The 7-Point Lighting Quality Checklist I Run on Every Fixture

I'm a quality compliance manager at a lighting company. Roughly 40 to 50 fixtures cross my bench every month—wall sconces, floor lamps, pendants, track heads. In 2024, I rejected about 14% of first-article samples before they got production approval. This article is the exact checklist I run on each one.

If you're an architect, interior designer, or someone specifying lights for a commercial project, this is for you. Seven checks, in order, each with a clear pass or fail outcome. Once you've done it a few times, a full evaluation takes about 20 minutes per fixture. I've built this list from years of reviewing Flos classics like the 265 wall lamp and the Toio floor lamp.

1. Verify the photometrics yourself

Datasheets lie. Not maliciously, but they're written from lab measurements on a single prototype under controlled conditions. What most people don't realize is that production units drift from the datasheet—sometimes noticeably—because of variations in LED binning, diffuser material, and driver tolerance.

So step one is simple: measure. I use a calibrated lux meter and a spectroradiometer. You don't need both for every fixture, but at minimum check:

• Luminous flux in lumens at nominal voltage
• Correlated color temperature (CCT)
• CRI (Ra), and R9 if you're lighting food, art, or skin

The Flos Toio lamp is a perfect example. Designed by Achille Castiglioni in 1977, it's an indirect floor lamp where a small lamp fires into a reflector, and the light passes through a perforated metal baffle. The result depends heavily on that baffle's alignment. In 2022 I tested a batch of eight Toio units—three had the beam cocked about 4 degrees off center. On a tall floor lamp aimed across a living room, that's the difference between lighting the seating area and lighting the edge of the bookshelf.

2. Check color consistency across the batch

When you're specifying lighting for a hotel corridor or a retail wall, you're not choosing one fixture—you're choosing twenty, or eighty, or three hundred. And people are remarkably good at noticing color temperature differences between adjacent lights.

The industry tolerance for color differences is expressed in Delta E (ΔE). For brand-critical colors, you want ΔE below 2. Between 2 and 4, a trained observer sees the difference. Above 4, most people can see it.

I've measured batches where unit A runs at 3000K while unit B sits at 3150K out of the box. Individually, both are within spec. Side by side, one is warm white and the other is slightly green. In our Q1 2024 quality audit, we rejected a batch of wall sconces—120 units, maybe 140, I'd have to check the record—because tint variation across the run exceeded our spec. The vendor had switched LED board suppliers without notifying us. The rework cost us a $14,000 chargeback and pushed the install schedule by a week.

3. Inspect the finish and mechanical build

This is where the Flos 265 wall lamp deserves attention. The 265 is a wall-mounted lamp with a curved steel stem and a weighted, adjustable shade. It looks simple, but the details matter:

• The cable runs inside the arm and determines how smoothly the shade glides. If it's pinched, the shade won't hold its angle.
• The weighted base needs a uniform painted finish, free of orange peel or small voids.
• The strain relief where the cable exits the arm has to be solid. A weak strain relief means the wire fatigues after weeks of pivoting.

Run a fingernail along the seams. Look at the diffuser edge under bright light. Open and close the switch forty times. These small details separate a fixture that looks good in the catalog photo from one that holds up on a job site.

(Should mention: check the mounting plate too. Wall fixtures should be installable by one person. If the mounting hole pattern doesn't align with a standard junction box, your client's electrician will be billing extra hours.)

4. Test dimming and flicker at low levels

Many LED fixtures dim acceptably to 20%. Below that, strange things start happening: strobing, color shift, or a low-frequency flicker that you perceive but can't consciously see.

Here's a trick that costs nothing: open your phone camera and point it at the fixture while you turn the dimmer down. Most phone sensors are sensitive to LED flicker that the human eye ignores. Moving bands or shimmering patterns on the screen mean the driver isn't producing clean high-frequency output.

