Guide · · By the tinysamples team
How to keep track of lab samples without losing your mind.
A workable sample-tracking system needs five things per sample: a permanent ID, an owner, a date, a location and a status — plus a log of what happened to it. Paper and spreadsheets can carry all five for a while; the usual breaking points are turnover and search. This guide builds the system tool-by-tool, starting free.
What does every sample record need?
Five fields: a permanent ID, an owner, a date, a location and a status — plus one log of what happened to the sample. That is the entire system. Paper can hold it, a spreadsheet can hold it, software can hold it. The tool matters far less than whether all six things exist.
- ID. Unique and permanent, written on the tube and in the record. Never reused, never renamed.
- Owner. A person, by name. Not a project, not the lab. Someone who can answer questions about the tube.
- Date. When it was made or collected, as YYYY-MM-DD. One format, everywhere.
- Location.Precise enough to walk to. “Freezer B” finds nothing; “Freezer B, rack 3, box 7, A4” finds a tube.
- Status. A short list agreed in advance — available, depleted, discarded. Three or four values, no more.
The log is the sixth thing, and the one most labs skip. It gets its own section below, because it is the part that saves you six months from now.
How should you name samples?
Short, unique, permanent and meaningless: a prefix plus a counter, like S-0142, written identically everywhere. Dates go in the date column, not the name. Meaning goes in the other fields, not the name — an ID that encodes the project breaks the day the project changes.
The tube that says “good one” is the failure mode. Naming has enough traps — Excel rewriting names into dates, hyphens versus underscores, missing leading zeros — that it has its own guide.
Why log actions instead of states?
Because states get overwritten and actions accumulate. “MK thawed S-042 on 2026-08-14” is a fact forever. A status cell that says “thawed” is only true until someone types over it, and every overwrite deletes a piece of history.
The difference shows up the day a result looks wrong. The question is never “what is the status of S-042” — it is “how many times has S-042 been thawed, by whom, and when.” A status column cannot answer that. A log answers it by being read.
The rule is append-only. Add a line for every action — made, moved, thawed, aliquoted, depleted — and never edit an old line. On paper, that is one dated line per action. In a spreadsheet, it is a second sheet with four columns: date, sample ID, person, action. The status field then becomes a summary you can trust, because the log backs it.
Paper, spreadsheet or software: which should you use?
The cheapest tool your scale allows. Paper works for one person and one box. A disciplined spreadsheet carries a small lab for years. Software earns its price when search, shared editing or turnover start costing you afternoons — not before. The table is the decision.
| Paper or spreadsheet | Tracking software | |
|---|---|---|
| Cost | $0 | Free tiers, then about $12–$29/mo |
| Search | Ctrl+F in the spreadsheet; page-turning on paper | Any field, from a phone |
| History | Only what someone remembers to write | Who and when, recorded with each entry |
| At the bench | Paper travels; the spreadsheet is at the desk | The phone already in your pocket |
| Survives turnover | With discipline and a named owner | By default — if the lab actually used it |
Disclosure: we make one of the software options. tinysamples is free for 100 samples and $12 a month after — that is the catch, printed. But if a spreadsheet with an owner and an append-only log is working, keep it. And if you suspect you need something much heavier, you almost certainly do not: take the five-question LIMS test.
What happens to samples when a student leaves?
They stay in the freezer, and the knowledge of what they are walks out the door. Samples outlive the people who made them — the tubes sit at -80 long after the defense. The fix is a handover checklist, run in the last month, not the last week.
- Every box the person owns is listed in the record, and the contents match reality. Open the freezer and check. Do not trust the box map.
- Every sample gets a new owner, by name, before the leaving date.
- Statuses are trued: depleted tubes marked depleted, discarded ones actually out of the boxes.
- The naming scheme is written down in one paragraph anyone can read.
- The record lives in a lab account, not a personal laptop or a personal Google Drive.
Two hours of this saves a successor two weeks of archaeology. The alternative — cryptic tube labels and a box map that was true two years ago — is not a hypothetical. It is the default.
Where do you start?
With the free template: a spreadsheet with the five fields as columns, an append-only activity sheet, and a short status list already set — the system from this guide, empty and ready. Download the sample tracking template, enter the samples you touch this week, and let the old boxes wait. When the spreadsheet starts costing you afternoons, reread the table above. Until then, it is enough.