The cheapest lamp on the quote is almost never the cheapest lamp in the building. After six years of tracking every lighting invoice we pay, I think the gap is wider than most procurement teams are willing to admit.
I manage the lighting and electrical consumables budget for a 400-person facilities services company — roughly $180,000 a year across 23 client sites (as of January 2025; the number drifts with fuel surcharges). I've logged every order, every credit note, and every "can someone come back, the lamps are flickering" call in a spreadsheet my team now unkindly calls the TCO file. It started as a way to defend myself in budget reviews. It's now the thing that decides which vendors we keep and which we quietly stop calling.
Here's the argument I want to make: for commercial lighting, the purchase price of the product is a rounding error next to the cost of getting it wrong. And the industry has moved far enough in the last few years that a lot of the folk wisdom we still buy on — cheapest equivalent is fine, connectors are connectors, a lumen is a lumen — is out of date. What was best practice in 2018 doesn't survive contact with a 2025 building.
Argument one: you're not buying lamps, you're buying callbacks
When I audited our 2023 spending, something annoyed me. Of the $212,000 we spent on lighting that year (yes, over budget), about 62% went to labor, access equipment and re-visits. Product was the small side of the ledger. Scaffolding and scissor lifts cost more per day than most of the fixtures we were installing on them.
That single number flips the whole calculation. A $9 connector versus a $4 connector is not a 125% price increase. On a 300-fixture job it's $1,500 extra, set against a single $220 emergency callout — and you rarely get just one callout. Do that math honestly, once, and you stop negotiating on unit price and start negotiating on documentation. Which spec sheet actually tells me the failure modes? Which supplier answers the phone in month fourteen?
I learned this the expensive way. Saved around $720 by going with the cheaper downlight connectors on a 180-fixture office refit (Spring 2023, so treat those prices as historical). Ended up spending just under $1,900 on two return visits, a batch of replacement connectors, and a very awkward conversation with the client's office manager. Net loss, call it $1,100, plus the credibility hit. The connectors themselves weren't defective. They just weren't the ones the installer had quoted against, and nobody checked.
Which is why downlight connectors stopped being a boring line item for me. The connector is where installers improvise. It's the part with the least glamour and the most variation across batches, and it's the part that decides whether your fit-out passes inspection the first time.
Argument two: the spec sheet is written for a room that doesn't exist
What most people don't realize is that the lifetime figures on a lighting datasheet are measured under conditions your ceiling void will never replicate. Ambient temperature, airflow, driver placement — all of it is controlled in the test and none of it is controlled in your building.
I used to believe cheap fixtures failed because the components were cheap. Then I sat down with our maintenance logs for two years and the causation ran the other way more often than I expected. People think price causes failure. In practice, installation conditions cause failure, and price only changes how quickly and how loudly it happens. A driver jammed above insulation and cooking at 20 degrees above its rated ambient will fail whether it cost $18 or $60. The $60 unit just fails later, and — this matters — fails with a documented warranty claim path instead of a shrug.
This is also where I've become a bore about substantiation. Per FTC advertising guidance, performance and environmental claims have to be truthful and backed by evidence — including "recyclable" claims under the FTC Green Guides (16 CFR Part 260). So when a vendor hands me a sheet promising 50,000 hours and full recyclability, I ask which test, at which ambient, and who verified the recycling pathway. Half the time the answer is a PDF from 2019. That's not a dealbreaker, but it does tell me what my support will look like at year four.
Argument three: the weird orders teach you the most
The strangest requests in our system have taught me more about supplier quality than the standard office refit ever did. Three examples from the last eighteen months:
- Black downlights. A retail client specified downlight black trims throughout — dark ceilings, exposed services, the whole look. We ordered "black" from three suppliers and got three different blacks, plus one batch that read charcoal under 3000K and brown under 4000K. Nobody died, but we ate the cost of a re-order on 40 units because color matching wasn't in the original spec. Now it's a written requirement, with a physical sample signed off before the PO goes out.
- The oven lamp question. A kitchen equipment client asked us to sort lamps for a batch of commercial ovens. Our buyer spent a week reading osram 25w halopin g9 oven lamp reviews online and came back with a verdict on color temperature. Wrong question. The reviews tell you what the light looks like in someone's kitchen. They don't tell you whether the appliance's thermal envelope is within the lamp's rated range, or whether the fitting is standard enough that you can still buy replacements in three years. That's what actually costs money — an obsolete lamp in a working oven.
- Grow lights. We picked up a small horticultural install through a client's wellness fit-out, and I had to learn what makes a good grow light almost from scratch. The answer isn't watts. It's delivered light over the actual photoperiod, uniformity across the canopy, and how many hours a year the thing runs. Grow lights run 12–18 hours a day, which changes the TCO math entirely — a 6% efficiency difference compounds into real electricity spend, and a driver failure mid-cycle can cost you a crop, not a complaint.
Why does this matter to a procurement person? Because these are the orders where the spec is vague and the supplier's competence shows. Anyone can ship 200 LED tubes. Not everyone can tell you why the black trim differs between lots.
"That's just paying a premium for the logo"
The pushback I get most often is that standardization on a known brand is laziness dressed up as strategy. Fair challenge. I've seen plenty of brand premiums that bought nothing but a nicer box.
But that's not what I'm arguing for. I'm arguing for documented, repeatable supply: a product family that's still in the catalogue in four years, a driver specification I can hand to a maintenance contractor, and a support line that answers when a site in a different city needs a like-for-like replacement at short notice. That's why I skim Osram LED news announcements the way other people skim sports scores — not out of brand loyalty, but because I need to know early if a family we've standardized on is being phased out. A discontinued lamp is a re-lamping project I didn't budget for.
The secondary objection is time: "we don't have bandwidth for a TCO model." You don't need a model. You need one column. Track, per product family, the number of return visits in twelve months. That's it. Six months of that data will tell you more than any vendor presentation.
Where this leaves us
I'm not saying price doesn't matter. It matters enough that we still tender competitively and still push back on quotes. But I've stopped treating the unit price as the decision and started treating it as the entry ticket.
The industry has genuinely changed — better drivers, better thermal design, better documentation, and a much wider spread between the good and the merely cheap. What hasn't changed is the fundamental question: who's coming back when it fails, and at what cost?Buy the answer to that question, not the cheapest line on the quote.
