A Chromatogram Is Evidence, Not a Complete COA
A chromatogram can be useful analytical evidence. By itself, however, it does not establish which sample was tested, how the result was calculated, or whether the reported conclusion is defensible.
Researchers comparing certificates of analysis often encounter a polished report with a chromatogram image attached. The graph may look technical and complete, but it represents only one layer of the analytical record. A defensible result depends on the chain connecting the physical sample to the method, raw data, calculations, review, and final report.
What a chromatogram actually shows
In chromatography, a detector records a signal over time. The resulting plot can show peaks, their retention times, and their measured responses. With an appropriate validated or otherwise fit-for-purpose method, reference materials, calculations, and review, those observations may contribute to an identity, purity, or assay determination.
The image alone does not explain what was injected, how the sample was prepared, whether the system performed acceptably, which peaks were included or excluded, or how the final number was derived. It is an output—not the whole analytical story.
Six record layers that connect a sample to a result
| Record layer | Question it answers | What to look for |
|---|---|---|
| Sample identity | What material was actually tested? | Product or compound name, lot or batch number, sample source, receipt or test date, and unambiguous sample identifier. |
| Method | How was the sample separated and measured? | Method identifier and version, relevant conditions, preparation steps, detector, and a statement that the method is suitable for its intended question. |
| System performance | Did the analytical system work acceptably during the run? | System-suitability criteria and results, calibration status where relevant, blanks, controls, and sequence context. |
| Raw data and processing | How did instrument output become integrated peaks? | Original data files, processing method, integration parameters, reintegrations or exclusions, and attributable change history. |
| Calculation | How was the reportable value obtained? | Formula, units, dilution factors, reference-standard values, conversion factors, and any acceptance criteria. |
| Review and report | Who checked the work and what conclusion was authorized? | Analyst and reviewer identity, dates, approved result, specification comparison, and a COA tied to the exact lot. |
Why an image can be misleading without context
Peak area is not automatically purity
A percentage printed beside a peak may describe area normalization under one set of conditions. It does not automatically establish mass fraction, identity, potency, or absence of every possible impurity. Compounds can have different detector responses, and some substances may not be visible under the selected method or wavelength.
Retention time is not identity by itself
A peak appearing near an expected retention time can support an identification, but retention time alone is generally not uniquely specific. A method may require comparison with an authenticated reference standard or an orthogonal technique, depending on the analytical question.
A clean graph does not prove complete data
A static screenshot cannot reveal the full injection sequence, deleted runs, manual integration, processing changes, audit-trail entries, or whether the displayed file belongs to the named lot. Those connections must be established by controlled records.
Chromatogram, raw data, reportable result, and COA are different things
| Term | Working definition |
|---|---|
| Chromatogram | A detector-response plot produced during a chromatographic analysis. |
| Raw data | The original records needed to reconstruct and evaluate the analysis, including instrument files and associated metadata. |
| Reportable result | The authorized value or conclusion produced after defined processing, calculation, review, and application of acceptance criteria. |
| Certificate of analysis | A summary document that reports tests, specifications, and results for a defined material or lot. Its reliability depends on the supporting records and exact lot linkage. |
A procurement checklist for chromatogram-backed COAs
When comparing suppliers advertising research peptides for sale, use the same questions for every candidate:
- Does the COA identify the exact product and lot shown on the received container?
- Is the testing laboratory identified, with a report number and test date?
- Are the test method and measured attribute stated clearly?
- Are units, specifications, and actual results distinguished from one another?
- Can the supplier explain whether the chromatogram is part of the original analytical package or only an exported image?
- Are reference standards, calculations, system suitability, and review records available when the research program requires deeper qualification?
- Does the document avoid making broader claims than the method can support?
How this fits with method suitability
A complete-looking record cannot rescue an unsuitable method. Before interpreting any result, define the measurand—the exact property the study needs to know—and confirm that the method is selective and capable enough for that purpose. Our earlier guide, How to Judge Whether a COA Method Fits the Research Question, explains that step in more detail.
Regulatory context and scope
The U.S. Food and Drug Administration’s data-integrity guidance states that data should be reliable and accurate, while current federal pharmaceutical CGMP laboratory-record rules call for complete test data, sample and lot identification, methods, graphs or spectra, calculations, results, and analyst and reviewer records. Those requirements apply in their specific pharmaceutical CGMP context. They are cited here as a useful framework for research-record literacy—not as a claim that every research material or supplier falls within that regulatory scope.
Primary sources
- U.S. FDA: Data Integrity and Compliance With Drug CGMP—Questions and Answers (December 2018)
- Electronic Code of Federal Regulations: 21 CFR 211.194, Laboratory records
Use documents as a connected record
Start with the lot, define the analytical question, then follow the evidence from sample identity through the authorized reportable result.
Editorial method: Published by bwell with AI-assisted drafting and human-directed editorial review. The article uses the cited primary sources and is educational, not a substitute for a laboratory quality system, method validation, or professional regulatory advice.