Test the exact sauce or spice blend you plan to sell, in the form you plan to sell it, before you print a single label — one sample, well mixed, measured for its free amino acid profile. That single result tells you the free glutamate behind the savoury taste, whether the batch matches the recipe in your notebook, and what wording your label can honestly carry. Everything else about selling — licensing, kitchen safety, shelf life — still has to be handled separately, and no composition test replaces it. This page covers the composition job, for the maker who is past "is it tasty?" and into "can I prove what it is?"
The job, in the maker's words
The recipe works. Friends reorder. A retailer or a market organiser asks what is in it, a co-packer asks for a specification, or you catch yourself writing "rich in umami" or a protein line on a label draft and realise you are guessing. Guessing is fine at the stove. On a product other people pay for, a guessed number is a liability that travels with every jar.
There is a second, quieter version of the same job. You scale the recipe — a bigger batch, a new supplier for one ingredient, a grinder that runs hotter — and the taste shifts slightly and you cannot tell whether the shift is real or imagined. Without a measured baseline for the batch that tasted right, you have nothing to compare the next batch against except memory. Memory is a poor specification.
Why blends and sauces are hard to pin down
A dry blend is not one ingredient; it is a mixture whose proportions vary scoop to scoop if mixing is uneven, and whose strongest components — salt, chilli, dark spices, dried mushroom or tomato, yeast-derived powders — dominate its character. A sauce adds water, oil, acid, and cooking, all of which change how ingredients present. In both cases the flavour you care about is carried substantially by free amino acids: amino acids already dissolved and available as eaten, among which glutamate is the best known for savoury depth.
That complexity matters analytically as well as culinarily. The matrix — everything in the sample besides the amino acids being measured — can distort an instrument reading, especially in a salty, dark, or oily product. A number produced without checking the matrix may describe the distortion as faithfully as it describes the blend. That is not a reason to avoid testing; it is a reason to test in the way described next.
The measurement that answers the job
Free amino acid analysis, panel AA-1 here, measures 19 amino acids plus ammonia by AQC (AccQ-Tag) pre-column derivatization with HPLC-UV: the amino acids are tagged, separated, and measured against a calibrated standard. Cysteine is excluded from the free panel and requires a separate AA-3 workup — relevant if a claim you have in mind turns on cysteine, unlikely for a sauce or rub, but the gap should be named rather than discovered later.
For a blend or sauce, commission the matrix check alongside the panel unless the lab advises otherwise for your specific product. Matrix-effect qualification adds a known spike of the analyte to your actual sample and measures how much is recovered — spike recovery. Good recovery means the matrix is leaving the reading largely alone. Poor recovery reveals the distortion, allowing a correction or, at minimum, an honestly qualified result. It is priced per sample and matrix on inquiry; a recent single-sample seasoning project with full matrix-effect qualification was $350. That figure is an example of one real project, not a flat rate for every sauce.
If your question is instead about the total amino acid makeup of the protein in the product, that is AA-T, total amino acid analysis, which hydrolyses the sample in 6 N hydrochloric acid at 110 °C for 24 hours before measurement, at $230 per sample. Most sauce and blend sellers asking about taste and glutamate need the free panel. The free-versus-total companion article will confirm which question is yours in five minutes; read it before ordering the more expensive test out of caution.
What it costs, and what to send
Single samples are welcome; you do not need a production run or a company. A liquid sauce is $150 per sample for AA-1. A dry blend, as a powder, is $180 per sample — $150 plus $30 powder preparation. First-time customers pay $150 per sample flat. If you settle into regular batch checking at 25 or more samples per month, the rate is $125 per sample. A $250 pilot covering 5 samples exists as a deposit, credited in full against a first contract signed within 90 days — a sensible shape if you are finalising several recipes at once and expect ongoing work.
Mix the batch thoroughly and sample representatively; a result describes the sample received, so an unmixed corner of the tub produces a faithful report about the wrong thing. Email before you post anything: describe the product, its main ingredients, and whether it is a liquid or a powder, and confirm the amount to set aside and the total price. The laboratory is in Davis, California, describes itself as validating now and reserving pilot slots, and is not GLP, GMP, or ISO/IEC 17025 accredited. Results are for your own decisions, including your label wording choices — not regulatory submission. If a retailer requires an accredited or specifically named test, obtain that requirement in writing first.
Your next step
Pick the batch that tastes right. Email info@openchemworks.com with the product, its ingredient list, and the claim or comparison you have in mind. Ask for the free panel, a view on matrix qualification for your product, and the all-in price. Keep the report with the recipe version it belongs to — a baseline is only useful if future batches can be compared against the same recipe record.
Frequently asked questions
Should I test before or after I finalise the recipe?
After. Testing a recipe you then change gives you a baseline for a product you no longer make. Finalise by taste, then measure.
My sauce is cooked. Does that matter?
Test it as sold, after cooking, because that is the product the customer eats and the claim describes. Tell the lab it is cooked and what is in it.
Can testing tell me why one batch tastes weaker?
A free-profile comparison between the baseline batch and the weaker batch can show whether free glutamate and related amino acids actually moved, or whether the difference lies elsewhere — salt, freshness, or perception. It narrows the search; it does not replace tasting.
Will this report satisfy my retailer?
Only if the retailer's written requirement matches a non-accredited composition report. Ask the retailer first; do not assume.
References
- OpenChemWorks — Pricing — Published service and pricing copy (2026).
- Waters — Derivatization of Amino Acids Using Waters AccQ-Tag Chemistry — Method background for AQC derivatization and HPLC-UV amino acid analysis.
OpenChemWorks Laboratory · Reviewed by the laboratory director, PhD chemist · Published October 16, 2026