A buyer sent me two vials last March with the same part number on the label. One was an OEM part from a Tier 1 supplier; the other was a private-label version sourced through a distributor. Same 12 x 32 mm dimensions, same 9 mm screw thread, same GPI 9-425 finish, same septa listed as silicone/PTFE. The difference showed up on the third injection: the private-label lot produced a silanol tailing factor of 1.94 on a moderately basic analyte, against 1.21 for the OEM lot, on the same instrument, same column, same method. The factory was identical. The tooling was not. That gap is the whole OEM vs private label analytical testing difference in one measurement, and it is the thing most procurement reviews never get to see because the paperwork looks the same on both sides.

When people ask me what actually differs between OEM and private label in analytical testing, I usually start by asking two questions: who owns the mold, and who owns the recipe. Everything else (certificates, branding, packaging, lead times) follows from those two answers. If you cannot answer them for a given part, you do not know what you are buying, regardless of what the spec sheet says.

Tooling ownership decides what the supplier can and cannot change without telling you

In an OEM arrangement, the brand that sells the part either owns the tooling outright or has an exclusive-use agreement that locks the mold to that brand. That sounds like a legal nicety. It is not. It is the mechanism that stops a supplier from quietly re-cutting a cavity, changing a gate location, or moving a family mold to a different press when capacity gets tight.

I have watched a septa supplier move a 20-cavity mold from a 110-ton press to a 150-ton press to free up capacity. Same mold, same compound, same cure schedule on paper. The clamp force change altered the flash profile on the outer 1.5 mm of the septa edge. On a 9 mm crimp cap that flash is invisible. On an autosampler needle piercing the center, it changed the puncture force from a mean of 4.8 N to 6.1 N with a much wider distribution, and the lab started seeing coring on about one in 400 injections. The OEM customer got a deviation notice and a lot hold. The private-label customer got a shipment.

That is the tooling question in practice. Who owns the mold determines who has to be told when the mold changes hands, changes presses, or gets re-cut.

What the private-label buyer typically does not own

In a private-label deal, the factory owns the tooling. The brand on the box is a label, not a manufacturing commitment. That means the factory can:

• Run the same mold for multiple brands on the same shift, with the same compound and the same cure, and only change the bag and the carton.
• Run a different mold for the same part number when the primary mold is down, as long as the drawing dimensions are met.
• Change resin lot, mold release agent, or annealing cycle without notifying the label holder, because the label holder is not the design authority.

None of those are automatically bad. A well-run factory with a solid QMS will control all of them. But the control is internal, and the private-label buyer has no contractual hook to see it. That is the structural difference, and it is the one that shows up as drift two years into a supply relationship.

The recipe is the second ownership question, and it is the one that bites in regulated work

The recipe is everything that goes into the part besides the drawing: the polymer grade and lot, the mold release, the annealing profile, the post-mold washing step, the packaging atmosphere, the sterilization cycle if any. In an OEM relationship the brand usually specifies the recipe and audits against it. In a private-label relationship the factory owns the recipe and the label holder buys against a performance spec, if there is one.

For analytical testing this matters because the recipe is what determines extractables. A polypropylene vial that meets the dimensional drawing can still leach a mold release that shows up as a ghost peak at 12.4 minutes on a 30 m x 0.25 mm x 0.25 µm DB-5 column, and the dimensional drawing will never tell you. I have written about ghost peaks as a material problem before, and the private-label trap is that the material is only specified at the family level (“polypropylene, virgin, USP Class VI”), not at the compound level.

OEM contracts for regulated work usually specify the compound by supplier and grade, and lock the mold release to a non-silicone, non-stearate type. Private-label contracts usually do not. If your method is validated under EPA SW-846 or a USP <661> extractables protocol, that difference is a revalidation trigger, not a purchasing detail.

Where the recipe shows up on the bench

Three places, in my experience, and all three are measurable:

• Extractables by GC-MS or LC-MS after a defined soak (typically 24 h at 40 °C in the mobile phase or a surrogate). OEM lots usually show a tighter profile lot to lot.
• Pierce force and coring on the septum, measured with a force gauge and a defined needle (I use a 22-gauge, 0.72 mm OD, at 200 mm/min). OEM septa usually hold a mean within 0.4 N of nominal; private-label lots can drift 1.5 N or more across six months.
• Dimensional stack on the vial neck and cap thread. A GPI 9-425 finish has a defined thread form; a private-label lot that meets it on paper can still fail the capping torque test because the thread flank angle is off by a degree or two.

