Search Strategy

How to Build an Effective Search Strategy for a Systematic Review

Your search strategy decides which evidence your review can find. Here is how to define keywords, choose databases, combine Boolean operators, and test and document every search.

PhD Scholar Desk Research Team6 min readUpdated November 2026

Key takeaways

  • Break your research question into core concepts and list keywords, synonyms, acronyms and spellings for each.
  • Pick databases that fit your field, such as Scopus, Web of Science, PubMed, IEEE Xplore or ScienceDirect.
  • Combine terms with AND, OR and NOT, and use quotation marks and parentheses for precision.
  • Test, refine and fully document every search so another researcher could repeat it.

Define keywords and research concepts

A good search strategy starts with a well-defined research question. Break the question into its main concepts and list the most important keywords for each. Include alternative words, acronyms and spellings that authors might have used.

Example: a review on online learning

Useful keywords might include e-learning, digital education, virtual learning and distance education. These variations help you capture a broader range of relevant studies.

Time spent choosing the right keywords at the start improves both the quality and the completeness of your results.

Use databases and Boolean operators effectively

Choosing the right academic databases matters as much as choosing the right keywords. Depending on your field, databases such as Scopus, Web of Science, PubMed, Google Scholar, IEEE Xplore or ScienceDirect give access to high-quality publications.

Boolean operators and search syntax help you control what your search returns:

OperatorWhat it doesExample
ANDNarrows results by combining conceptse-learning AND engagement
ORBroadens results by including synonymse-learning OR “virtual learning”
NOTRemoves irrelevant topicse-learning NOT corporate
” “Searches an exact phrase“distance education”
( )Groups terms in complex search strings(e-learning OR “online learning”) AND engagement

Combining these techniques returns more relevant studies and filters out work that does not relate to your research.

Document, test and refine your search strategy

A search strategy should not stay fixed throughout the review. After the first search, check the results and make sure key studies are being found.

Too many irrelevant results?

Make the search more focused by tightening keywords or adjusting your Boolean queries.

Too few results?

Broaden the search with more synonyms and by searching additional databases.

Your search log should record

  • Databases searched
  • Search dates
  • Complete search strings
  • Filters applied
  • Number of records retrieved

This documentation improves transparency and lets another researcher repeat your review. It also feeds straight into your PRISMA flow diagram.

An effective search strategy is an ongoing process rather than a single search. Testing and refining your approach helps make sure your systematic review rests on comprehensive, relevant and high-quality evidence.

Final thoughts

These recommendations are based on widely accepted practices in academic writing and systematic reviewing, including structured search strategies and recognised reporting standards such as PRISMA. Review methods vary across disciplines, but these principles give you a reliable foundation for a high-quality systematic literature review.

Ready to put it into practice? Start with our guide to conducting a systematic literature review from scratch, and if you plan to map a field quantitatively, see how to do bibliometric analysis.

Struggling to build your search strategy?

We can help you turn your research question into tested, documented search strings across the right databases for your field.

Keep reading

This guide is for learning and planning. Review methods vary by discipline and university, so always check your programme handbook and supervisor’s expectations. Any work you submit must be your own, with every source properly attributed.