I'm a quality compliance manager at Artemide. I review production samples before anything ships—roughly 200 unique SKUs a year. Maybe 180, I'd have to check the system. As of January 2025, I've been doing this for four years. In Q1 2024, I rejected 6% of first factory deliveries due to inconsistent finishes, wrong driver currents, or missing photometric files. That may sound picky. It's supposed to be.
My opinion after all these reviews is simple: the size of an order tells you almost nothing about how much specification discipline you need. A 20-unit trial from a small architect deserves the same scrutiny as a 400-fixture hotel contract. Actually, it deserves more.
Why I Start with an Artemide Floor Lamp, Not a Datasheet
When someone says they want an Artemide floor lamp, most people forward the product page. I do the opposite. I ask for a sample. I look at the mechanical details: hinge, base, switch, cable strain relief. On an articulating arm, I'll move the head to full extension and let go. If it drifts, the tension mechanism is off. We measure joint torque because a 0.5 mm pin variation is enough to change the feel—and for me, feel is part of the spec.
This is also why I like reviewing the Artemide Dioscuri wall light. It looks like a simple cast cylinder, but the approval process is not simple. We measure the finish layer, verify the thermal bond between the LED board and housing, and compare the beam distribution to the original photometric file. If the hot spot shifts by more than 2 degrees, we request another sample. Most people can't see 2 degrees. The wall can.
Table Lamp Specifications Are a Contract, Not a Memo
A table lamp supplier will happily send a one-page table lamp specifications sheet with lumens, watts, CRI, and IP rating. That sheet means nothing until you verify it. I've audited table lamp suppliers where the declared lumen output was 25% above what we measured on an integrating sphere. Often the declared CRI was based on the LED chip, not the finished fixture. In a real lamp, the optics absorb and shift the output.
Looking back, I should have insisted on physical samples during the RFQ stage. At the time, I wanted to save time. It cost us two extra weeks of qualification. Now every supplier evaluation includes a sample, even for small runs.
Here is what I check in a table lamp specifications review:
- Test the fixture at 0 hours, then again after 500 hours. Lumen decline is normal; instant falloff is not.
- Measure CCT and CRI at full brightness and at 10% dimming. Many drivers change color when dimmed.
- Ask for LM-80 test data on the LED module and a TM-21 projection. According to IES LM-80-08 and TM-21-19, lumen maintenance forecasts should be based on long-term tested data, not a marketing guess.
- Check the driver. If the driver is sealed inside the body with no access, ask why. Replacement cost matters more than first cost.
One caveat: I don't expect a small customer to buy an integrating sphere. But you can ask a table lamp supplier for the LM-80 report and an IES file. If they hesitate, that's a signal. (Should mention: some small suppliers don't have the reports, but many don't even know what they are. That's an even bigger signal.)
How to Evaluate Track Lighting Manufacturers
Track lighting is where I see the largest gap between promises and performance. The reason is thermal. A track head is a small enclosed space with an LED module, a driver, and usually no active cooling. Manufacturers test components in open air and then wonder why the installed product fails.
In our own audits, we've seen track lighting manufacturer samples look perfect for the first hour. After 200 hours inside a closed housing, the driver temperature climbed past 92°C, the CCT shifted, and the fixture emitted an audible buzz. None of that showed up in the supplier's catalog.
Here are the four checks I use when evaluating track lighting manufacturers:
- Thermal contact. Is the LED module mounted against a metal heat sink, or is it floating on a plastic housing? Touch the back of the head after 30 minutes. It should be warm, not hot.
- Ground continuity. Track heads are meant to be removed, reinstalled, and reoriented. Ground continuity needs to survive that motion. A continuity tester costs $50. The cost of finding out later is not $50.
- Color consistency from head to head. Put two "3000K" track heads side by side. If you can see a difference, that batch is not matched. On a 50-meter track, you'll see it on the wall.
- Driver access. Ask if the driver can be replaced without replacing the entire track head. If yes, request a spare driver with the order. That single line item can save a project.
(I should add that this applies to recessed downlights and decorative pendants too. But track lighting is where manufacturers get away with the most because there are so many connection variables.)
At a project in 2023, a wrong track fitting wasn't caught until installation. The redo cost the installer just over $22,000 and delayed the opening by three weeks. I mention that not to scare you, but to remind you that small details have large consequences.
Small Order, Big Leverage
Now for the part that gets me into arguments with people who treat small orders as a nuisance. The conventional wisdom in sourcing is that big buyers get preferential treatment. Everything I'd read about procurement said the same thing. In practice, I've found the opposite is often true.
Small orders are where you can be thorough. On a 5,000-unit purchase, you rely on statistical sampling and hope your batch evaluation is good. On a 30-unit order, you can inspect every single unit before it goes to the customer. A small order has better traceability, not worse.
If a table lamp supplier treats your small order as an inconvenience—if they won't share photometric files or quote a reasonable lead time for two samples—then they're telling you how they'll behave when your order grows. Believe them.
I know what you're thinking: Easy for you to say. Artemide is a premium brand with in-house labs. I'm just comparing catalog prices. That's fair. But I'm not saying you need a lab. I'm saying you need a threshold. Use the same questions whether you're buying one Artemide Dioscuri wall light or 40 track heads. What is the actual luminous flux? Is there an LM-80 report? Can I get a physical sample before a production order? If a supplier can't answer those questions, no contract price is low enough to make it a good deal.
The Short Version
A good specification is not a punishment. It's a tool that lets a small customer compare apples to apples. And it's the best filter for finding serious lighting partners. I've never regretted losing a supplier who couldn't provide basic documentation. I've regretted accepting a fixture that looked fine in a sample but couldn't survive a week of actual use.
So next time you receive a table lamp supplier catalog or three track lighting manufacturer quotes, ignore the glossy photos for a minute. Open the spec sheet. Ask for the LM-80 report. Order a sample. If the supplier treats you with respect when your order is small, imagine what they'll do when it is not.
