What to write down
The database, the exact string, the date you ran it, any limits applied, and how many results came back. Five items, and they fit on one line of a document you keep beside your reading.
The result count matters more than it looks, because it is how you will know later whether a re-run genuinely found new material or whether you have rebuilt the search differently.
Why the exact string
Because a search paraphrased from memory is a different search. Truncation, phrase marks and which fields were searched all change the results substantially, and none of that survives being remembered.
Recorded properly, the string is a reproducible object. That is what turns a literature search from an activity into a method.
Database accounts save searches too
Most academic databases let you save a search to an account and re-run it in one click, and many will email you when new results appear. That alerting is genuinely useful during a long project.
It is not a substitute for writing the string down, because accounts expire, institutions change providers, and you cannot paste a saved search into a methods section.
Translating between databases
Each database has its own syntax and its own controlled vocabulary, so a string is adapted rather than copied. Recording both the original and each translation means the work is done once.
For any systematic review that record is required. For ordinary work it is what stops you rebuilding the same search from scratch for the second database.
It becomes part of your method
Where your work has a methods section, the search belongs in it, and having recorded it as you went means writing that section is transcription rather than reconstruction.
Reviewers and committees ask about searches specifically, because a search nobody can repeat is a search nobody can evaluate.
The re-run before submission
Long projects should search again shortly before submitting, because a study published while you were writing is exactly the one an examiner will know about. A recorded search makes that a short task.
Finding something new at that point is inconvenient and it is far better than being asked about it in a viva having not seen it.
Keep it with the reading
One document, holding the searches and the annotated sources they produced. That way the provenance of every source in your library is visible: you know where each came from and how you found it.
This is the same principle as a reference list, applied one step earlier in the process.
What it costs
About fifteen seconds per search. Set against reconstructing a search months later, or being unable to answer a question about how you found your literature, it is among the cheapest habits in academic work.
It is also one of the few that makes the next project easier as well as this one, since a good string on a topic is reusable.
The general principle
Everything this letter recommends comes down to recording what you did while you are doing it. Sources verified as they are collected, notes written as you read, searches saved as they are run.
Each one costs seconds in the moment and hours if deferred, and the deferral always happens at the point when hours are what you have least of.
Questions this raises.
Do I need to record searches for coursework?
It is rarely required and it is still worth about fifteen seconds. The value arrives later: when you need more recent sources, when a related assignment covers similar ground, or when a marker asks how you found your literature. For any systematic work it stops being optional.
Is a saved search in a database enough?
Useful and not sufficient. Accounts expire, institutions change database providers, and a saved search cannot be pasted into a methods section or translated to another database by somebody reading your work. Save it in the database for convenience and write the string down for the record.
How do I re-run a search for new results?
Run the identical string with a date limit covering the period since you last searched. That is why recording the date matters. Most databases also offer an alert that emails new matches automatically, which is worth setting up at the start of any project lasting more than a few months.
What if I refined my search several times?
Record the final version and note that earlier iterations were narrower or broader. For a systematic review the full development is usually reported, often as an appendix. For other work the string you actually used is what matters, provided it is the one that produced the sources you cite.
Where should I keep all this?
Wherever your reading lives, so that a source and the search that found it sit together. A single document alongside your notes is enough. What matters is that the provenance of everything in your library is recoverable, which is the same standard this letter applies to citations.
Dr. Marisol Quintanilla
PhD, Public Health, on the mph and research side of the hall. Writes for The Footnote and coaches the students who bring this work to the desk. The rest of the specialists.
Elsewhere in the hall.
Send the assignment and the deadline.
A specialist credentialed in your own field reads what you actually have and replies within two hours with a plan and a price. The reading costs nothing, and if the honest answer is that we are not the right hall for it, you will get that instead.
