Home / Reading a report

How to read your certificate

Page 1 of a Cupel report, element by element, for someone holding one for the first time. The rendering is a page layout: the format exactly as issued, with illustrative figures. The numbered marks point at the explanations.

Cupel CL-2026-99901-6 Issued

Certificate of analysis

Semaglutide · purity 98.8% area, meets claim · identity confirmed

Example report · illustrative dataNo sample was analysed to produce this document. Serials 99900–99999 are example reports.

Laboratory

Name: Cupel Laboratories
Address: 2599 Speakman Dr
Mississauga, ON L5K 1B1
Web: cupel.ca
Email: help@cupel.ca

Report

Accession: CL-2026-99901-6
Status: Issued
Issue date: 16 October 2026, 09:10 UTC
Revision: 1 (first issue)
Analysis: 13–15 October 2026

Sample

Analyte: Semaglutide / CAS 910463-68-2 / C₁₈₇H₂₉₁N₄₅O₅₉ (free peptide)
Sender’s lot: EX-LOT-2026-0412
Sample source: Submitted by the seller, who is not named on this report under the terms (Publication)
Matrix: Lyophilised powder, 5 mg nominal fill per vial
Appearance: White lyophilised cake, intact; clear, colourless solution on reconstitution
Label claim: purity at least 98% (customer-supplied, not verified by the laboratory except as tested below)
Received: 12 October 2026, 15:20 UTC; vials sealed and intact; courier packaging undamaged; stored at 2–8 °C
Vials: 6 received, 5 consumed; 1 vial retained at −20 °C for 30 days from issue
Salt form: Trifluoroacetate as stated by the supplier, not determined by this laboratory

Results

Results, one row per test.
Test	Method	Result	Specification	Conclusion
Purity	RP-HPLC–UV 214 nm, area normalisation, USP <621>	98.8% area	at least 98% (label claim)	Meets claim
Identity	LC–MS, ESI+, accurate mass (Q-TOF)	Confirmed, max 5.6 ppm	within 10 ppm (laboratory criterion)	Confirmed
Mass per vial, free peptide	RP-HPLC–UV, external standard	4.30 mg/vial	–	–
Water content	Coulometric Karl Fischer, USP <921>	3.40% w/w	–	–
Bacterial endotoxin	Kinetic chromogenic LAL, USP <85>	0.41 EU/mg	–	–

– No specification supplied: reported as measured, for information. Purity 98.8 ± 0.3% (U, k = 2): meets 98% with the uncertainty included; decision rule with the method parameters. In plain terms and the mass balance on the mass-per-vial page are opinion and interpretation (ISO/IEC 17025 7.8.7); In plain terms was generated from the results above.

In plain terms

The material is confirmed as semaglutide. Purity is 98.8% by peak area. That meets the label’s claim of at least 98%. Each vial contains about 4.3 mg of semaglutide as free peptide, against a nominal fill of 5 mg. The rest of the powder is water (0.17 mg, 3.4% of the powder), the trifluoroacetate counter-ion (at most 0.47 mg by stoichiometry) and minor impurities (0.05 mg); 0.05 mg (0.9% of the fill) is not assigned. Bacterial endotoxin was 2.1 EU per vial, 0.41 EU per mg of the 5 mg nominal fill. No endotoxin limit was set for this material, so the figure is given for information.

An Open point row sits here when there is one. This report has none.

Analysed by

Example analyst Analytical chemist, for and on behalf of Cupel Laboratories Signed in the result record 15 October 2026, 17:42 UTC

Reviewed and released by

Example reviewer Quality manager, for and on behalf of Cupel Laboratories Signed in the result record 16 October 2026, 09:10 UTC

Verify

CL-2026-99901-6 cupel.ca/verify

SHA-256: 334e5df3217d8cf0992683c8d473c19a71d0c417ec5bcfa09601d7230f9abdca
Record: d30ab315c7a228302d4bd1b41c5e056237329d57953f70b2f03303cb3351790c

SHA-256 hashes the data summary (payload.json); Record, the complete result record (record.json, 6 478 values, the names and times above included). Both are served at cupel.ca/r/CL-2026-99901-6/ and checked at /verify, no account needed. The names above were typed into the result record and are not authenticated; this file carries no digital signature. Cupel holds no ISO/IEC 17025 accreditation; this is not an accredited result.

This result describes the composition of the sample as received under this accession only. It is not an assessment of safety, suitability for human or animal use, sterility, or regulatory status, and nothing in it is clearance to use any material. Results relate only to the item tested. Customer-supplied information is reproduced as given. This certificate shall not be reproduced except in full.

Cupel Laboratories · CL-2026-99901-6 · issued 16 October 2026 · revision 1 Page 1 of 9

Page layout. The figures are illustrative: no sample was analysed to produce them, and the number on it is not a report in the register. The layout is the composer's own page 1, checked word for word against the PDF it writes.

