Kjeldahl and Dumas protein tests measure total nitrogen, not protein — so cheap nitrogen-rich amino acids like glycine can inflate the number. A free amino acid panel (nineteen amino acids plus ammonia, by HPLC) separates each one individually, exposing the spiking signature the nitrogen test can’t see.
The COA says 85.2% protein.
You sign off. You pay whey prices.
But somewhere in the back of your mind, a question you can’t unhear: is this number real?
You’ve heard the term. Amino spiking. Cheap nitrogen-rich amino acids dumped into the batch to push the protein number up — without adding a gram of actual protein.
Glycine. Creatine. Urea. Pennies per kilo.
And the lab test your supplier uses can’t tell the difference.
You’re the QA manager. The number is on your desk. And you have no way to verify it.
The test everyone uses is blind by design
Here’s the problem nobody puts on the COA.
Kjeldahl — or its faster cousin, Dumas — doesn’t measure protein. It measures total nitrogen, then multiplies by 6.25.
Nitrogen in, protein number out.
The math assumes all the nitrogen comes from protein, which is roughly 16% nitrogen. As long as that’s true, the number is honest.
The moment someone adds free glycine, the math breaks. Glycine is 18.6% nitrogen — richer than protein itself. Every gram of it counts for more than a gram of real protein on a Kjeldahl test.
The test was never built to catch this. It’s a surrogate — a stand-in for protein, not a measurement of it. It cannot distinguish protein nitrogen from free amino acids, creatine, taurine, or urea. A free amino acid panel takes the opposite approach — see what an amino acid panel tells you.
The economics favor the spiker
Glycine costs a fraction of whey per kilo — pennies against dollars.
A 5% glycine addition can lift the apparent protein number by three to four percentage points. The COA still reads 85.2%. The batch cost drops. The margin moves.
You can’t see it. Your supplier’s routine lab can’t see it. The number looks perfect — because the test is measuring exactly what the spiker wants it to measure: nitrogen.
And the number has your name under it.
If the product gets flagged — a retailer audit, a competitor’s test, a customer with a lab — the COA you signed is the paper trail. The question won’t be whether you ran a test. It’ll be whether you ran the right one.
If you have a COA you’re not sure about, that’s the exact question our pilot answers.
The one measurement that sees it
A free amino acid panel doesn’t give you one nitrogen number. It separates and measures each amino acid individually — all nineteen in our panel, plus ammonia — by HPLC.
A clean ingredient has a profile that looks like itself. A spiked one doesn’t. The glycine peak towers where it shouldn’t — far above what’s natural for the ingredient.
One number becomes a full profile, and the lie has nowhere to hide.
That’s what an amino acid spiking test is: not a new kind of protein test, but the measurement that makes the old one’s blindness visible.
Why you can’t just buy this test today
Here’s the strange part. Everyone in the industry knows about spiking. Brand blogs explain it. Forums argue about it endlessly.
But try to buy a test called an amino acid spiking test, and you’ll find almost nothing. Barely any lab sells it by that name.
Vendors sell you the instrument. Certifiers sell you a badge. Nobody sells the answer to your question.
That’s the gap.
What it shows, concretely
Take the glycine example. Five kilos of glycine in a hundred-kilo batch costs a few dollars and buys three to four points of “protein” on paper.
On a panel, that same batch shows its glycine fraction sitting far above what’s natural for the ingredient — the signature of spiking, not of protein.
No single nitrogen number can draw that distinction. Only the individual profile can.
So the next time a COA lands on your desk reading 85.2%, you’ll know the right question. Not “is the number high enough” — but “what did they actually measure?”
If you’re wondering what your label is actually worth: $250 for 5 samples — a deposit, not a discount: credited in full against your first contract if you sign within 90 days of the report. We’re validating now and reserving pilot slots. See our per-sample prices, or start by verifying amino acids on a supplement label.
References
- Nutritional Outlook — “Protein products vulnerable to amino acid spiking” — Kjeldahl described as a “surrogate test”; industry discussion of amino acid spiking.
- Molecules 2020 — nitrogen content and protein conversion factors — Glycine at 18.6% nitrogen vs ~16% assumed for protein; nitrogen-rich adulterants lower the true conversion factor.
- Ayah Labs — nitrogen spiking in protein raw materials — Worked example of glycine addition lifting apparent protein on Kjeldahl; one lab’s figures, not universal across matrices.
OpenChemWorks Laboratory · Reviewed by the laboratory director, PhD chemist · Published October 2, 2026