Six Months, Forty-One Vials: Putting a Number on What It Costs to Give Materials Away

Nobody objects to sharing in principle. In practice, sharing is a small administrative industry that a lab runs without a budget line, a full-time employee, or a performance metric that rewards it. Between March and September of 2026, we decided to find out what that industry costs us. We logged every external request for a plasmid or strain that arrived in the lab, then timed each one from the initial email to the moment a tracking number went out. We did not change our practices midway. We simply kept a stopwatch and a spreadsheet.

The Ledger, in Plain Numbers

Forty-one requests arrived in roughly six months. Twenty-nine were plasmids, nine were bacterial strains, three were both. Thirty-three came from academic labs in the United States; eight were international, from six countries. We shipped thirty-six. Two requesters withdrew after reading our terms, one strain had been retired without our noticing, and two are still open at the time of writing because the corresponding material transfer agreement has not cleared.

The median elapsed time from request to shipment was thirty-four days. The median hours of hands-on staff time per fulfilled request was 1.7. Total staff time across the period: approximately seventy-eight hours, or roughly two full working weeks of a postdoc's attention. Nothing about that figure is dramatic in isolation. What matters is that it is invisible in every account we keep.

Where the Clock Actually Goes

Requesters tend to imagine a sequence of three steps: ask, receive, thank. Our log shows sixteen distinct handoffs, and only four of them involve anyone who knows how to streak a plate.

• Triage and reply (about 40 minutes). Reading the inquiry, confirming we still hold the construct, checking whether the person is asking for the material or for advice about the material. The two get confused constantly.
• The agreement (median 19 days). Once a request crosses institutional lines, the MTA usually governs. Our own template, filed with our sponsored research office, moves quickly; a counterparty's template rarely does. Three requests consumed more than eight weeks apiece, mostly in exchanges between two contracts offices over language about progeny, derivatives, and commercial rights.
• Physical recovery (about 2.5 hours). Pulling the glycerol stock from the −80, reviving it, verifying identity by colony PCR or sequencing, growing enough biomass to ship safely, and preparing the tube. This is the part with scientific risk, and it is the smallest part of the calendar.
• Outbound logistics (about 45 minutes). Dry ice, secondary containment, a courier account, a customs declaration for the international shipments, and a follow-up email with revival instructions that we rewrite every time.

The gap between 1.7 hours of labor and thirty-four days of elapsed time is the central finding. Our bottleneck is not the bench. It is correspondence, contractual vocabulary, and the absence of any shared convention about how long any of this should take.

What It Costs Beyond the Hours

Three costs do not appear in any of our tallies.

The first is delay as a form of quiet rejection. Two of our international requesters were graduate students on deadlines. By the time their agreements cleared, one had moved to a different project. We shipped to the other, but the experiment for which the strain was intended had been abandoned. We will never know what that collaboration might have produced, and neither will anyone else.

The second is the freezer's ledger. Every request we fulfill requires that we keep the stock healthy, re-verify it periodically, and maintain shipping documentation for years. The obligation runs in both directions; a lab that shares generously is also a lab that must maintain an archive. That maintenance has never appeared in a grant budget we have written.

The third is asymmetry of effort. Large, well-resourced labs receive more requests and typically have a technician or a repository deposit that absorbs them. Smaller labs absorb the same workflow into the principal investigator's inbox. Sharing, in other words, currently taxes the groups least equipped to pay.

Five Changes We Made Before September Ended

Our audit was uncomfortable enough to prompt changes, and each one was inexpensive.

1. A public sharing page. Our lab website now lists the plasmids and strains we distribute, whether an agreement is required, our typical turnaround, and the exact information we need in a first email. Requests that follow the template move about a week faster than those that do not, because nothing has to be negotiated twice.
2. Deposit before publication. For constructs likely to be requested, we now deposit with a nonprofit repository at the time of manuscript submission rather than after the first inquiry arrives. Roughly a third of our requests can then be answered with a link, and the repository, not a postdoc, handles the logistics.
3. A pre-negotiated institutional template. With our sponsored research office, we filed a standing default agreement and a short cover note explaining it. Counterparties that accept it skip the longest step in the process; the share has risen from about a quarter of requests to just over half.
4. A fixed shipping day. We ship on Tuesdays. This sounds trivial and is not: batching recovery and logistics lowered our per-request handling time by roughly twenty minutes and eliminated the midweek scramble that used to delay everything else.
5. A spreadsheet column nobody asked for. We now record requests we could not fulfill and why. In six months, that column has produced more useful information about our own record-keeping than any internal audit.

The Argument for Counting

Open science is often framed as a matter of conviction: share because sharing is right. We hold that conviction, and the audit did not weaken it. It clarified it. Policies that require sharing without funding the labor of sharing simply relocate the cost onto whichever trainee happens to be answering the inbox that month.

Our modest recommendation to other molecular biology labs is to measure first. Log a single season of requests, time the steps, and note where the calendar goes rather than where the effort goes. The numbers will be unflattering, and they will also be actionable. We cannot ask the field to treat materials as public infrastructure while continuing to run distribution out of a personal mailbox and a shared calendar. The vials are not the hard part. Everything wrapped around them is.