If the fixture will ever be on camera, this is critical. During a spotlight photoshoot, a flickering fixture will produce banding in video and erratic exposure in burst shots. No amount of post-processing can fix it. I once bought a 'pro-grade' LED lamp from another brand—I won't say which—and its flicker was so bad that anything above 1/250s shutter produced visible stripes.

5. Evaluate the beam: the skin spotlight test

No datasheet number will tell you whether a fixture makes people look good. You have to look at it with your own eyes.

My method: place the fixture at face height, aim it the way it will actually be used, and stand in front of it. Try it with a light-colored shirt and a dark one. Notice:

• Spectral highlights on your forehead and nose—too harsh?
• Deep shadows in the eye sockets—a bad sign for a conversational setting.
• A flat, waxy quality to skin—likely a low R9 and a poor diffuser.

This 'skin spotlight' evaluation matters most in hospitality and residential settings. It's also essential in photography. A true spotlight photoshoot needs a fixture with a controllable beam, not a source that flushes the whole scene with flat light. Being able to narrow the beam and dim without shifting color is what separates useful fixtures from decorative ones.

6. Calculate total cost of ownership, not just unit price

I've said this until I'm blue in the face: the sticker price is not the cost. If a fixture has a non-replaceable integrated LED module, you throw the entire product away when the emitters age out. A fixture with a standardized light engine and a replaceable driver can be serviced in minutes—at a fraction of the cost.

A $500 fixture that fails in year four is more expensive than a $650 fixture that runs for a decade with a $40 driver replacement in year seven.

The $500 quote turned into $800 after shipping, setup, and rework fees. The $650 all-inclusive quote was actually the cheap one by the time we finished the install.

I learned that lesson in 2021 on a project where the 'budget' fixtures had a driver compatibility issue with the client's dimmer system. We spent more in labor swapping, adjusting, and testing than the price difference between the two options would have been. This was accurate as of Q1 2025. The market changes fast, so verify current pricing and LED module lifetimes before you budget a large project.

7. Run a 24-hour burn-in test

Most people skip this step. I don't, and I do not plan to change that.

Run the fixture at 100% output for at least 24 hours before you approve it. Then check:

• Heat at the driver enclosure and shade. If it's too hot to hold your palm against it, that's a problem.
• Discoloration on the diffuser or housing.
• Light output drop over the period.
• Any odor—the smell of hot electronics means the fixture is cooking itself.

Heat is the leading cause of LED failure. I once specified an interior sconce without checking its thermal behavior in a ceiling corner—it ran 25°C hotter than in the vendor's open-air test. That cost us a full re-install.

One of my biggest regrets from earlier in my career: not running a burn-in on a small batch of recessed lights because the client was rushing. Three of fifteen units visibly dimmed after two weeks. We were on the hook for replacement, and the client relationship never fully recovered. So glad I implemented a mandatory burn-in protocol in 2022. We've caught three driver failures and one connector issue before shipment since then—probably saved us six figures in field rework.

Common mistakes to avoid

These are the errors I see even from experienced specifiers:

Trusting the spec sheet without testing. The manufacturer can substantiate their claims and still ship units that don't match the sample. Per the FTC's advertising guidance, claims must be truthful and substantiated—but that means the claim has lab support, not that every production unit is identical.

Sampling only one unit. Batch consistency is the real question. Take at least one sample from the beginning, middle, and end of the production run.

Ignoring the driver. The driver is the most failure-prone component in an LED fixture, and it's the most likely part to be incompatible with the dimmer you specified. Test the combination you will actually install.

Forgetting the environment. A fixture rated for open-air operation may behave differently in a sealed cavity or a humid bathroom. Match the product to the real installation environment, not the catalog's ideal conditions.

And one practical note on documenting your work. I keep spec sheets and inspection notes in a shared folder and use Spotlight Search on my Mac—Cmd+Space, or the magnifying glass icon in the menu bar—when I need to pull up a datasheet without digging through folders. If you're managing dozens of fixtures across multiple projects, being able to find that critical file in two seconds instead of twenty minutes adds up.

Write down what you find. The fixture you don't document is a fixture you'll have to re-test next month.