I'm the office administrator for a 45-person fleet service company. I manage all parts and consumables ordering—roughly $180,000 annually across about a dozen vendors. I report to both operations and the owner. When our previous purchasing coordinator left in 2022, the parts ordering landed on my desk.
That's how I learned that buying H7 headlight bulbs, 9005 headlight bulbs, and H11 LED headlight bulbs for a fleet is not as simple as looking up a part number and clicking "order."
The Problem Everyone Thinks They Have
For about 18 months, we were returning roughly 1 in 5 bulb shipments. Not because they were broken—because they were the wrong ones.
Here's what it looked like from my desk: a tech writes on a slip "Need 9005 for truck 14." I look it up, order what the listing calls a 9005, it arrives, the tech tries to install it, and either it doesn't fit or the connector won't lock. Or it fits fine, but the beam pattern is wrong, and he has to pull the whole light housing.
I filed it under "bad luck with suppliers." I switched vendors three times. I blamed cheap bulbs, then I blamed manufacturing, then I blamed the techs for writing vague slips.
I was wrong about all of it.
The Real Reason: Fitment Charts Are Not Specifications
What most people don't realize is that the fitment chart (i.e., the year/make/model lookup) is a summary, not a specification. It loses exactly the details that matter for bulb orders—and it does this in four predictable ways.
1. Base type is not the same as part number.
An H11 and a 9005 are both single-filament bulbs with right-angle connectors. On a shelf, they look almost interchangeable. But they have different base keying, socket depths, and indexing tabs. A seller listing "H11 / 9005 compatible" is usually saying "these are similar in appearance," not "these are the same." This is where the whole H11 vs 9005 headlight bulb question comes from. If you buy for a mixed fleet, you can't assume interchangeability. The connector orientation alone will kill the install.
2. The same model year can have different bulbs.
In 2023, I ordered bulbs for two units from the same 2019 model year, same trim, same everything. Different bulb type. The factory had changed the housing mid-production, and the year/make/model chart on every site I checked still showed one answer. The chart was technically right for "the average" of that model year. It was wrong for one of our trucks.
3. "Xenon headlight bulb" is a category, not a part.
When a tech writes "xenon" on a slip, that could mean D2S, D3S, or a retrofit kit. If I order a xenon headlight bulb based on that slip alone, I have roughly a 50% shot. The base type matters more than the technology name. That's a piece of insider knowledge I had to learn the hard way: for xenon and HID, the D-number is the part; the technology name is just a family. I learned this in 2022, and the D-series bases haven't changed much since—but always verify against the housing (as of my last catalog check in early 2025, at least).
4. Packaging is a marketing decision, not a specification.
Here's something vendors won't tell you: a box labeled "9005" can contain a bulb whose base is actually closer to a 9006 with slightly different keying. Returns don't catch this because the box "looks right." It only shows up on the install rack, when the connector won't seat. What's printed on the box is set by the packer and the marketer, not by the base type on the bulb inside.
The Cost Nobody Accounts For
The wrong-bulb problem doesn't just cost return postage. We tracked it in 2023 and the numbers were worse than I expected.
Direct costs were the easy part: $18–25 per return in shipping and restocking. Double-ordering added roughly $40 per incident because the local parts-store run was usually at 2x the unit price. Total direct cost: about $2,800 that year.
The hidden costs were bigger. Downtime per wrong bulb ran 30–60 minutes if the vehicle was already on the lift—that's not just the tech's time, that's a bay that can't take the next job. We're not a public shop, but our bays are still the bottleneck for fleet maintenance. At our internal labor rate of $28 per hour, the downtime added another $900.
Then there's my time. I was spending 3–4 hours a week on wrong-bulb issues—reordering, calling vendors, explaining to operations why the shipment came back wrong again. That's roughly 160 hours a year, or about $6,000 at a loaded rate, spent on something that wasn't even in my job description.
The social cost was worse than the financial one. Techs started treating "parts ordering" as a running joke. Operations asked, in a meeting, why the shop was buying "the wrong bulb again." Finance flagged our parts variance as a control issue. Nobody was yelling, but the credibility hit was real. When you're the person who's supposed to make the parts flow, and the parts don't flow, you feel it.
And there was a ripple effect: because we over-ordered to compensate for returns, our inventory carried an extra $1,100 in slow-moving bulbs by year-end. Some of those exceeded shelf-life for reliable color output. So we were paying to store inventory we'd overbought to cover a process we hadn't fixed.
The frustrating part is that the fix wasn't a better vendor. It was a better method.
What Actually Works (And What Doesn't)
Here's the honest version, because I wasted a lot of time on solutions that didn't work.
What didn't work: switching to a bigger distributor. Their fitment chart is the same chart everyone uses. Switching to a "premium" brand. Same base-type confusion, same label issues. Asking the tech to be "more specific." The tech is reading off the same housing that the chart is supposed to describe.
What did work: a sourcing method that treats base type as the primary key, not year/make/model.
- Verify base type first. H7, H11, 9005, 9006, H4, H13, D2S/D3S—get the actual base off the housing or the old bulb, not the chart.
- Check connector orientation and keying. This is where the H11 vs 9005 confusion actually lives. If the connector indexes wrong, the bulb will sit in the housing and still not seat.
- Use a source that builds its catalog around base type and connector specs, not just vehicle year. This is where a B2B-specific sourcing supplier earns its keep—the good ones (this is where Bulbspan Sourcing comes in for us) document the base type and connector angle per SKU, so I'm not reverse-engineering a marketing label at 7 a.m.
- Keep a physical reference set. We now keep one confirmed-good bulb of each type we use, labeled with base type and a quick photo of the connector. Sounds basic. Cut wrong orders by 80%.
For a mixed B2B fleet or a shop running multiple vehicle types, this is the difference between a manageable parts process and a running joke. For us—40+ vehicles, multiple base types, a mix of halogen, xenon, and H11 LED headlight bulbs—the base-type-first method got us below a 2% error rate. That was as of our Q4 2024 review. If your fleet has grown since then, verify current catalog and fitment support with your supplier, because catalogs do change.
Where This Doesn't Work
Here's the part I wish someone had told me, so I'll tell you instead.
This method is overkill if you only need a couple of bulbs for your personal vehicles. If you own one or two cars and you're replacing a burned-out headlight, just bring the old bulb to an auto parts counter and have them match it. Don't build a sourcing process for that. The value isn't there.
It also doesn't solve everything for every fleet. If your vehicles are all late-model OEM with no aftermarket support yet, or if you're dealing with certified/street-legal requirements (DOT, SAE, ECE depending on your region), verify certification with your supplier before you standardize. I'm not going to pretend a base-type-first method solves that—it doesn't. That's a compliance problem, not a fitment problem, and it needs its own check.
And one more: if your order volume is under, say, 50–100 bulbs a year, the "sourcing relationship" value isn't really there. You'll get more out of a good local parts store and a printed base-type cheat sheet. Scale matters, and being honest about that is better for everyone.
The real lesson wasn't "find a better bulb supplier." It was "stop treating year/make/model as a specification." That's the thing I didn't expect, and it's the thing that actually fixed the problem.