I started reviewing rectangular eyeglasses deliveries in 2021 — first as a QC contractor for a regional optical chain, then as brand compliance manager for a mid-size B2B buyer. The number one source of trouble I see, and the one I've caused myself, is applying a single procurement checklist to every order. A first production run, a restock, and a supplier switch all fail for different reasons. The sourcing approach has to change with the order type.
Below are the three scenarios I now plan for. If you only read one section, read the one that matches where you are in the buying cycle. If you're not sure, the last section gives you a rough way to figure out which one you're actually in.
Scenario A: First order from a new factory
The first order is where most quality problems get baked in — not because suppliers are dishonest, but because the spec hasn't been tested yet. A rectangle glasses frames spec that says "gloss black, rectangle, acetate" is not a spec. It's a wish.
A working spec sheet for rectangular glasses needs the boxing dimensions — written as lens width □ bridge – temple length in millimeters, for example 52□18-145 — plus vertical box height, material grade, hinge type (5-barrel or 7-barrel), end-piece angle, temple core material, nose pad style, and a color reference (Pantone or a physical chip, not "black"). The vertical box dimension is the one most buyers skip. On a rectangle frame, that vertical box is what separates a square from a proper rectangle in your customer's perception.
I learned this the hard way in Q1 2024. I knew I should have written the end-piece angle into the contract, but thought "we've worked with this factory for two years." That was the one time the verbal agreement got forgotten. The first shipment of black rectangle glasses had roughly a 1.2 mm drift against our ±0.5 mm spec. The supplier said it was "within their normal production tolerance." We rejected 600 units and they redid the run at their cost.
(Should mention: the rejection was a good outcome — most factories will absorb that cost once as a relationship investment. Don't count on a second time.)
My rule for first orders: no verbal agreements, no "trust the sample," and no approval until you've measured a random selection yourself with calipers. If the factory can't hold dimensions across a 30-piece pilot, they won't hold them across 3,000.
Also worth flagging on a first order: ask the factory what's not included in their quote. Logo engraving, custom nose pad colors, branded temple tips, and individual packaging are common add-ons that only surface on the invoice. I now request a full landed-cost breakdown before approving any first-order pricing.
Scenario B: Replenishing a proven SKU
The failure mode here is different. The spec is settled, the factory is known, and the price is set. The risk is drift — small changes that accumulate between production runs and that nobody notices until a customer complains.
My working tool for restocks is a "golden sample." One pair, kept in a sealed bag, signed and dated by both the buyer and the factory, referenced on the purchase order. Every restock gets compared against it: temple weight, hinge tension, surface roughness on the bridge, end-piece fit. If the factory pushes back on keeping a golden sample, that's a signal in itself.
I also run a small blind test on every restock — usually 10 pairs from the new run mixed with 10 from the approved sample. If more than 2 of my team pick the new pairs as "lower quality" without being told which is which, the run gets flagged. The cost of fixing whatever's causing the drift is almost always smaller than the cost of a customer returning their favorite frame.
If I remember correctly, the worst restock we had was in 2023 — same factory, same spec, but the temples felt "lighter" when I picked them up. Turns out they'd switched their temple-wire supplier mid-year. They hadn't told anyone. That one cost us a redo on about 800 pairs.
Scenario C: Testing a new supplier to cut costs
The third scenario is the messy one, and it's where I see buyers make the most mistakes. A cheaper quote on rectangle eyeglasses can be a real opportunity — or it can be a preview of what the next two years will feel like.
The playbook I use for a new supplier is a "stress pilot" rather than a full order. Thirty to fifty pairs, from the exact production line that would make the full run (not a sales-sample line). Then:
- Cut one frame in half and check wall thickness at the end pieces.
- Run a hinge fatigue test — 200 open/close cycles minimum. A properly spec'd 5-barrel hinge shouldn't loosen.
- Test the nose pads for adhesion (a 10 N pull, per the ISO 12870:2016 requirements, is a reasonable starting point).
- Leave a pair in a hot car for a day. Acetate and TR-90 behave differently under heat, and that difference shows up later as customer complaints.
I went back and forth between our existing acetate supplier and a cheaper TR-90 option for about two weeks on our rectangle eyeglasses line. On paper, the TR-90 quote saved us roughly 22% per unit. In the end, we stayed with acetate because our customers associated the heavier feel with quality. I'm still not 100% sure that was the right call — the TR-90 was arguably the better product for people who wear their frames all day.
I should add: every spreadsheet analysis I ran pointed to the cheaper supplier. My gut said their samples felt slightly off. Turns out their temple cores were thinner than the spec sheet claimed, which is why our first test batch developed loose hinges within six weeks of customer wear. The numbers didn't catch that because I hadn't put "temple core gauge" into the spreadsheet.
How to tell which scenario you're actually in
Most buyers aren't in a clean scenario. They're in some hybrid — replacing a supplier on one SKU while restocking three others, for example. The quickest way to place yourself is to check three things:
- Do we have a signed golden sample for this SKU? If no, you're in Scenario A, whether it's technically a first order or not.
- Has the factory changed anything about their process since the last run? Material source, machine, line, sub-supplier — if any of those changed, treat the restock as a Scenario A order with a small pilot first.
- Is the price on this order lower than the last one? If yes, you're in Scenario C, even if it's the same supplier. Price cuts rarely come from efficiency. They usually come from somewhere in the spec.
The honest answer is that I've treated orders as "just a restock" when they weren't, and it cost me. The shop that looks the same on the invoice often isn't the same on the bench.
What this means for the price you actually see
One thread runs through all three scenarios: the quote a factory shows you is rarely the final number. Tooling, packaging, custom nose pads, logo engraving, and shipping all show up later if you don't lock them into the spec sheet. I now ask "what's not included" before I ask "what's the price" — and the suppliers who answer that clearly, even at a higher total, are the ones I end up trusting with the bigger orders.
That rule applies whether you're buying direct or working with a rectangle frames sourcing partner. If the partner can't walk you through the landed cost line by line, that's worth a second conversation before you sign anything.
There's no universal right answer for rectangular glasses sourcing. But there is a right next question to ask, and it depends on where you sit in the buying cycle. If you're not sure which of the three scenarios you're in, the safe default is Scenario A: assume the spec is untested until you've measured it yourself.