COA Recordkeeping: Building an Audit-Ready Analytical Document File

A Certificate of Analysis that cannot be located, or that cannot be matched to the experiments it supports, provides no more assurance than no certificate at all. Recordkeeping is the operational half of analytical documentation: retention, organization, and the connective information that links a document to a lot and a lot to a result. This guide covers how long to keep what, how to structure the file, and what to log alongside each certificate so the archive answers questions later instead of raising them.

Why the file matters after the purchase

Certificates get requested at purchase and then, in many labs, forgotten. The moments when they matter come later. An anomalous result raises the question of whether the compound was the variable, which is answerable only if the lot and its documentation can be retrieved. A manuscript under review draws a request for compound characterization details. A collaborator asks which material a shared dataset used. A batch changes mid-study and the question becomes whether the change is visible in the data. Each of these is trivial with an organized archive and effectively unanswerable without one, since reconstructing lot identity from memory or from an old invoice rarely works. The archive is also what allows a supplier's consistency to be assessed over time, because comparing this batch to the last three requires having kept the last three.

Retention periods

Retention should be set by the longest applicable requirement rather than by convenience. The working floor is the duration of the research plus the period during which questions can still arise, which for published work extends well past the experiments themselves. Institutional research data policies commonly specify multi-year retention after publication or project close, funder terms frequently impose their own minimums, and journals may require characterization data to remain available for a defined period. Where work is subject to any regulatory or contractual framework, that framework's retention clock governs and is usually longer. The practical guidance is to identify which of these applies, take the longest, and set the retention period there rather than reasoning from the material's shelf life; the compound expires long before the questions about it do. Certificates occupy negligible storage as digital files, so over-retention costs little while under-retention can be unrecoverable.

What to keep alongside the certificate

The certificate alone is thin. What makes an archive useful is the connective information stored with it. Keep the certificate itself, and the raw chromatogram and mass spectrum where they were provided, since the summary document loses the detail that later questions tend to concern. Record the supplier, order date, lot number, and quantity received. Note receipt condition, including whether cold-chain shipments arrived cold, along with the date, since handling context is the part no document supplies. Log the storage location and any subsequent handling events such as freeze-thaw cycles and container openings. Most importantly, record which experiments used which lot, because that cross-reference is what turns the archive from a pile of documents into an investigative tool. A short entry linking lot number to experiment identifier at the time of use costs seconds and is nearly impossible to reconstruct afterward. Where cold-chain integrity is part of the record, it connects directly to lyophilization and stability considerations.

Organizing the file

Structure matters less than consistency, but some conventions hold up better than others. Filing by compound, then by lot number, matches how questions usually arrive, since the question is typically about a specific compound and then narrows to a batch. A consistent filename carrying compound, lot, and receipt date makes the archive searchable without opening anything. Digital scans or supplier-provided PDFs are appropriate as the primary record; the value is in retrievability, not in physical originals, though anything received on paper should be scanned promptly rather than filed loose. Storage should be backed up and shared rather than living on one person's local drive, because staff turnover is the most common way an archive becomes inaccessible. A single index, a spreadsheet is sufficient, mapping lot numbers to compounds, receipt dates, storage locations, and the experiments that used them provides the cross-reference layer that file structure alone cannot. That index is the piece most labs skip and most often wish they had.

Keeping it audit-ready

Audit-ready means a specific question can be answered from the file without reconstruction. A useful test is to pick a past experiment at random and try to produce, in a few minutes, the compound's lot number, the certificate for that lot, its purity result, the receipt date, and the storage history. If that takes an afternoon, the archive needs work. Common failure points are predictable: certificates saved with default filenames that identify nothing, lot numbers recorded in a notebook but never linked to documents, files on a departed researcher's drive, and material received without a certificate that was never followed up. Each is cheap to avoid and expensive to fix later. Building the habit at receipt, filing the certificate and logging the lot the day material arrives rather than at some future cleanup, is what keeps the file coherent. A quarterly check that recent acquisitions are all filed and indexed catches drift before it accumulates.

FAQ

How long should Certificates of Analysis be retained?

Take the longest applicable period among institutional policy, funder requirements, journal expectations, and any regulatory framework governing the work. That is typically several years past project close, well beyond the material's usable life.

Are digital scans acceptable as records?

Yes. Retrievability matters more than physical originals for most research contexts. Scan anything received on paper promptly and store it with backup.

What is the most commonly skipped record?

The link between lot number and experiment. Certificates get filed while the cross-reference to which work used which batch does not, and that link is the one that makes later investigation possible.

Should raw chromatograms be kept, not just the certificate?

Yes where they were provided. Later questions usually concern impurity profile or peak shape, which the summary document does not preserve.

What if material arrives without a certificate?

Request it immediately and record the gap if it is not supplied. An undocumented lot should be identifiable as such in the archive so its use can be assessed later.

---

Research Use Only: All compounds sold by Onward Aminos are intended exclusively for laboratory research. Not for human or animal consumption. These products are not drugs, supplements, or food. Statements have not been evaluated by the FDA. Must be 21+ to purchase.