The number
CL-2026-99901-6 is the accession: our prefix, the year the sample was logged, a serial assigned on receipt and a check digit that catches a wrong digit or two neighbouring digits swapped before anything is looked up. The check digit is no defence against forgery, since anyone can compute one; the lookup is, because a number we did not issue is not in the register.
The headline
The line under the title is the finding in a few words, written from the same values as the results table. Where the vial carried a label claim it says whether the measured value meets it, which tests the seller’s statement, not a specification of ours.
The laboratory block
Who issued the document and how to reach them: the name, the web address and the email today, and the registered entity, postal address and telephone on every certificate issued after we publish them. A certificate with no way to reach the laboratory that signed it is one nobody has to answer for.
The report block
Status reads Issued or Retracted, and the issue date is the moment the report was released, after the reviewer had signed. Revision says whether this document replaced an earlier one, and Analysis is the span of days the tests ran.
The sample block
The sender’s lot is the reference on the vial, printed as given so a buyer can match the two, and the label claim is the sender’s statement, marked as theirs and tested only by the rows below. Appearance and Received record what the powder looked like and the state the box arrived in, Vials how many came, how many the tests used and how many we keep, and Salt form what the supplier stated, because the CAS number and the formula refer to the free peptide. Match three things before you rely on any report: the number, the lot reference and the analyte; if the vial’s lot is not the one printed, the report is not about your vial.
The results table
One row per test: the method in a chemist’s shorthand, the result as measured, the specification it is compared against with its source, and the conclusion that comparison gives. A dash in the last two columns means nobody supplied a limit, so the figure is reported as measured with no verdict; a dash is not a pass.
In plain terms
The same findings as the table, written for the person who bought the vial rather than the chemist who ran it. It is composed from the measured values, so it cannot say more than the table does.
The open point
When the method could not settle something, a peak it could not resolve or an impurity it could not identify, the front page says so in an Open point row under the plain terms, not in a footnote. This report has none; when one is printed, read it before you rely on the figure above it.
Signatures and titles
The analyst who ran the tests and the reviewer who released the report, each by name and title with the date and time they signed, both for and on behalf of the laboratory. The reviewer signs after the analyst; the report is then released by an authorised person who is not the analyst, and that moment is the issue date.
The two hashes and the QR code
SHA-256 is the hash of the report’s data summary (payload.json) and Record that of the complete result record (record.json), both served beside the report; a match shows the served file is the one the hash was printed for, and nothing more, since we publish the data and the hash alike; whether the paper in your hand says what that file says is settled by comparing its figures with the published report. The QR opens this number’s entry in the public register, which proves the number resolves, not that the paper in your hand is the one it resolves to: compare the figures.
The closing statements
The left column says what the hashes are of, where to verify, and that we hold no ISO/IEC 17025 accreditation, so this is not an accredited result. The right column limits the use of the result: it relates only to the item tested, customer-supplied information is reproduced as given, and the certificate is not to be reproduced except in full.
The footer
Every page carries the laboratory, the number, the issue date, the revision and its place in the whole, here Page 1 of 9. A page handed on alone still says which report it belongs to and how much of it is missing.

What the other pages hold

A report with these tests runs to 9 pages. Page 1 is the certificate; the rest is the evidence it rests on, one section per test, in this order.

Purity by RP-HPLC–UV. From page 2. System suitability against its criteria, the chromatogram as acquired at 214 nm with every peak at or above the reporting threshold, the peak table with relative retention and area percent, and the three injections with their mean and spread. SOP-02 (PDF, 10 pages, 112 KB)
Identity by LC–MS. From page 4. Each observed charge state beside its theoretical m/z and its error in parts per million, the spectrum drawn, and the instrument class and resolving power the 10 ppm criterion needs. SOP-03 (PDF, 7 pages, 96 KB)
Mass per vial. From page 5. The external-standard calibration with the sample response marked, the reference standard’s supplier, lot and stated content, and the mass balance of the vial. SOP-06 (PDF, 9 pages, 99 KB)
Water content. From page 6. The Karl Fischer replicates, one whole vial each, with the drift, the blank and the water-standard recovery. SOP-05 (PDF, 7 pages, 87 KB)
Bacterial endotoxin. From page 7. The LAL standard curve on log-log axes with its fit, the positive product control recovery, the negative control, and where the sample fell on the curve. SOP-04 (PDF, 7 pages, 88 KB)
Method parameters. From page 9. Column, gradient, flow, wavelength, source voltage and the rest, for every test, as this sample was run.

The procedure behind each section is published on the methods page; what the methods cannot determine is on the limits page.

Three things to check first

Before you trust any certificate, ours included.

Verify the number where the laboratory publishes it. Type it at /verify or scan the QR, and compare what comes back with the paper: the same analyte, the same lot, the same figures. A number that resolves nowhere, or only on the seller’s own site, leaves the certificate an assertion.
Recompute a hash. Download payload.json or record.json from the report’s page and hash it with SHA-256, or press the button there that does it in your browser. A match shows the file you hashed is the one the hash was printed for; it cannot show that the figures on your paper match it (compare them with the published report), nor that the laboratory measured correctly.
Read the specification column and the open point. A dash means no limit was supplied and no verdict was given, which is not a pass, and the open point is where a laboratory says what it could not resolve. A certificate with neither is one on which nobody had to say.