What Is a Lab Notebook? The Record That Makes Work Checkable
By the BrainSnail editorial team. How these articles are written and checked, and how to tell us when one is wrong.
A laboratory notebook records what was actually done, when, by whom and what happened, in enough detail that someone else could repeat it. It is a legal document, a defence against error and the thing most researchers keep worst.
What it is for
The record serves several purposes that pull in the same direction. It allows the work to be repeated, which requires recording what was done rather than what was planned, including the deviations, the failures and the conditions nobody thought mattered. It establishes priority, since a dated and witnessed record of when something was first done has settled disputes about who discovered what and is evidence in patent proceedings. It supports the published account, since a paper is a summary and the notebook holds what it summarises, which is what an investigation into a disputed result examines. It protects against a researcher's own memory, which reconstructs and improves the past in the way memory generally does. And it transfers knowledge when a person leaves, which is otherwise simply lost.
What good practice looks like
The conventions are consistent across fields and are mostly about making alteration visible:
- •Bound pages with numbers, so that nothing can be removed or inserted without it showing
- •Permanent ink, with errors struck through once and left legible rather than erased
- •Entries dated and made at the time rather than reconstructed later
- •Enough detail that a competent person in the field could repeat the work without asking
- •Raw data recorded directly rather than after calculation, with the calculation shown separately
- •Failed attempts recorded, since knowing what did not work is information and omitting it wastes the next person's time
Where it matters most
The record becomes decisive in situations nobody anticipates when keeping it. Patent disputes have turned on dated notebook entries establishing when an invention was made, and in jurisdictions that once awarded priority to the first inventor rather than the first filer, the notebook was the evidence. Investigations into misconduct examine notebooks first, since a paper reporting results with no underlying record is itself evidence, and several high-profile cases have turned on notebooks that were missing, altered or inconsistent with what was published. Regulated industries require records meeting specified standards, with pharmaceutical and clinical work subject to inspection. Disputes about authorship and contribution are settled by who recorded what. And ordinary errors are caught by going back, which is impossible without a record that says what was actually done.
Why it gets neglected
Almost everyone agrees the record matters and the standard is met unevenly, and the reasons are worth naming rather than moralising about. Writing up costs time immediately and the benefit arrives much later or never, which is the structure of every task people put off. The person doing the work knows what they did and cannot easily imagine not knowing, so the detail that will be needed is exactly the detail that seems unnecessary now. Nobody checks routinely, so the discipline depends on the individual. Training is frequently informal, with students learning from whoever supervises them and inheriting whatever that person does. And the incentives reward publication rather than documentation. Institutions that have improved matters have generally done so by making the record part of a workflow that produces something the researcher wants anyway, rather than by exhortation.
The electronic version
Electronic notebooks have replaced paper in many settings and the transition involved solving problems paper solved automatically. Tamper evidence required cryptographic timestamps and audit logs, since a file can be edited silently while a bound page cannot. Long-term readability is a genuine concern, since a notebook from thirty years ago is readable and a file in a discontinued format may not be, which makes format choices a preservation decision. Integration with instruments removes transcription errors and introduces a dependence on the instrument's own records. Search and sharing are the clear gains and are substantial. Backup is trivially easy and is frequently not done. The general finding from institutions that have made the change is that the discipline matters more than the medium, since a badly kept electronic record is exactly as useless as a badly kept paper one and is easier to produce.
The takeaway
The record must say what was actually done including the failures, in enough detail for someone else to repeat it, which memory cannot supply afterwards. Bound numbered pages, permanent ink and errors struck through rather than erased all exist to make alteration visible. Notebooks settle priority disputes and are the first thing a misconduct investigation examines.