If you cannot measure those three, you are buying a private-label part on faith. That is a defensible position for a general-purpose sample cup. It is not a defensible position for a validated method.

What changes when the brand changes but the factory does not

This is the part that confuses people, because the factory genuinely does not change. The press does not move. The operator does not change. The mold cavity is the same cavity. So why does the part perform differently?

Three reasons, and they are all about control rather than capability.

1. The specification moves from recipe to dimensions

An OEM spec typically says “compound X from supplier Y, mold release Z, annealed at 120 °C for 2 h, washed in DI, packed in a nitrogen-flushed bag.” A private-label spec typically says “12 x 32 mm, 9 mm screw thread, 1.5 mL, PP, septum silicone/PTFE, 100 per pack.” The factory can meet the second spec with a dozen different recipes. Most of the time it uses the same one it uses for the OEM. Some of the time it does not, and you find out on the bench.

2. The change-control obligation shifts

In an OEM deal, the change-control clause usually requires written notice before any change to compound, mold, press, or process. In a private-label deal, the factory’s change-control obligation is to its own QMS, not to you. There is no notice requirement because there is no design authority relationship. That is the whole point of private label from the factory’s side: it can optimize without a customer conversation.

3. The certificate of analysis changes meaning

An OEM CoA usually traces to a lot-specific measurement against a locked spec. A private-label CoA often traces to a lot-specific measurement against a generic spec, or worse, to a certificate of conformance that says “meets specification” without a number. I have seen both. The one that says “meets specification” without a number is not a certificate, it is a promise, and I have written about that distinction elsewhere.

None of this means private label is bad. It means the buyer has to do the work the OEM brand would otherwise do. For a full picture of how these parts sit in a regulated workflow, see analytical testing coverage on the site.

A comparison that holds up in a procurement review

Here is the table I use when a lab manager asks me to score an OEM quote against a private-label quote. The numbers are typical values from my own supplier audits over the last six years, not industry averages, so treat them as calibration points rather than guarantees.

The table is not a verdict. It is a list of the things you need to verify before you can treat a private-label part as interchangeable with an OEM part. If you can verify them, private label can be a perfectly sound choice. If you cannot, you are buying on the assumption that the factory did not change anything, and that assumption has a shelf life.

Two field cases where the difference was measured, not argued

Both of these are from my own work, and both turned on a number rather than an opinion.

Case 1: The tailing factor that moved with the label

Symptom: A contract lab running a USP method for a small-molecule API saw tailing factors on a basic analyte drift from 1.24 to 1.87 over three lots, with no change to the column, mobile phase, or instrument. The lab had switched from OEM vials to a private-label equivalent six weeks earlier to save on unit cost.

Measurement: We ran a controlled comparison. Twenty injections from each lot on the same column, same method, same autosampler. The OEM lot gave a mean tailing factor of 1.21 (RSD 2.8%). The private-label lot gave 1.84 (RSD 6.1%). We then ran a 24 h soak of both lots in mobile phase and analyzed by LC-MS. The private-label extract showed a siloxane series at m/z 281, 355, 429 that was absent from the OEM extract. The septa were listed as the same silicone/PTFE laminate, but the private-label lot had a different silicone cure system, likely a different platinum catalyst loading, and it was leaching.

Fix: The lab moved the validated method back to the OEM vial and septum, and wrote a functional spec into the private-label contract for the non-regulated screening work: extractables by soak test below a defined threshold, and tailing factor within 1.3 on a reference analyte. The private label stayed in the building, but only on the methods where a 0.6 tailing shift would not sink a system suitability check.

Case 2: The cap torque that read fine on the drawing

Symptom: An environmental lab running volatiles by headspace started seeing intermittent leaks on a 20 mL headspace vial, about one in 250 vials, with a crimp cap. The leak showed up as a 30% drop in response for a 2 µg/L standard, which is the kind of number that gets blamed on the instrument first.

Measurement: We pulled 200 vials from the suspect lot and 200 from the OEM lot and measured capping torque with a torque tester set to a 1.5 N·m target, plus seal integrity by pressure decay at 1.0 bar. The OEM lot held 1.48 to 1.55 N·m with zero pressure-decay failures. The private-label lot held 1.30 to 1.62 N·m with four failures. The neck finish on the private-label lot was within the GPI 9-425 drawing on diameter but the thread root radius was undersized by about 0.08 mm, which let the cap seat at a lower torque on some vials and over-compress on others.

