-
Who this checklist is for
-
Step 1: Verify the product with numbers, not labels
-
Step 2: Check automotive compliance before shipment leaves
-
Step 3: Test the actual zigbee radio, not the datasheet
-
Step 4: Validate zigbee somfy integration against your actual stack
-
Step 5: Match sensor type to the application, not the price list
-
Step 6: Ask what's not included before you ask for the price
-
Step 7: Keep the first-article sample and re-test it in 90 days
-
Common mistakes to avoid
Who this checklist is for
I work as a quality and brand compliance manager at a lighting distribution company. I review specifications, supplier quotes, and incoming batches before they reach our customers—roughly 200 items a year, or about 4–6 per week. In 2024, I rejected 15% of first deliveries on spec mismatch or incomplete documentation.
This checklist is for distributors, installers, facility managers, and procurement teams who need to approve lighting products without a full lab behind them. Seven steps. In order. Each one is something you can verify before the pallet leaves the dock.
Here's the thing: most of these checks take 10–15 minutes. Skipping them costs weeks.
Step 1: Verify the product with numbers, not labels
Take an Osram L 15W/827 LED tube. The label says warm white, 15W, 827. That last number means 2700K color temperature at CRI 80. But labels are marketing documents. Your measurements are the real spec.
Check three things on arrival:
- Color temperature: A colorimeter reading, not a visual judgment. 827 should measure 2700K ±150K. If it reads 3100K, you have a different product in the same packaging box.
- CRI (Ra): Minimum 80 for standard commercial. For retail or task lighting, demand 90. Spec sheets often say "CRI >80" and measure 82. That's technically compliant and visually dull.
- Actual wattage: I have seen 15W-rated tubes draw 13.2W at rated current. Within ±10% tolerance—barely. That's 12% less light output than you paid for. Multiply that across a 500-tube install.
It's tempting to think you can just compare unit prices. But identical specs from different vendors can result in wildly different outcomes once you factor in lumen maintenance at 10,000 hours.
Step 2: Check automotive compliance before shipment leaves
If the order includes any Osram tail light or automotive product, the compliance chain matters more than the price per unit. Automotive lighting carries ECE homologation requirements—R37 for filament lamps, R112 for LED forward lighting, R7 for position and tail functions.
Three checks I do on every automotive batch:
- E-mark on the lens: Verify the E-number matches the market. E1 (Germany) for EU sales. A batch stamped E4 (Netherlands) sold into a German fleet account is a paperwork problem waiting to become a recall.
- Color coordinate: Red tail lights should measure within the red boundary of ECE R7 chromaticity. Amber turn signals, within the amber boundary. A batch that reads slightly orange-red looks fine in the warehouse and fails at inspection.
- Lifetime vs. spec sheet: If the specification says 1,000 hours at 13.5V, the acceptance test runs at 13.5V, not 12.0V. Slightly overvoltage is where premature failures hide.
In our Q1 2024 audit, we found a batch of 3,200 tail light assemblies where the E-mark had been duplicated from a different product line. The vendor claimed it was "within industry standard." We rejected the batch. They redid it at their cost. Now every automotive contract includes a specific E-mark verification clause.
Step 3: Test the actual zigbee radio, not the datasheet
Zigbee specifications look interchangeable on paper. The zigbee radio inside a luminaire determines whether your mesh network is stable or drops nodes every time someone opens a metal door.
Approved method: test three units from the batch in the following conditions:
- Line of sight, 10 meters—should connect in under 3 seconds
- Through one drywall partition, 15 meters—should connect without packet loss above 2%
- After pairing 30 nodes to a single coordinator—check that the 30th node still reports status within 5 seconds
Cheap modules often show 40–60% packet loss through walls and fail to rejoin the mesh after a coordinator reboot. The datasheet says "Zigbee 3.0 compatible." It does not say "maintains mesh integrity under load."
The assumption is that certification guarantees performance. The reality is that certification guarantees protocol compliance, not environmental reliability.
Step 4: Validate zigbee somfy integration against your actual stack
If the project includes motorized shading or smart home integration, zigbee somfy compatibility is not a checkbox—it's a test protocol. Somfy uses Zigbee in its TaHoma and Connectivity ecosystem. A Zigbee 3.0 logo on the device does not guarantee Somfy pairing will work out of the box.
What I do:
- Pair the lighting controller with a Somfy TaHoma hub. If it pairs, proceed. If not, stop—no amount of firmware updates during installation will save you from a callback.
