Buying aluminum windows and doors for one house is a shopping decision. Buying them for a real estate development is a scheduling decision. The product is the same; the risk is multiplied by every unit, every floor, and every plot on the site. This guide is written for property developers, main contractors, and architects who specify window and door packages at scale and want the factory conversation to start from the right documents rather than a photograph of a showroom sample.
A single showroom unit tells you whether a profile looks right. It tells you nothing about what happens when that unit is repeated forty, four hundred, or four thousand times across a programme. Every choice — opening type, frame colour, glass make-up, handing, threshold height — is decided once in design and then copied onto every elevation. A good decision copied is a saving. A marginal decision copied is a fixed cost. A wrong decision copied is a defect programme, not a single defect.
That is why, across multi-unit schemes, windows, doors and the related envelope items are usually the first major packages a developer buys from a source factory: they are high in quantity, they repeat across every unit, they are specified by schedule and drawing rather than by taste, and they are the most expensive to correct precisely because by the time the error is visible the building is already closed in. The factory is not the last step in the supply chain; on a development it is the first constraint on the programme.
On a multi-unit scheme the window and door package is normally the first major envelope item to be frozen, because almost everything else — cladding, internal partitions, mechanical penetrations, the sill detail — keys off the structural opening. Specifying early but loosely is worse than specifying late but tightly: a transposed dimension on a drawing that is repeated across forty plots becomes forty units of rework, not one. The discipline that protects a development is to treat the opening schedule as a controlled document, not a take-off that is finalised at bid time.
A price list tells you what a product costs in isolation. It does not tell you whether the product fits your drawing. Before a competent system-window factory can return a real number, it needs the documents that describe your building, not theirs. The minimum set:
Send the schedule and the drawing, and a competent supplier returns a priced line-by-line schedule against your own opening references — not a re-numbered price list — together with a queries list of the lines that cannot be quoted as written. That queries list is where most of the value sits: it is the factory reading your drawing harder than you did.
Most ordering errors on a development do not originate in the factory. They originate from a dimension that was assumed rather than measured, a wall thickness taken from the architect's intent rather than the built-up wall section, or a handing written at the wrong end of the schedule. The cheapest defence is to coordinate the architectural drawings with the manufacturing schedule and to sign off one sample plot first. One proven reference unit on site is far cheaper than discovering the error at plot thirty-nine.
The most expensive version of a drawing error is one that passes factory quality control and then fails on site. That is why a competent casement window contractor should be part of the sample-plot sign-off, rather than being brought in only when the programme is already slipping at plot forty.
Agree the finish early and keep the development to a single powder-coat run. A single finish across the whole scheme looks sharper and avoids the mismatched-batch problem that appears when different phases are painted in separate production windows. ALPES holds four standard finishes — Ice Cream White, Aurora Gray, Starry Sky Gray and Brown Coffee — with custom colour matching available, which means the colour decision can be locked at schematic design rather than discovered during snagging. If the scheme also includes entrance sets, put the entry door factory on a separate line of the schedule and insist on one powder-coat run across the entire development so the entrance and the windows read as one building.
On a phased development, release timing matters as much as unit price. Release too early and you pay to store a site's worth of frames; release too late and the glazing holds the entire programme. The practical pattern is a master purchase order against the total unit count, with phased releases tied to each phase's close-in or interior-trim dates, rather than dozens of individual orders. A factory that understands phased delivery is worth more to a developer than the same factory quoting a fraction lower on a single unit.
| Buyer role | What it optimises | What it most often gets wrong | What to lock first |
|---|---|---|---|
| Developer / owner | Whole-project margin, programme, exit value | Treats the openings as a commodity line item | One schedule, one colour run, sample-plot sign-off |
| Main contractor | Install sequence, snagging, programme | Assumes wall build-up instead of measuring it | Confirmed wall build-up per opening before order |
| Architect / specifier | Appearance, compliance, performance | Writes hardware by brand where the equivalent is unavailable | Configuration-locked spec, not brand-only spec |
| Property investor | Durability across the portfolio, service cost | Buys on first cost, ignores repeatability | Spare-parts policy and finish consistency |
When you shortlist a casement window manufacturer, the question worth asking is not whether the first sample is good, but whether the four-hundredth unit is identical to the first. That is a factory-control question, not a showroom question, and it is the one that decides whether a development hands over clean.
None of this reduces the value of buying from a source factory. It simply places the value where it belongs: in repeatable manufacture, documented performance, and a schedule that arrives when the programme needs it — not in a certificate that travels with the product.