Fix: The lab moved headspace work to the OEM lot and added a torque check to incoming inspection: 10 vials per lot, target 1.5 N·m ± 0.1, reject the lot if any vial falls outside 1.35 to 1.65 or fails pressure decay. That is a 15-minute check that has caught two bad lots since.

Both cases share a shape. The private-label part met the drawing. It did not meet the function, because the drawing was never the whole spec. For the vial side of that argument, HPLC vial resources are a useful reference on the dimensional stack and the neck finish.

What to write into a private-label contract so it behaves like an OEM one

You can close most of the gap with a contract, and it costs less than the OEM premium on a validated line. Here is the checklist I use, in the order I negotiate it.

Control clauses

• Tooling disclosure: which mold, which press, which cavity count. Named, not generic.
• Recipe lock: compound by supplier and grade, mold release type, annealing profile, wash step, packaging atmosphere. All named.
• Change notification: 60 days written notice before any change to the above, with a right to re-qualify or reject.
• CoA content: lot-specific measured values for pierce force, dimensions, extractables, and torque where applicable. Not conformance-only.
• Retention samples: 12 months minimum, available on request.

Functional specs with numbers

• Pierce force (septa): mean within 0.4 N of nominal, RSD ≤ 8% across 30 samples.
• Coring: zero coring events in 500 consecutive punctures with a 22-gauge needle at the method’s injection rate.
• Torque (screw caps): 1.5 N·m ± 0.1, no vial outside 1.35 to 1.65.
• Extractables: soak 24 h at 40 °C in mobile phase, no peak above the method’s LOQ that is absent from the OEM reference.
• Dimensional stack: GPI 9-425 finish per drawing, plus a functional fit test on the lab’s own autosampler.

That list is not exhaustive, but it is the one that has caught the most problems for me. If a factory will not sign it, that is your answer on the OEM vs private label question for that part. For a broader view of how consumables fit into the method, chromatography consumables coverage on the network is a good starting point, and the Agilent consumables reference and Thermo Fisher sample handling pages give you the OEM baseline for comparison. Restek technical resources are also useful for septa and inlet liner selection criteria.

Troubleshooting checklist when a private-label part starts to drift

When a lab calls me because a method has gone sideways after a supplier switch, I run this in order. It takes about two hours and it separates a part problem from an instrument problem before anyone touches the instrument.

1. Pull retention samples from both lots. If you do not have them, you have already lost the comparison. Start retaining from now on.
2. Run 20 injections from each lot on the same column and method. Compare tailing factor, response, and retention time RSD. A shift in tailing with stable retention points to a part; a shift in retention with stable tailing points to the column or mobile phase.
3. Measure pierce force on 30 septa per lot. Force gauge, 22-gauge needle, 200 mm/min. Anything above 0.4 N spread between lots is a flag.
4. Run a 24 h soak extractables screen. Compare against the OEM reference. New peaks are the answer.
5. Check capping torque on 10 vials per lot. 1.5 N·m target, reject outside 1.35 to 1.65.
6. Check the dimensional stack on the neck finish. Thread root radius, thread pitch, and seal seat depth. A 0.08 mm deviation is enough to cause intermittent leaks.
7. Only then look at the instrument. If steps 1 to 6 are clean, the problem is upstream or downstream of the vial, and you can stop blaming the part.

That checklist has saved me from replacing an autosampler syringe more than once. The syringe was fine. The septa were not.

The decision rule

Buy OEM when the method is validated and the part sits inside the validated envelope, because the tooling and recipe control are what keep the method inside that envelope. Buy private label when the method is general-purpose, the part is not in the critical path, and you have either signed the control clauses above or you have the incoming inspection to catch drift yourself. The factory is the same factory. What changes is who owns the mold, who owns the recipe, and who has to be told when either one moves. If you cannot name all three, you are not buying a part, you are buying a hope. For the broader supply picture, the lab vials resources on the network cover the dimensional and material side in more detail.

References & further reading

Related resources from our lab consumables network:

• Analytical Testing at testinganalytical.com
• Hplc Vial at vialforhplc.com
• Chromatography Consumables at chromatographyconsumables.com
• Lab Vials at labvials.com

External references and standards:

• Agilent chromatography consumables and columns
• Thermo Fisher chromatography and sample handling
• U.S. EPA SW-846 test methods for waste and materials
• Restek technical resources for chromatography