Ask ten lighting buyers for the right spec and you'll get twelve answers. Not because they're wrong—because "right" depends on where the lamp is going. A tube aimed at a warehouse ceiling has almost nothing to do with a tail lamp bolted to a car, and neither has much to do with a Zigbee control system running an office. Same industry. Completely different conversations.

I review around 200+ lighting SKUs a year as part of a compliance function, and I've stopped giving generic advice. Instead I break it down by scenario. Here are the three I get asked about most, and what I actually check in each.

Scenario A: Commercial / Facility Retrofit

This is where most questions land—office relights, warehouse upgrades, hospitality refits, schools. The products are linear LEDs, panels, downlights, high bays. A common example is the Osram L 15W/827 LED tube: 15 watts, and the 827 code means CRI 80+ with a nominal color temperature around 2700K. Warm, residential-adjacent white. Fine for a hotel corridor, questionable for a freezer aisle.

What do I actually review on a tube like that?

First, lumen output against the target. A 4-foot fluorescent replacement typically lands around 1,800–2,000 lm. If the datasheet says 1,500, that's not a direct replacement—it's a downgrade wearing the same tube shape. Second, color consistency. The 827 number tells you nominal CCT and CRI, not consistency. Look for macAdam 3-step binning or better. Two tubes side by side in the same fixture should look like the same tube (note to self: I owe the team that one-page binning guide).

Third, driver quality. This is the part buyers skip. The LED chips themselves are pretty reliable across brands now. The driver is where you get flicker, early failures, and dimming surprises. Ask for the driver datasheet, ask about minimum operating temperature, ask about compatibility with the existing dimming protocol (0-10V, DALI, and TRIAC are not interchangeable, whatever the catalog implies).

Fourth, LM-79 and LM-80 reports. If a vendor can't produce an LM-80 report on the LED package they're using, they're repeating a marketing number—"50,000 hours"—without evidence behind it. I'm not asking for a guarantee. I'm asking what the test data shows.

One thing I've stopped caring about: brand name alone. It took me a few years and about 180 SKU reviews to understand that a spec sheet from a big manufacturer and a spec sheet from a small one can both be accurate, or both be useless. What matters is whether the numbers are testable.

We also didn't have a formal sample-testing step for retrofits initially. That cost us. The first time a 500-tube order arrived with a visible color shift—two-thirds of the pallet read bluish next to our reference lamp—we'd already installed 80 of them. Now every retrofit order gets a 5-piece pull tested on a lux meter before the pallet leaves the supplier.

Scenario B: Automotive Lighting

Different world entirely. When the topic is an Osram tail light or any other automotive lamp, the spec conversation isn't about preference—it's about whether the product is legal to put on the road.

Automotive lighting is one of the few categories where the useful spec is defined by regulation. Tail lamps, brake lamps, reverse lamps, side markers—each has an ECE category (R37 for filament, R128 for LED, R148 for signalling devices). Color coordinates, intensity, and viewing angle are prescribed, not negotiated. If you're buying aftermarket for a passenger vehicle, "fits" and "legal" are two different questions.

What I check here:

  • Base and voltage—an H4 is not an H7, and a 24V truck bulb is not a 12V passenger bulb
  • Color point—the visible color has to fall inside a specific chromaticity box, and red is not "sort of red"
  • Thermal behavior—especially for LED retrofits going into housings designed around a filament source; heat has to leave the lamp, not cook it
  • Labeling—if the packaging doesn't say the product is road legal in your market, treat it as off-road only

I made a classic mistaken assumption early on: I assumed an aftermarket LED bulb that fit an original socket would perform like the OEM piece. It won't, necessarily. Different thermal design, different beam pattern. A socket is a mechanical fitment, not a performance guarantee. That lesson cost us an exchange on a 400-unit order and a week of explaining to a client why their "upgrade" actually reduced output.

Scenario C: Smart Lighting Controls

This is where the most confusion lives, and where spec numbers matter least. Three things get mixed up constantly: the radio, the ecosystem, and the sensor.

