By Cristian Bucsa

Cristian Bucsa is the founder of ReviveLab, a Canadian commercial supplier of research-use-only materials.

A specialized research material arrives at a Canadian university.

Finance can usually show that it was purchased. The supplier may have provided a certificate or other documentation. Someone may have checked the package when it arrived. Months later, that same material may appear in an experiment, a dataset or work moving toward commercialization.

The problem is that those records often live in different places.

The purchase order is in a financial system. Supplier documents are stored somewhere else. 

Receiving information may be in a log or an email. The researcher uses an internal sample name in a notebook or dataset. By the time someone needs to reconstruct the trail, the connection between those records may depend on one person remembering what happened.

That is a surprisingly important blind spot in research procurement.

We tend to treat procurement as an administrative transaction: request, approve, buy, receive, pay.

For ordinary office supplies, that makes sense. But some research materials become part of the evidence behind the research itself. A particular lot, storage condition, supplier document or receiving discrepancy may matter when someone later asks how a result was produced.

Canada already has strong expectations around the records on either side of this gap. What is less obvious is who is responsible for keeping the connection intact.

The rules already point in the same direction

Canada’s federal research funding system does not treat recordkeeping as incidental. 

Under the agreement governing administration of Tri-Agency grants and awards, institutions are required to keep complete and accurate records of the use of agency funding, including verifiable audit trails and supporting documentation for transactions.

On the research side, the Tri-Agency approach to research data management emphasizes documentation, storage and preservation throughout the research lifecycle. Tri-Agency guidance also makes an important distinction: research materials are the objects used in an investigation, while research data are created from those materials through research methods or practices.

That distinction matters. 

If a research material helps generate the data, then knowing which material was actually used can sometimes be part of understanding how those data were produced.

The responsible-conduct framework points in the same direction. Research integrity depends on complete and accurate records of data, methodologies and findings that allow work to be verified or replicated.

None of this means every box delivered to a laboratory needs a chain-of-custody dossier. 

It does mean we should pay more attention to the point where the administrative record ends and the research record begins.

The missing link is continuity

Consider a fairly ordinary scenario.

A researcher orders a specialized material. The purchasing system records the supplier, price and invoice. The supplier documentation identifies a catalogue number and lot. Receiving confirms when the package arrived and perhaps whether there was visible damage or a temperature issue.

Then the material enters the lab.  Someone gives it an internal identifier. It moves into storage. A portion is used several weeks later. 

The experiment is recorded under the internal identifier.

Each individual system may be doing exactly what it was designed to do.

But if the internal identifier is not connected back to the supplier and lot, the institution has separate records rather than an evidence trail.

That distinction becomes visible when something goes wrong.

Perhaps two experimental runs produce different results. A collaborator asks which lot was used. A supplier later corrects a document. A research-integrity review requires the work to be reconstructed. A technology-transfer team is performing diligence before licensing an invention. A researcher leaves the institution and someone else inherits the project.

At that point, the question is no longer whether the records exist.

It is whether another person can follow them without guessing.

This does not require another giant software system The obvious institutional response would be to build another database. I do not think that is necessary in most cases.

The purchasing record can stay in the purchasing system. Supplier documents can remain in document storage. Research data can remain in the appropriate research environment.

What institutions need is a reliable bridge between them when the material matters to the research.

A practical approach could start with four things.

First, define which materials require greater traceability. A routine consumable should not create the same administrative burden as a specialized reagent, reference material or other input where batch identity, storage or documentation could affect interpretation of the work.

Second, preserve a few identifiers at receiving. Supplier, catalogue or product identifier, lot or batch, purchase reference, date received and documentation location will often be enough to establish the starting point.

Third, maintain the translation between the external identifier and whatever identifier the research team uses internally. That small step is easy to overlook, yet it is what allows someone later to connect a result to the material that entered the institution.

Finally, record exceptions. A damaged package, temperature excursion, lot mismatch, missing document or supplier correction is more important to preserve than another uneventful delivery. 

The goal is not to document everything.

The goal is to make the important records connect.

Procurement is part of the research infrastructure

Canada spends considerable effort improving research integrity, data stewardship, commercialization and the responsible administration of public research funding.

Procurement rarely appears in those conversations except as a cost-control function.

For specialized research materials, that view is too narrow.

The purchasing process creates some of the earliest records about an input that may later contribute to research data, publications, intellectual property or commercial development. Losing the connection at the laboratory door makes every later review harder than it needs to be.

A useful test for an institution is simple.

Choose a specialized material that entered a laboratory six months ago. Can someone who was not involved in the original purchase determine what was ordered, what lot arrived, whether there was a receiving problem, where the relevant supplier documentation is, and which research work used the material?

If answering those questions requires several people, multiple systems and a search through old email threads, the institution may not have a documentation problem.

It has a continuity problem.

Canada does not need to turn every research purchase into a compliance exercise. But when a material becomes part of the path from public funding to research evidence, we should be able to follow that path.

The evidence trail should not end at the purchase order.

R$