-
Who This Checklist Is For
-
Step 1: Put Zigbee Compliance Requirements on the Table Before You Get a Quote
-
Step 2: Match Sensors to Fixtures by Model Number, Not by Category
-
Step 3: Run the LED Driver Compatibility Check Against Dimmer Limits
-
Step 4: Lock the Spec Sheet — Don't Order From Memory
-
Step 5: Calculate Total Landed Cost, Not Unit Price
-
Step 6: Run the Small-Order Test — Even if the Vendor Groans
-
Step 7: Do an Unannounced Pre-Ship Inspection
-
Final Notes — Things That Bite Buyers
Who This Checklist Is For
If you're sourcing lighting controls from a wholesale supplier — Lutron motion sensor switches, Lutron LED lighting products, dimmers, track lighting systems — this checklist is for you. I focus on contractors and distributors placing orders between 50 and 5,000 units.
I've been handling lighting controls orders for 9 years, starting in 2016. I've personally made (and documented) 6 significant mistakes, totaling roughly $14,200 in wasted budget. After each one, I added a line to this list. Right now it has 7 steps. The full pass takes about 20 minutes.
Don't skip any of them. I learned that the hard way, multiple times.
Step 1: Put Zigbee Compliance Requirements on the Table Before You Get a Quote
Zigbee compliance requirements are what most buyers handle last — and that's exactly why they become the most expensive problem. Requirements vary wildly between projects because of regional regulations, project specs, and the end customer's jurisdiction.
Back in September 2022, we had a shipment go sideways: 132 occupancy sensors with Zigbee modules. The client assured us their gateway would handle any Zigbee 3.0 device. Three weeks later, when they tried to commission, the gateway was locked to a specific profile. Three weeks of back-and-forth, plus an expedited air freight redo that cost $1,200 in shipping alone.
What I do now: get the project spec in writing (actual spec sheet, not "the client said"). Cross-reference against the manufacturer's declaration of conformity. Ask the OEM three questions before you even request a quote:
- What Zigbee profile is this batch actually on?
- What firmware version, and will it work in an environment with deployed devices that are 3+ years old?
- What testing lab issued the certification you're handing me?
If the vendor can't answer those three questions, that's a deal-breaker. Full stop.
Step 2: Match Sensors to Fixtures by Model Number, Not by Category
This is the dumbest mistake I've made. "Occupancy sensor" and "compatible occupancy sensor" are not the same thing.
In my first year (2017), I did the classic: submitted a 340-piece sensor order meant to work across 3 different fixture models. I ordered by category — "PIR sensor, LED fixture, should be fine, right?" Wrong.
One fixture model had a 0-10V dimming control interface. The sensor didn't support it. Another used line-voltage switching. Took me three days to figure out why a third of the installs were failing.
What I do now: create a cross-reference table. Fixture model on the left, driver model in the middle, specific sensor number on the right. Get an electrical engineer to sign it off. Not me guessing.
Lesson from this one: you track by model, not by category.
Step 3: Run the LED Driver Compatibility Check Against Dimmer Limits
I'm not a controls engineer, so I can't speak to the underlying math of dimmer compatibility. That's not my expertise. I'd recommend consulting a licensed electrical engineer for the calculations.
What I can do from a procurement perspective is the routine: take the LED driver spec sheet, take the dimmer compatibility list, and cross-check them line by line. If the dimmer says 150W LED load and your project loads 40W, you're fine. If it says 60W and you're stuffing 8 drivers at 20W each, you're not.
We lost about $3,400 on a batch in 2020 because of this exact gap. The client wanted to mix different LED driver brands to save on cost. Five different drivers, one dimmer. Two of those drivers worked. The other two flickered or wouldn't dim at all.
This step happens regardless of how tedious it feels. Two hours with a spreadsheet is cheaper than a failed install.
Step 4: Lock the Spec Sheet — Don't Order From Memory
Memory is the most expensive tool in procurement.
We got three rejections in Q1 2024 for the same reason. Each time, I had three projects running with three different specs, and I pulled the wrong one from two weeks ago when submitting the order.
After the third rejection, I built the pre-check list. One page, spec sheet attached. Now: before I submit any order, the procurement coordinator has to upload a current spec sheet PDF with a date stamp from within the last 30 days. No exceptions. No "I think it was 14 inches."
Simple. Takes about 90 seconds. Saves more errors than any other line item on this list.
Step 5: Calculate Total Landed Cost, Not Unit Price
Unit price comparison lies to you. Here's why.
Most distributors quote you FOB somewhere, which excludes freight, duties, and their own handling fees. So I always add these back in before comparing:
- Ocean or air freight
- Duties and customs clearance
- Expediting fees if you need it at a specific time
- Warehousing costs if it sits before shipping to the client
- Break-bulk or minimum-order penalties on smaller quantities
Last year, on a single shipment, we paid an extra $2.40 per unit because we needed it faster for a client install date. Quoted unit price looked amazing. Total landed cost was higher than the local supplier we almost passed on.
Step 6: Run the Small-Order Test — Even if the Vendor Groans
I'm going to be blunt about this one: even if a vendor acts annoyed by your 12-piece test order, place it anyway.
When I was starting out, the vendors who treated my $200 orders seriously are the ones I still use for $20,000 orders. Small doesn't mean unimportant. Small means potential.
In 2019, we wanted to source 8 sensors for a site test. One vendor refused: "MOQ 500" — no flexibility, no sample program, nothing. We went with a different supplier who took a test order of 8 units. That project later needed 4,200 units. Guess who got the order.
Small-order treatment is the single biggest predictor of who's going to actually pick up the phone when a $40K project goes sideways. Test before you scale. Always.
Step 7: Do an Unannounced Pre-Ship Inspection
Nobody does this step. Most procurement teams stop at the order confirmation.
But we added it after a process gap in April 2023. We placed an order for Lutron motion sensor switches — 120 units for a Midwest distributor client. The factory shipped with mislabeled units: outer cartons had the correct model, inner boxes had a different one. When the client received them, they almost sent 120 units back.
We caught it before the shipment left our warehouse because we had someone scan 10% of the units by hand with a barcode reader. Two of those 10% were wrong. That scales to roughly 20% of the batch. Dodged a bullet — we were one click away from a full RMA cycle and a credibility hit with a distributor who buys yearly.
So now: someone manually checks at least 10% of the physical product against the purchase order before anything ships. Not the packing list — the actual product, model number, and quantity against what the PO says.
Final Notes — Things That Bite Buyers
The steps in this checklist are more boring than most people want to hear. They're also cheaper than most people want to admit.
A few reminders:
- This was accurate as of April 2026. Zigbee compliance requirements and LED driver compatibility shift fast. Verify your reference documents quarterly.
- Don't take "the vendor assured me" as documentation. I've made the mistake of trusting verbal confirmations. Get the paper trail, always.
- If a step feels skippable, double down on it. That feeling is what caused every error I've made.
- Know your limits. I'm not an electrical engineer, and I stopped pretending to be. When in doubt on any engineering-adjacent step, bring in someone who actually knows.
The most frustrating part of this whole process is how often the same errors come back in slightly different shapes. You'd think written specs and checklists would catch everything, but interpretation varies wildly across teams and factories.
I learned these lessons the expensive way. I hope you don't have to.

