- Aug 15, 2018
- 602
- 346
- 63
Alright everyone, OLYMPIC here.
We talk about lab reports often enough that it's worth having one reference thread on how to actually **read** a Certificate of Analysis (COA). This is not another argument about whether a particular category of compounds is genuine. We already have that discussion elsewhere. This is about understanding what the paperwork says, what it doesn't say, and how to spot a report that deserves more questions.
A COA is a record of particular tests performed on a submitted sample. It is not a universal guarantee, and the words “lab tested” tell you very little without the supporting details. Two documents can look equally professional while answering completely different analytical questions.
## First, check what the report is attached to
Before looking at a percentage, look for the boring identifying information:
- Laboratory name and report identifier
- Sample name or description
- Unique sample or batch/lot number
- Date received, date tested, and date issued
- Analytical methods and results with units
- Report version, sign-off, and relevant limitations
Now compare the lot or batch identifier on the document with the one printed on the container or vial. **An exact match is necessary for a claimed connection, but it isn't proof that the vial supplied the sample.** Sometimes a laboratory assigns its own sample ID while the submitter uses a separate batch ID; a proper traceability record should link those identifiers.
No matching identifiers? Don't assume the report covers that container. A date by itself is not a batch reference. And a report for one submitted sample cannot automatically establish the composition of everything ever made under a similar label.
## The panels: what are they actually measuring?
### 1. Identity and mass spectrometry
An identity panel asks whether the detected material is consistent with the proposed substance. Mass spectrometry measures ions according to mass-to-charge characteristics. Depending on the method, it can support identity through molecular mass, fragmentation patterns, or other reference comparisons.
But “mass spec passed” isn't a complete method description. Was it a targeted method? Was a reference standard used? Were related molecules distinguishable? A reported matching mass does not, alone, identify every component or demonstrate complete purity.
### 2. HPLC purity percentage
HPLC means high-performance liquid chromatography. It separates components under defined conditions and produces a chromatogram showing detected signals. A report might state **98.9% purity**, but ask how that number was calculated.
One common result is chromatographic **area percent**: the target peak represents a fraction of the total integrated signal detected under that method. That is not automatically the same as 98.9% of the sample's mass being the target substance. Response factors, undetected materials, water, salts, and method scope all matter.
An informative result describes the method, detector, calculation basis, and any relevant limitations. A beautiful single peak is not a substitute for that context.
### 3. Related substances and degradation products
This panel looks for specified impurities or breakdown products. It can be separate from the headline HPLC purity result, or it may be evaluated with a related analytical procedure. Check whether individual impurities are named and whether the report includes detection or quantitation limits.
“Not detected” means *not detected under the stated method and limit*. It never means every imaginable impurity is absent. A blank impurity table without a method or reporting threshold tells you less than it appears to.
### 4. Residual solvents
Residual-solvent testing concerns volatile substances potentially remaining from processing. Reports may list individual solvent names, numerical findings, reporting limits, and analytical methods. A general purity percentage does not establish whether residual solvents were tested.
You may see values in parts per million or other units. Without a named analyte, method, and stated specification, “solvent tested” is vague. Equally, an analytical guideline designed for regulated pharmaceuticals should not be mistaken for proof that an unapproved research compound has regulatory clearance.
### 5. Endotoxins and microbiological testing
These are **different questions**. Endotoxin testing targets bacterial endotoxin components associated with certain bacteria. A microbial test assesses specified forms of viable contamination under its method. A sample may be negative under one category of test without having been assessed under the other.
Check the exact panel name and the reported limit. The word “sterile” cannot be inferred from an HPLC chromatogram, a mass spectrum, or an endotoxin result alone. No single panel establishes universal safety or suitability for human use.
## Third-party versus in-house: what changes?
**In-house testing** means the submitting organization or an associated operation performed the analysis. **Third-party testing** means a separate laboratory performed it. Independence may improve confidence against certain conflicts of interest, but neither label guarantees good sampling, suitable methods, or honest reporting.
For a third-party report, look for a named testing facility, unique report number, authorized sign-off, and a way to authenticate the original document. If a facility claims accreditation, check whether the **specific test** is within its assessed scope. Accreditation is not a blanket endorsement of every possible assay.
Also ask who selected and submitted the sample. An independent laboratory can honestly analyze what it receives without independently knowing where that sample came from. That distinction is often missed in forum debates.
## Red flags worth stopping at
- No batch number, or the report's batch doesn't match the container
- Missing testing or issue dates
- No laboratory identity or unique report number
- “HPLC tested” without a method or calculation basis
- A headline purity number presented as if it covers solvents, microbes, and endotoxins
- An old report presented for a different or later batch
- Cropped screenshots that hide qualifications, signatures, or page numbers
- A claimed third-party result with no way to check its authenticity
- “Not detected” without a stated detection or reporting limit
An old COA isn't automatically invalid merely because time has passed; its relevance depends on the batch, stability, and claims being made. But **an expired, superseded, or obviously unrelated report** should never be passed off as current evidence for a different sample.
## My five-minute reading order
Read the **sample and batch identity first**, then the **date**, then the **specific panel and method**, then the **result and units**, and finally the **limitations and authorization**. If something doesn't line up, mark it as unanswered instead of filling the gap with assumptions.
I want us to be a community that can discuss evidence without treating a screenshot as gospel or assuming everyone with questions is attacking someone. Technical literacy means knowing what a good report can establish *and* where it stops. None of this is medical guidance, an endorsement of any compound, or an instruction for using one.
## Community discussion
1. What missing field on a COA makes you stop and ask questions first: batch number, method, dates, or something else?
2. Have you seen a report where “purity” was confused with identity, microbial testing, or residual-solvent testing?
3. What would make independent report verification clearer for forum members without exposing anyone's private information?
