Home › Terpenes › Cannabinoid Testing Methods
⚐ Flag an error How this page was made
Also known as: potency testing · cannabinoid quantification · cannabinoid profiling

Cannabinoid Testing Methods

An overview of how labs measure THC, CBD and other cannabinoids in cannabis.

Automated fact-check pass
7 cited sources
Published
Last reviewed
How this page was made
↯ The honest take

Lab testing for cannabinoids is useful and real, but the numbers on your jar are not as precise as they look. Results vary between labs by several percentage points, samples can be routed to labs known to report higher numbers, and sampling methods matter hugely. A COA tells you roughly what's in the product—not an exact measurement.

What cannabinoid testing measures

Cannabinoid testing quantifies the concentration of individual cannabinoids in a cannabis sample—most commonly Δ9-THC, THCA (the acidic precursor), CBD, CBDA, CBG, CBNA, CBN, and CBC. Results are typically reported as percent by weight (% w/w) for flower and concentrates, or milligrams per package or serving for edibles and tinctures [1].

A critical distinction: raw cannabis flower contains mostly acidic cannabinoids (THCA, CBDA), which are non-intoxicating until heated. Decarboxylation—driven by heat from smoking, vaping, or baking—converts THCA to Δ9-THC Strong evidence[2]. 'Total THC' is usually calculated as: Total THC = THC + (0.877 × THCA), where 0.877 accounts for the mass loss of CO₂ during decarboxylation Strong evidence[2]. Labels that report only 'THC' without distinguishing the acid form can be misleading because they obscure how much of the cannabinoid is active before heating.

HPLC: the industry standard

High-performance liquid chromatography with UV or diode-array detection (HPLC-UV/DAD) is the dominant method in cannabis labs Strong evidence[1][3]. The key advantage: it separates and quantifies cannabinoids without heating the sample, preserving the native ratio of acidic to neutral cannabinoids—essential for accurate flower analysis Strong evidence[1][3].

A typical workflow: grind sample → solvent extraction (often methanol or ethanol/water) → filter → inject onto a reversed-phase C18 column → detect ultraviolet absorbance at ~220–230 nm Strong evidence[1]. Well-validated HPLC methods can resolve 10 or more cannabinoids in under 20 minutes Strong evidence[3].

Limitations: HPLC-UV cannot easily distinguish Δ8-THC from Δ9-THC without careful method development, and it cannot identify unknown compounds. For those tasks, labs use HPLC coupled to mass spectrometry (HPLC-MS), which adds molecular-weight confirmation Strong evidence[4]. HPLC-MS is more expensive and less routine but increasingly common in regulated markets.

Gas chromatography and alternatives

Gas chromatography with flame ionization detection (GC-FID) or mass spectrometry (GC-MS) was historically used for cannabis analysis but has a major drawback: the injection port heat (~250°C) decarboxylates THCA to THC during analysis Strong evidence[2][3]. This means GC measures total THC by default and destroys information about the acid/neutral ratio unless the sample is chemically derivatized beforehand Strong evidence[2][3].

GC remains valuable for terpene profiling—terpenes are volatile and well-suited to GC analysis—and for forensic work. Supercritical fluid chromatography (SFC) and near-infrared spectroscopy (NIR) are emerging alternatives; NIR is fast and non-destructive but less accurate, and is mostly used for in-process screening rather than compliance testing Weak / limited[1].

Why lab results vary

Multiple peer-reviewed studies and investigative reports have documented systematic problems with commercial cannabinoid testing Strong evidence[5][6]:

Lab-to-lab variability: Round-robin studies splitting identical samples across laboratories have found THC results varying by several percentage points in absolute terms Strong evidence[5]. A 2023 analysis of Colorado labs found significant inconsistencies in reported potency across the state Strong evidence[6].

Lab shopping and potency inflation: Producers can route samples to labs known to report higher THC numbers, which command higher wholesale prices. Investigative reporting in multiple states has documented this practice Strong evidence[7].