The radio: Zigbee

When a datasheet lists a Zigbee radio, that tells you the physical layer—2.4 GHz mesh, IEEE 802.15.4. It does not tell you what the device talks to. There are pre-Zigbee 3.0 profile variations, manufacturer-specific clusters, and certified vs non-certified implementations. Zigbee 3.0 certification is the closest thing to a "safe default" you can look for.

Practical check: ask for the certification ID. If they can't give you one, they're using Zigbee silicon without going through the certification program, and "works with our gateway" is the only claim they can actually defend.

Zigbee and Somfy

Somfy makes motorized blinds, shutters, awnings—and their TaHoma or Connexoon hubs can bridge into Zigbee. When someone asks about Zigbee Somfy compatibility, the honest answer is: depends on which Somfy controller and which Zigbee device. Somfy's own Zigbee 3.0 line works inside the TaHoma ecosystem directly. Third-party Zigbee sensors often need the right integration profile, not just the Zigbee logo on the box.

This trips up installers. They buy a Zigbee occupancy sensor assuming it pairs with the Somfy hub because both logos appear on the packaging. Half the time it doesn't, because the sensor is on a different profile. Read the integration list, not the logo list. Honestly, that's the whole point.

Occupancy vs motion sensors

This is the spec question I get most, and the one where choosing wrong actually costs money in energy code compliance.

Occupancy sensors vs motion sensors—the names get used interchangeably in marketing, but they aren't the same device:

  • Motion sensor: detects a distinct change in infrared or microwave signature. Great for hallways, stairwells, parking garages. Turns off when the room goes still. Frustrating in an office where people sit and type.
  • Occupancy sensor: picks up smaller signals—breathing-level motion, minor shifts, sometimes ultrasonic or dual-tech (PIR plus ultrasonic). Required by many energy codes in spaces where people stay stationary for long periods: offices, classrooms, restrooms.

Put a motion-only sensor in an open-plan office and the lights go out while people are at their desks. Put an occupancy sensor in a warehouse aisle nobody walks through for an hour, and you're burning energy on an empty corridor. The technology has to match the space. Dual-tech is usually the right call where compliance matters; single-tech is often fine for corridors and parking.

One more thing worth mentioning on the marketing side: if a category datasheet leans on words like "eco-friendly" or "recyclable," remember the FTC Green Guides require those claims to be substantiated. "Recyclable" means recyclable in areas where at least 60% of consumers have access to recycling—which for lighting products is not always true, since drivers and legacy tubes complicate the waste stream. Per the FTC's 16 CFR Part 260, vague environmental claims without evidence are a compliance risk for the seller, and a red flag for the buyer.

How to Figure Out Which Scenario You're In

Three quick questions before you even open a datasheet:

Is the product regulated? Automotive lighting is. If the answer is yes, start with the ECE or SAE category, not the brand catalog.

Is it a like-for-like swap or a system change? A tube replacement is like-for-like—focus on electrical and photometric specs. A smart lighting project is a system change—focus on the ecosystem and the integration list before brand.

Is the space occupied by people who sit still or people who move through? That decides motion vs occupancy, which is where a surprising number of energy code problems start. It's not a lighting question. It's a usage question.

The one thing that cuts across all three: ask for evidence, not adjectives. "Industry standard" and "high quality" are not specs. A lumen figure, an LM-80 report, a certification ID, or a regulation number—those are.

I've said this to more suppliers than I care to count, and the ones who get it—the ones who actually send the data—are usually the ones who take a $200 sample order as seriously as a $20,000 rollout order. That pattern matters more than people think. The buyer placing a sample order for a small school district today is often the same person writing the spec for a 40-building rollout next year. I've watched vendors treat small orders like an inconvenience and lose the volume later over it.

There's no universal right answer on lighting specs. There's only the right answer for your scenario. Figure out which one you're in, then ask the questions that actually apply.