Scoping Search vs. Systematic Search: Which One Fits Your Question?

A scoping search is a fast, exploratory pass to learn the shape of a literature and decide whether a question is worth pursuing. A systematic search is a planned, documented, reproducible pass built from eligibility criteria and run across multiple sources. If you are finishing a course paper, a thesis chapter, or a first manuscript and starting the next project, the right choice depends less on ambition than on what you already know and what you need to produce next.

This article compares both options on the same task, shows where each one breaks down, gives you a decision table, and then combines them into one workflow you can run over a winter break or a January project start.

Before you choose a search type, take stock of your starting state. The two searches answer different questions, so the same reader can need either one depending on the day.

You are probably ready for a scoping search if:

• You have a broad topic but no settled question.
• You are unsure which terms the field actually uses.
• You need to know whether a gap is real before committing a semester to it.
• You are cleaning up loose ends from a finished chapter and want to see what sits adjacent to it.

You are probably ready for a systematic search if:

• You can already state eligibility criteria for what counts as relevant.
• You need a search another reader could rerun and reach the same records.
• You are writing a review, a thesis chapter, or a manuscript section where selection must be defensible.
• You have a supervisor or reviewer who will ask how you found your studies.

A useful diagnostic: try to write your eligibility criteria in three lines. If you can, you are closer to a systematic search. If you cannot, that inability is itself the signal to scope first.

A scoping search is exploratory. Its job is orientation: mapping vocabulary, spotting the major authors and review articles, and testing whether your question has enough behind it to justify a deeper pass. It is deliberately loose, and that looseness is the point.

Good scoping moves include searching by author and title, sorting by date, following citation trails, and checking related articles. Google Scholar documents exactly these behaviors: author and title search, date sorting and date limits, Cited by and Related articles, and ways to locate alternative versions of a work. PubMed offers field tags, phrase searching, filters, Advanced Search, saved searches, Similar Articles, and citation lookup, and its search details show how terms were translated and combined. Those features make a scoping pass efficient without requiring a formal protocol.

What a scoping search produces is a map, not a dataset: a shortlist of key papers, a working vocabulary, a sense of which sub-questions are crowded and which are thin, and a first guess at your eligibility criteria.

The Cochrane Handbook describes systematic searching as planned from eligibility criteria, using multiple sources, combining free-text terms with subject headings, documenting the process, deduplicating records, screening in stages, linking multiple reports of one study, and rerunning searches before publication. That list is a good checklist for what makes a search systematic rather than merely thorough.

The defining feature is reproducibility. A systematic search is an artifact someone else could rerun. That means:

• Eligibility criteria written down before searching.
• A concept-by-concept term plan, with free-text terms and subject headings where the database supports them.
• More than one source, so a single database's coverage does not silently define your evidence base.
• A record of the exact searches run, including dates.
• Deduplication and staged screening, so you can report how many records moved from identified to included.
• A rerun before you submit, so the search reflects the literature as of your writing date.

If you cannot yet write the eligibility criteria, a systematic search will produce a large, well-documented pile you cannot yet screen. That is the most common failure mode.

Imagine a concrete case: you finished a thesis chapter on a topic in your field, your supervisor's feedback asked you to "situate the contribution more firmly," and you have a January project start ahead. You want to know whether to scope or systematize.

Run both against the same task and the difference is one of sequence, not of quality. The scoping search tells you what to systematize. The systematic search tells you what to include.

A scoping search fails when it is asked to do a systematic job. If you present a scoping pass as the evidence base for a review or a thesis chapter, a reviewer will ask which sources you searched, which terms you used, and how you decided what counted. A loose pass has no good answer to those questions.

It also fails by drift. Because a scoping search has no stopping rule, it can expand indefinitely: one more Cited by list, one more related article, one more alternative version. Without a written question to converge on, the search becomes a hobby rather than a step.

And it fails by single-source dependence. If your entire map comes from one database, your sense of the field is shaped by that database's coverage and ranking, not by the literature. That is acceptable for orientation and unacceptable for a defensible selection.

A systematic search fails when it is run before the question is ready. If your eligibility criteria are vague, a carefully documented search will faithfully retrieve a large, well-organized set of records that you then cannot screen consistently. The documentation does not rescue the question.

It also fails on cost. Multiple sources, term planning, deduplication, staged screening, and a rerun before submission are real work. Run that machinery on a question you have not yet validated and you may spend a month systematizing a gap that a week of scoping would have shown to be crowded.

A third failure is premature closure. A systematic search is only as good as its term plan. If you skipped the scoping phase, you may not yet know the field's subject headings or the synonyms your free-text terms are missing, and the search will be reproducible but narrow.

If you are between rows, scope. The cost of a scoping pass is low, and its output — a draft question and draft criteria — is exactly the input a systematic search needs.

The two searches are strongest in sequence. Here is a workflow that moves from map to evidence base, with each tool placed at the step it actually supports.

Start broad. Search by author and title, sort by date, and follow citation trails to see which papers keep reappearing. Use a scholarly search interface to run these exploratory queries and collect candidates in one place; the WisPaper search interface is built for finding academic papers, which is exactly the job at this stage. Keep a running note of the terms that keep showing up — those become your subject headings later.

A scoping pass generates more candidates than you can hold in your head. Save the promising ones and upload the PDFs you already have so they sit together. My Library is where you save or upload papers and organize them, which turns a scattered shortlist into a working set you can return to.

Once your shortlist is assembled, ask questions across it: which methods dominate, where the findings disagree, which sub-questions are already saturated. Library QA lets you ask questions based on papers uploaded to My Library or added from search results, which is a fast way to test whether your draft question survives contact with the actual literature.

Now formalize. Write the inclusion and exclusion criteria in plain language, then build a concept-by-concept term plan combining free-text terms with subject headings. This is the step where the scoping pass pays off: you already know the field's vocabulary, so your systematic search will be both reproducible and wide enough.

Run your planned searches in more than one source, record the exact strings and dates, and export the records. Deduplicate before screening. Screen in stages — title and abstract first, then full text — and keep counts at each stage so you can report the flow.

Before your bibliography hardens, check it. TrueCite checks BibTeX against real academic databases to flag hallucinated references, which is a concrete safeguard at the moment your citation list becomes final.

Cochrane's guidance to rerun searches before publication applies to any systematic pass. Set a calendar reminder for the week before submission and rerun your saved searches so the included set reflects the literature as of your writing date.

Readiness looks different for each search type, and you should judge each against its own job.

Your scoping search is ready when:

• You can state your question in one or two sentences.
• You can list the field's main terms and the subject headings your databases use.
• You can name the major papers and the main review articles.
• You can write draft eligibility criteria in three lines.
• You have stopped adding candidates because new searches return familiar names.

Your systematic search is ready when:

• Your eligibility criteria are written and dated.
• Your term plan covers each concept with free-text terms and subject headings.
• You searched more than one source and recorded the strings and dates.
• You deduplicated and screened in stages, with counts at each stage.
• You linked multiple reports of the same study rather than counting them twice.
• You reran the search before submission.

If you cannot tick the systematic list, you are not failing — you are probably still in the scoping phase, and the right move is to go back one step rather than push forward.

Pick the row of the decision table that matches your situation today. If it says scope, block two hours this week: run author, title, and citation-trail searches, save the shortlist in My Library, and finish by writing your question and three lines of eligibility criteria. If it says systematize, block a day: write the criteria and term plan, then run the planned searches across multiple sources and record every string and date.

Either way, end the session with one written artifact — a draft question or a dated criteria list — because that artifact is what tells you, and your supervisor, that the search did its job.