ALPES manufactures on two production centres in Foshan and Zhaoqing, Guangdong, with a combined 100,000 square metre facility, and builds around 6060-T66 aluminum (wall thickness 1.8 to 2.5 mm) with PA66GF25 multi-chamber thermal breaks and German-system hardware from ROTO, Winkhaus, CMECH and Wehag. Across a development the relevant systems are:
| System | Type | Profile | Thermal K | Notable for repeat-build |
|---|---|---|---|---|
| C120 | Outward casement | 6060-T66, 1.8-2.0 mm | 1.6 | 119 mm frame wall, 62 mm sash, full glue-injection corners |
| C113 | Outward casement | 6060-T66, 1.8-2.0 mm | — | 113 mm frame, 34 mm insulation strip, CMECH hardware tested to 30,000 cycles |
| C116 | Outward casement | 6060-T66, 1.8-2.0 mm | 1.7 | 116 mm frame, 81 mm sash, 34-44 mm multi-chamber strip, 304 stainless hinges |
| AW90 | Tilt-and-turn / passive | 6060-T66, 1.8 mm | 1.3 | 90 mm frame wall, 360 degree locking, 8-layer sealing, Winkhaus hardware |
| TS130 | Side-press sliding window | 6060-T66, 1.8-2.0 mm | — | 130 mm frame wall, ROTO 200 kg pulleys, 8-point lock, casement-level seal |
| SL155 | Thermal-break sliding door | 6060-T66, 2.2-2.5 mm | — | Four-track 155 mm, 400 kg 20-wheel, single sash 2800 x 2600 mm / 7.5 m2 |
| SL112 / 167 | Heavy-duty sliding door | 6060-T66, 1.8-2.5 mm | — | 112 / 167 mm, 220 kg driving, 45 mm ｄｒｏｐ + 85 mm barrier, 24A glass |
| SL126 / 182 | Thermal-break / lift-slide | 6060-T66, 2.0 mm (3.0 mm local) | — | 126 / 182 mm, lift-slide 2500 x 3000 mm / 7 m2, 27A glass, 8 mm low or 43 mm high track |
Copy this into the enquiry and fill one row per distinct opening. Empty cells are where the factory will send queries, so a blank here is a delay, not a silence.
| Field | What to fill | What happens if left blank |
|---|---|---|
| Opening reference | Your own tag per opening | Line cannot be tracked against your drawings |
| Overall size (mm) | Frame outer dimension, stated as frame or masonry | Factory assumes; one wrong assumption repeats across the plot |
| Wall build-up | Layer-by-layer from outside to inside | Wrong reveal, wrong anchor length, on-site rework |
| Configuration | Fixed / casement / sliding / tilt-turn | Quoted against the wrong system |
| Glass make-up | Thickness, cavity, gas, laminate | Thermal and acoustic target not verifiable |
| Colour / finish | Exact finish call-out | Split batches across phases |
| Handing + convention | Handing and the convention used | Wrong-handed units arrive and cannot be flipped |
| Hardware set | Locking and handle reference | Mixed ironmongery across the scheme |
| Quantities by unit type | Count per building or phase | No phased-release plan possible |
| Performance basis | Standard, wind pressure, thermal target | Compliance cannot be evidenced |
| Item | ALPES publishes | You must still request | Why the gap exists |
|---|---|---|---|
| Profile alloy and thickness | 6060-T66, 1.8-2.5 mm | Wall thickness for your exact series | Depends on the series and local reinforcement |
| Casement / tilt-turn K-value | 1.3 / 1.6 / 1.7 (per system) | K for your glazing and size | A thermal value belongs to a tested configuration, not a name |
| Sliding-door K-value | Not published | Calculated value per configuration | Sliding systems are sealed by contact, not compression |
| Coating durability | 240 h salt spray (AW90) | Your climate exposure statement | Hours describe a test, not your site's life |
| Hardware cycles | 30,000 cycles (CMECH) quoted | Cycles for your hardware set | Bench cycles are not field service life |
| Local approval | Factory test reports | Certificate in your market | The certificate holder is the importer or developer |
| Perimeter seal | Factory-injected sash seal | Site-applied weatherproofing | Installation is a separate contract from manufacture |
The product can be the same; the documents must not be. At ten units a drawing error is annoying. At a hundred units the same error is a programme event. The supplier that earns the volume order is the one that returns a queries list and supports a sample-plot sign-off, not the one with the lowest single-unit price.
Early enough that the structural openings, lintels and sills are coordinated with the manufacturing schedule, but only after the wall build-up is confirmed. Freezing on intent rather than on the built-up section is the usual cause of on-site rework.
Yes. Powder-coat batches can read differently when painted in separate production windows. A single agreed finish, locked at schematic design and run across the whole development, removes that risk and usually simplifies the order.
Only if your configuration — leaf or sash construction, frame, glazing and hardware as a set — matches the tested one. A report for a different hardware set or glass make-up does not carry over, which is exactly why the performance basis belongs in the schedule, not in a generic brochure.
The window and door schedule plus the head, jamb and sill details. Those two items let a factory return a priced, line-by-line schedule and a queries list — the two things that turn a quote request into a buildable package.
Not at all; sliding systems are the workhorse of balcony and patio openings precisely because they repeat well. The only honest caveat is that the sliding families do not carry a published K-value, so the thermal figure must be requested for your exact glazing and size rather than borrowed from a casement system.