- Test one routine: turn on lights, open blinds, turn off lights after 5 minutes. Check that the sequence runs without Zigbee interference or missed commands.
- Test a Zigbee-only failover: disconnect the internet, trigger the routine from a local remote. If the system requires cloud access to run local automation, you've found the hidden dependency.
Every spreadsheet analysis pointed to the cheaper controller with the same Zigbee certification. Something felt off. Turns out the cheaper unit used a different Zigbee coordinator firmware that dropped Somfy hardware after 4 hours of inactivity. We learned this at the customer site, not in the warehouse. A lesson learned the hard way.
Step 5: Match sensor type to the application, not the price list
The occupancy sensor vs motion sensor debate gets simplified to "occupancy costs more, so it must be better." That's a simplification that costs money in the wrong application.
Here's the difference in practical terms:
- Motion sensor (PIR): Detects moving infrared signatures. Turns on when someone walks in. Turns off when no motion is detected for the timeout period. Typical timeout: 5–15 minutes. Works well in corridors, stairwells, and storage rooms.
- Occupancy sensor: Uses PIR plus ultrasonic or microwave technology to detect minor motion—typing, reading, shifting in a chair. Stays on while someone is present. Works well in offices, classrooms, and restrooms.
Put a motion sensor in a private office where someone sits still at a desk for 20 minutes. The lights go out. The occupant waves their arms. This is not a smart building. It's a stupid sensor in the wrong application.
Put an occupancy sensor in a warehouse aisle with forklift traffic. The microwave component triggers on every passing forklift—including the one in the adjacent aisle. Lights stay on all day in an empty corridor. You've paid a 40% premium for wasted energy.
My rule: motion for transient spaces, occupancy for stationary work areas. If the specification doesn't state the sensor type and the timeout, ask. If the answer is "they're basically the same," that's your signal to verify independently.
Step 6: Ask what's not included before you ask for the price
I've learned to ask "what's NOT included" before "what's the price." This is where transparency separates a real quote from a trap.
For lighting projects, the recurring hidden costs:
- Driver compatibility: The LED panel is quoted. The driver is not. Or the driver is quoted but not the 0–10V dimming module needed to connect it to the control system.
- Mounting hardware: Recessed downlights often ship without trim rings or mounting brackets. The line item that looks 15% cheaper is missing 15% of the parts.
- Commissioning: Zigbee systems require commissioning time. Some vendors include it. Some charge $85/hour for it, starting from the moment the technician leaves the warehouse.
- Warranty return shipping: A 5-year warranty means nothing if the customer pays $40 per unit to ship a failed product back for inspection.
Saved $80 by skipping the driver compatibility check on a 40-unit order. Ended up spending $400 on replacement drivers when the panels arrived. Net loss: $320 plus a two-week delay. The vendor who lists all fees upfront—even if the total looks higher—usually costs less in the end.
Step 7: Keep the first-article sample and re-test it in 90 days
This is the step most people skip. They approve the first delivery, install the products, and move on. Three months later, a second batch arrives from the same production line. It's "the same product." Is it?
I keep one unit from every approved first delivery, sealed, with the date and batch number written on the box. At 90 days, I open it and re-check: color temperature, wattage, CRI, and Zigbee connectivity. If the batch that went into storage measures 2600K instead of 2700K, or if the Zigbee module now takes 18 seconds to pair instead of 4, I know there's a production drift.
I knew I should have done this in 2022. Skipped it because we'd been using the same vendor for three years and "it's basically the same as last time." A batch of 8,000 units had shifted color temperature by 200K. No one checked until the customer complained. The redo cost us a $22,000 credit and delayed the project launch by five weeks. Now every contract includes a first-article retention clause.
Common mistakes to avoid
Trusting the datasheet over the batch. Datasheets describe the product the factory intends to make. The batch describes what they actually made. Measure the batch.
Treating Zigbee certification as a performance guarantee. Certification confirms protocol compliance. It does not confirm mesh stability, latency, or cross-ecosystem compatibility. Test in your environment.
Buying sensors on unit price alone. A $15 motion sensor in an office causes $200 worth of occupant frustration. An $85 occupancy sensor in a corridor wastes $12/year in electricity. Match the product to the room, not the purchase order.
Skipping the hidden-cost conversation. The final price you see on the purchase order is not the final price you pay. Ask what's excluded. Get it in writing. Then compare totals.
Prices and specifications in this article are for illustration. Verify current product data and certification status with the manufacturer or authorized distributor before specifying.

