can call themselves an accreditation body; the test is compliance with ISO/IEC 17011 and acceptance as an MLA signatory. | The IAF MLA is split into sub-scopes. The accreditation body must be a signatory for the sub-scope that relates to your certificate, not merely a signatory. |
| 2 | That the certification body's own accreditation actually covers this management system. Check the accreditation body's site for the certification body's scope of accreditation. | A body can hold accreditation that does not include the standard on your certificate. |
| 3 | That the certificate itself is current, on the certification body's own register, with matching entity, address, standard and edition, scope and dates. | Certificates run on a three-year cycle with annual surveillance audits, and can be suspended. The document you were emailed may have lapsed since. |
There is a global database, IAF CertSearch, that lets you search by certificate number or company name. Use it. But be honest about what it can and cannot tell you: public guidance notes that it is not comprehensively populated, and that populating it is not mandatory for certification bodies. Absence from that database is a reason to go to step three, not a finding of fraud. A page that tells you a certificate is definitely fake because it did not appear in a search is a page to stop reading.
Red flags that are worth a second look, all drawn from public guidance: a certificate issued in a few weeks when a legitimate process is normally described as taking several months; a scope statement so broad it could describe any factory; a certification body that appears in no accreditation register anywhere; and a supplier who cannot point you to their own entry on the issuing body's register. That last one is the shortest test of all.
A certificate is a document about a system. A factory visit, or a written questionnaire answered with documents attached, is a test of whether that system is running on your order. The single best question we have seen in public guidance written for industrial buyers is also the simplest: ask to see the last three nonconformity reports and their corrective actions.
The reasoning in that guidance is worth quoting because it is so clean: a factory with zero recorded nonconformities is either not recording them or not finding them, and both are problems. A functioning quality system produces these regularly and handles them systematically. A factory that cannot produce recent nonconformity records is running the paperwork without running the system.
| Ask for | A good answer looks like | Sounds reasonable, carries no information |
|---|---|---|
| The last three nonconformity reports and their corrective actions | Dated records, named causes, named corrective action, a closure date | "We have very few quality problems." |
| In-process inspection records, not only final inspection | A signed check at cutting, machining, corner assembly and glazing, tied to an order number | "Every window is inspected before packing." |
| Calibration records for the measuring equipment used | A list of gauges and tapes with current calibration dates | "Our workers are very experienced." |
| The mass grade and durability grade of the hardware set on your order | A four-box code, the edition, and the sash mass it was chosen for | "We use German hardware." |
| Whether a mishandling device is fitted, and of what type | A named part on the bill of materials for your order | "All our windows are safe." |
| What sash diagonal difference is held, and what happens to a sash outside it | A number, a gauge it is measured with, and a rework route | "We make them square." |
| Who signs the drawings your order is built from, and whether design is in the certificate scope | The scope sentence quoted verbatim, plus a named responsible person | "Our engineering department is very strong." |
| The test report behind any performance claim, with laboratory and standard named | Standard, edition, specimen configuration, laboratory, date | "It is tested." |
The honest version of this table has a third column explaining the gap rather than filling it. An unexplained blank in a catalogue is where assumptions get made, and assumptions turn into site problems.
| Item | What we publish | Why there is a gap, and what to do |
|---|---|---|
| Thermal value | K 1.3 for the AW90 passive window | That is a system value, not a whole-window value for your opening, and not a rating in your national scheme. Ask for a calculated value at your size, division and glass build. |
| Profile | 6060-T66, 1.8 millimetre on AW90 | Alloy and temper tell you what the material is, not what the section does. Ask for the section drawing with dimensions. |
| Salt spray | Two hundred and forty hours on a named series | It tests coating and substrate. It is not a rating for air, water or wind, and it does not transfer to another series. |
| Shear strength | Five thousand newtons on a named series | It measures the composite bond between insulation strip and aluminium. Useful, and about something else entirely. |
| Hardware | Winkhaus, with C-slot ROTO as an option | A brand is not a grade. Ask for the four-box classification code of the set on your order, and the edition. |
| Air, water and wind | Not published as a class across the range | Absent does not mean worse; it means there is nothing to compare. Ask for a report in the standard your project uses, on the configuration you are buying. |
| Maximum sash size | Not published | It is a function of the glass build, and the glass build is your choice. Give the glass and ask for the size, not the other way round. |
| Sash mass | Not published | It depends on your glass. This is the number the hardware grade is chosen from, so ask for both together. |
| AW113 series | Listed, with incomplete published detail | We would rather say that than let you build a schedule on it. Ask for the current drawing set. |
| Our own certificate | No certificate number, issuing body or scope is published on this page | Ask us for all three and run the checks above against them, exactly as you would against anyone else. |
| Your project approval | Nothing | Approval attaches to a building and to the authority that governs it. We can supply evidence, drawings and specimens. We cannot be the applicant. |
We have more motive to say yes than anyone reading this has to hear yes, which is precisely why the following three should be said by us.
That last point connects directly to the argument of this article. A supplier that answers every line of your schedule with yes is not demonstrating a quality system. It is demonstrating that its scope statement has not been read.