— OLYMPIC
We talk about lab reports often enough that it's worth having one reference thread on how to actually **read** a Certificate of Analysis (COA). This is not another argument about whether a particular category of compounds is genuine. We already have that discussion elsewhere. This is about understanding what the paperwork says, what it doesn't say, and how to spot a report that deserves more questions.
A COA is a record of particular tests performed on a submitted sample. It is not a universal guarantee, and the words “lab tested” tell you very little without the supporting details. Two documents can look equally professional while answering completely different analytical questions.
## First, check what the report is attached to
Before looking at a percentage, look for the boring identifying information:
- Laboratory name and report identifier
- Sample name or description
- Unique sample or batch/lot number
- Date received, date tested, and date issued
- Analytical methods and results with units
- Report version, sign-off, and relevant limitations
Now compare the lot or batch identifier on the document with the one printed on the container or vial. **An exact match is necessary for a claimed connection, but it isn't proof that the vial supplied the sample.** Sometimes a laboratory assigns its own sample ID while the submitter uses a separate batch ID; a proper traceability record should link those identifiers.
No matching identifiers? Don't assume the report covers that container. A date by itself is not a batch reference. And a report for one submitted sample cannot automatically establish the composition of everything ever made under a similar label.
## The panels: what are they actually measuring?
### 1. Identity and mass spectrometry
An identity panel asks whether the detected material is consistent with the proposed substance. Mass spectrometry measures ions according to mass-to-charge characteristics. Depending on the method, it can support identity through molecular mass, fragmentation patterns, or other reference comparisons.
But “mass spec passed” isn't a complete method description. Was it a targeted method? Was a reference standard used? Were related molecules distinguishable? A reported matching mass does not, alone, identify every component or demonstrate complete purity.
### 2. HPLC purity percentage
HPLC means high-performance liquid chromatography. It separates components under defined conditions and produces a chromatogram showing detected signals. A report might state **98.9% purity**, but ask how that number was calculated.
One common result is chromatographic **area percent**: the target peak represents a fraction of the total integrated signal detected under that method. That is not automatically the same as 98.9% of the sample's mass being the target substance. Response factors, undetected materials, water, salts, and method scope all matter.
An informative result describes the method, detector, calculation basis, and any relevant limitations. A beautiful single peak is not a substitute for that context.
### 3. Related substances and degradation products
This panel looks for specified impurities or breakdown products. It can be separate from the headline HPLC purity result, or it may be evaluated with a related analytical procedure. Check whether individual impurities are named and whether the report includes detection or quantitation limits.
“Not detected” means *not detected under the stated method and limit*. It never means every imaginable impurity is absent. A blank impurity table without a method or reporting threshold tells you less than it appears to.
### 4. Residual solvents
Residual-solvent testing concerns volatile substances potentially remaining from processing. Reports may list individual solvent names, numerical findings, reporting limits, and analytical methods. A general purity percentage does not establish whether residual solvents were tested.
You may see values in parts per million or other units. Without a named analyte, method, and stated specification, “solvent tested” is vague. Equally, an analytical guideline designed for regulated pharmaceuticals should not be mistaken for proof that an unapproved research compound has regulatory clearance.
### 5. Endotoxins and microbiological testing
These are **different questions**. Endotoxin testing targets bacterial endotoxin components associated with certain bacteria. A microbial test assesses specified forms of viable contamination under its method. A sample may be negative under one category of test without having been assessed under the other.
Check the exact panel name and the reported limit. The word “sterile” cannot be inferred from an HPLC chromatogram, a mass spectrum, or an endotoxin result alone. No single panel establishes universal safety or suitability for human use.
## Third-party versus in-house: what changes?
**In-house testing** means the submitting organization or an associated operation performed the analysis. **Third-party testing** means a separate laboratory performed it. Independence may improve confidence against certain conflicts of interest, but neither label guarantees good sampling, suitable methods, or honest reporting.
For a third-party report, look for a named testing facility, unique report number, authorized sign-off, and a way to authenticate the original document. If a facility claims accreditation, check whether the **specific test** is within its assessed scope. Accreditation is not a blanket endorsement of every possible assay.
Also ask who selected and submitted the sample. An independent laboratory can honestly analyze what it receives without independently knowing where that sample came from. That distinction is often missed in forum debates.
## Red flags worth stopping at
- No batch number, or the report's batch doesn't match the container
- Missing testing or issue dates
- No laboratory identity or unique report number
- “HPLC tested” without a method or calculation basis
- A headline purity number presented as if it covers solvents, microbes, and endotoxins
- An old report presented for a different or later batch
- Cropped screenshots that hide qualifications, signatures, or page numbers
- A claimed third-party result with no way to check its authenticity
- “Not detected” without a stated detection or reporting limit
An old COA isn't automatically invalid merely because time has passed; its relevance depends on the batch, stability, and claims being made. But **an expired, superseded, or obviously unrelated report** should never be passed off as current evidence for a different sample.
## My five-minute reading order
Read the **sample and batch identity first**, then the **date**, then the **specific panel and method**, then the **result and units**, and finally the **limitations and authorization**. If something doesn't line up, mark it as unanswered instead of filling the gap with assumptions.
I want us to be a community that can discuss evidence without treating a screenshot as gospel or assuming everyone with questions is attacking someone. Technical literacy means knowing what a good report can establish *and* where it stops. None of this is medical guidance, an endorsement of any compound, or an instruction for using one.
## Community discussion
1. What missing field on a COA makes you stop and ask questions first: batch number, method, dates, or something else?
2. Have you seen a report where “purity” was confused with identity, microbial testing, or residual-solvent testing?
3. What would make independent report verification clearer for forum members without exposing anyone's private information?
— OLYMPIC