Sampling heterogeneity: Cannabinoid content varies between buds on the same plant and even within a single bud. If only a small aliquot is tested, the result may not represent the entire batch Strong evidence[1][5].

Reference standard quality: Accurate quantification requires certified reference materials. Not all labs use them consistently Strong evidence[1].

ISO/IEC 17025 accreditation (international standard for laboratory competence) helps but does not eliminate these issues Strong evidence[5][6]. Consumers should interpret label potencies as approximations rather than exact measurements.

What a good Certificate of Analysis includes

A defensible Certificate of Analysis (COA) for cannabinoids should include Strong evidence[3][4]:

For consumers and patients, the practical takeaway: COAs are useful for confirming a product contains roughly what it claims, for verifying CBD:THC ratios, and for screening contaminants (when included) Strong evidence[5][6]. They are not precision measurements. A flower labeled '24% THC' might genuinely test anywhere from ~20% to ~28% if re-analyzed at a different accredited lab Strong evidence[5][6].

How to read a COA

When reviewing a Certificate of Analysis Strong evidence[3]:

  1. Check for method transparency: Reputable labs list their chromatography technique. HPLC-DAD or HPLC-MS are standard for flower; GC-MS is acceptable for concentrates where decarboxylation is less of a concern.
  2. Look for THCA and THC separately: A COA that reports only 'total THC' or fails to distinguish acidic from neutral forms is less informative.
  3. Verify accreditation: ISO/IEC 17025 accreditation is a good sign, though absence of it does not guarantee poor work.
  4. Note the date: Cannabinoids degrade over time, especially in warm or humid storage. A COA from 6+ months before purchase may not reflect the current product.
  5. Cross-reference if possible: If you've purchased the same product before, compare COAs for consistency. Large swings between batches can indicate sampling or lab variability.

Sources

  1. Peer-reviewed Citti, C., Braghiroli, D., Vandelli, M. A., & Cannazza, G. (2018). Pharmaceutical and biomedical analysis of cannabinoids: A critical review. Journal of Pharmaceutical and Biomedical Analysis, 147, 565–579.
  2. Peer-reviewed Dussy, F. E., Hamberg, C., Luginbühl, M., Schwerzmann, T., & Briellmann, T. A. (2005). Isolation of Δ9-THCA-A from hemp and analytical aspects concerning the determination of Δ9-THC in cannabis products. Forensic Science International, 149(1), 3–10.
  3. Peer-reviewed Giese, M. W., Lewis, M. A., Giese, L., & Smith, K. M. (2015). Development and validation of a reliable and robust method for the analysis of cannabinoids and terpenes in cannabis. Journal of AOAC International, 98(6), 1503–1522.
  4. Peer-reviewed Aizpurua-Olaizola, O., Omar, J., Navarro, P., Olivares, M., Etxebarria, N., & Usobiaga, A. (2014). Identification and quantification of cannabinoids in Cannabis sativa L. plants by HPLC-MS/MS. Analytical and Bioanalytical Chemistry, 406(29), 7549–7560.
  5. Peer-reviewed Jikomes, N., & Zoorob, M. (2018). The cannabinoid content of legal cannabis in Washington State varies systematically across testing facilities and popular consumer products. Scientific Reports, 8, 4519.
  6. Peer-reviewed Schwabe, A. L., Johnson, V., Harrelson, J., & McGlaughlin, M. E. (2023). Uncomfortably high: Testing reveals inflated THC potency on retail Cannabis labels. PLOS ONE, 18(4), e0282396.
  7. Reported Schroyer, J. (2022). Marijuana potency inflation: How lab shopping drives inflated THC numbers. Marijuana Business Daily.

How this page was made

Generation history

Oct 30, 2026
Fact-check pass — raised 8 flags
Oct 30, 2026
Revision
Jun 3, 2026
Initial draft
Jun 2, 2026
Fact-check pass — raised 2 flags

Drafted with AI assistance and checked by a separate automated fact-check pass. Not every article has been read by a person. See how articles are made.