Every one of these is a real thing that happens to real tilt-and-turn windows, and in almost every case the first thing people do makes it worse. The last column matters more than the second.
| Symptom | Most likely mechanism | What people usually do first, and should not |
|---|---|---|
| Sash hanging on one bottom corner with the top swung in | Handle moved while the sash was open; two modes engaged at once | Letting it hang and swing while deciding what to do. Take the weight first, then reset. |
| Handle turns but nothing happens | Gearbox stripped or disengaged from the track | Forcing it. Force is how a repairable gearbox becomes a replaced one. |
| Sash catching the frame at the bottom, harder to lock over time | Sash dropped on its hinges under glass weight; locking points no longer meet their keeps | Lifting with a lever or screwdriver. That damages frame, hinge and sealed unit at once. |
| Draught when shut | Compression lost at one or more locking points, or gasket set | Adding more sealant round the outside. The leak is at a locking point, not the perimeter. |
| Handle stiff, then suddenly easier | Track dry or a corner drive wearing; the sudden change is the part that matters | Oil-based spray that collects dust. Use the lubricant the hardware maker specifies. |
| Water on the inner face of the outer frame | Drainage path blocked, or the outer chamber doing what it is designed to do | Sealing the whole outer perimeter. That traps water until the chamber fills and comes inward. |
Fill these in before you ask for a price. Every empty line becomes an assumption, and assumptions get priced optimistically and delivered literally.
| Line | Fill in | What happens if you leave it blank |
|---|---|---|
| 1 | Opening sizes in millimetres, and whether they are structural opening or frame size | Two people measure two things and the difference appears at installation |
| 2 | Wall build-up from outside in, layer by layer | A single depth number gets read as wall thickness, which it is not |
| 3 | Glass build-up, including thickness of every pane and the cavity | Sash mass is assumed, and the hardware grade selected from an assumed mass |
| 4 | Estimated sash mass in kilograms, and who calculated it | Nobody owns the number that the hardware grade depends on |
| 5 | Required hardware classification: durability grade, mass grade, corrosion grade | "Heavy duty" is supplied, and cannot be shown to be wrong |
| 6 | Which edition of the hardware standard those grades are written to | A four-box code and a nine-box code get compared as if they were the same ladder |
| 7 | Whether a mishandling device is required on every opening sash | It becomes an option, and options get value-engineered out |
| 8 | Performance standard and class required, in your market's own scheme | A result arrives in a different scheme and cannot be compared |
| 9 | Certificate requirement: standard, edition, issuing body, and whether accreditation is required | An unaccredited certificate arrives and looks identical to the real one |
| 10 | Documents required with the shipment, and who signs each | The paperwork argument starts after the container lands |
| 11 | Spare parts to ship in the original container, listed by part | Year five needs one cam and gets a freight discussion instead |
| 12 | Freeze date for changes, and what a change after that date costs | A glass upgrade late in the run silently invalidates the hardware grade |
| Question | Answer to be suspicious of |
|---|---|
| Are you ISO certified? | "Yes, ISO certified." No number, no year, no body. |
| Can I see the certificate? | A scan with the scope sentence cropped out of the frame. |
| Is the certification body accredited? | "Of course." Ask which accreditation body and for which sub-scope. |
| Does your scope cover design? | "We have our own engineering team." That is not what the scope sentence says. |
| What hardware grade is on my sash? | "Top grade German hardware." A brand is not a grade. |
| What does my sash weigh? | "Our hardware handles up to a large weight." Unrelated to your sash. |
| Is there a mishandling device? | "The window cannot be mis-operated." It can. |
| Can I see quality records? | A tidy folder of final inspection sheets and nothing before glazing. |
| Will the test report cover my window? | "The series is tested." A series is not a specimen. |
| Can you supply everything on my schedule? | An unqualified yes to every line, including the products they have never built. |
No, and that is not a criticism of the standard. The joint ISO and IAF communiqué states in numbered terms that accredited certification does not imply a superior product, and that the product itself is not certified as meeting any ISO or other standard. What it should give you is confidence in the organization's ability to consistently provide product that meets customer and applicable statutory and regulatory requirements — for the defined certification scope.
Yes, mostly after the fact. The expected outcomes include correcting product nonconformities detected after delivery, analysing their cause, taking corrective action and addressing customer complaints. Those are obligations you can point at when something goes wrong. The certificate is worth more as a lever than as a promise.
Because the failure modes of this window type are design-shaped. The gearbox does not know whether the sash is open; the thing that stops a wrongly turned handle from dropping the sash onto one corner is a small block chosen on a drawing. Whether it is fitted, which hardware mass grade was selected, where water goes, how the corner joint is made — those are design outputs. If design is not in the scope, they were not inside the audited system.
The four-box classification code of the hardware set on your order, with the edition of the standard it was written to, and the sash mass it was selected for. Then ask for the calculated mass of your sash with your glass build. If those two do not meet, nothing else in the file matters very much. You can see how we approach this on our own tilt and turn window range.
Not necessarily, and anyone who tells you otherwise is overstating it. Public guidance notes the database is not comprehensively populated and that uploading certificate data is not mandatory for certification bodies. Treat absence as an instruction to go to the issuing body's own register and check entity, address, standard and edition, scope and dates there.
Because we are the party with the strongest motive to say yes, and that is exactly why a yes from us is the least useful answer you can get. We do not publish a certificate number, issuing body or scope statement on this page. Ask us for all three and run the three checks against them. If a supplier is annoyed by that request, the certificate was never the point of the certificate.
