KEVOS
ArticlesServicesCase studiesAboutContact
ArticlesServicesCase studiesAboutContact
← ArticlesLiterature Search StrategiesProject Delivery · Research ProjectsLesson 55/85← PrevNext →
GuidePublished 16 Aug 20269 min readBy KEVOS Editorialliterature searchsearch strategykeywordstruncation
On this page

Ask about this page

KEVOS AILiterature Search Strategies

KEVOS knowledge first · trusted web sources when needed

KEVOS/Project Delivery/Research Projects/Reviewing the Literature
Project DeliveryResearch ProjectsCoreLiterature Review

Literature Search Strategies

Six rules govern whether a search returns the work that already answers your question or a wall of noise. None of them is difficult. All of them are routinely skipped.

Reading time10 minutes
LevelCore
Topic streamLiterature Review
Source materialReviewing the Literature
Updated2026-08-16

In brief

  • The supplied source gives six search rules and names five kinds of search target. It never explains the two Boolean operators it tells you to check.
  • Search is a skill with a technique, not a matter of typing your topic into a box and reading what comes back.
  • Lower case, synonyms, most important word first, quotation marks, truncation, and advanced filters — that is the whole rule set.
  • The single highest-value habit is generating synonyms before you search, not after you fail.

Why search technique decides the quality of your review

A literature review is only as good as the literature that reaches it. If your search misses the three papers that already engage with your question, your review will confidently claim a gap that does not exist — and an examiner who knows the field will find those three papers in a minute.

The supplied teaching source is blunt about the payoff: "By adopting effective research strategies you can save yourself a lot of time and obtain better quality articles." Both halves matter. Better technique is faster and returns better material, which is unusual — most efficiencies cost quality.

From the source

The source's framing of the task

"There are lots of research engines, but they work in different ways. To select the most relevant Database for your research, look in the library databases."

The operative phrase is they work in different ways. A strategy that works in one discovery layer can return almost nothing in another, which is why the source treats database selection as a step in its own right rather than an afterthought.

The five database types the source names

The supplied material names five kinds of search target. It describes what they are only by name, so treat the descriptions below as a working orientation rather than as the source's own definitions.

SEARCH TARGETS NAMED IN THE SUPPLIED SOURCE

Named in the sourceWhat it isWhen it is the right first stop
CatalogueNamed onlyBooks and physical holdings — where you go when you need a monograph or a classic text rather than a paper
SummonNamed onlyA single-search discovery layer across many databases at once — broad, fast, and noisy
Research ArchiveNamed onlyAn institution's own research outputs — theses, conference papers and case studies produced in-house
TroveNamed as a national databaseNational-scale aggregation including newspapers, archives and digitised material
A dissertation and theses databaseNamed onlyDoctoral and masters theses specifically — the fastest route to a full worked research design in your field

The source names these five and instructs you to "select the most relevant data base using your concepts/keywords". It gives no selection criteria beyond that.

Practice note

Start with theses, not journals — a synthesis, not the source's instruction

The source does not prioritise among the five. In practice, if you are writing a thesis, a thesis database is the highest-yield first stop, because a thesis shows you an entire worked research design — question, review, methodology, instrument, limitations — at the length you are being asked to produce. A journal article shows you a compressed result. Thesis structures compared works through five of these end to end.

The six search rules

As stated in the supplied source

  1. Type in lower case

    "Most search engines ignore capitals or may only pick up the Capitalised word." Capitalising is at best neutral and at worst silently narrows your result set.

  2. Generate synonyms before you search

    "Think of similar words of phrases that may help to access the information. Use a thesaurus to help." This is the rule with the largest effect on recall, and the one most often skipped.

  3. Put the most important word first

    Word order affects relevance ranking in many engines. Lead with the concept you cannot compromise on.

  4. Bind phrases with quotation marks

    "Combine words or phrases with 'quote marks'." Without them, a two-word concept is treated as two unrelated words that may appear pages apart.

  5. Truncate with a wildcard

    The source's own examples: globali*ation picks up both the American and the English spelling; cook* returns cook, cooks, cooking and cookery. One character replaces a list.

  6. Use advanced filters and categories

    "Use advance searches and categories of searches e.g. 'no' books, 'no' newspaper articles, 'yes' to research paper etc." Filtering by publication type is the fastest way to cut a large result set without losing the material you actually want.

Building a synonym set before you search

Rule two carries more weight than the other five combined, because it is the only one that increases what you find rather than tidying what you already have. The source tells you to use a thesaurus and stops there. The expansion below is a synthesis for this page, not the source's.

Four places synonyms come from

  • A thesaurus, as the source says — but a general thesaurus will miss field-specific usage
  • The keyword lists attached to papers you have already found useful
  • The subject headings a database assigns to a relevant record
  • The terminology of adjacent disciplines — the same phenomenon is often studied elsewhere under another name

Four kinds of variation to cover

  • Spelling variants — one wildcard covers both, exactly as the source's own globali*ation example does
  • Grammatical forms — manage, manages, managing, management, manager (one truncation covers all)
  • Register — practitioner language against academic language for the same idea
  • Era — a concept renamed a decade ago will hide older literature under its old label
Caution

The failure this rule prevents

The supplied source reports a recurring pattern: students claim there is nothing in the literature on their topic, and in virtually every case the reason is that they searched only for work that was exactly the same as their topic. A weak synonym set produces exactly that outcome — a thin result set, and a false conclusion that the field is empty. Why review the literature early works through the pattern in full.

Operators the source names but never explains

The database technique material instructs you to "check your use of OR & AND" when a search returns too many results. That is the only appearance of Boolean logic anywhere in the supplied material — the two operators are named, the instruction assumes you already know what they do, and no definition, example or truth table is given.

Source gap

A named technique with no explanation attached

AND and OR are named as things to check but are defined nowhere in the supplied source. This page will not invent a Boolean tutorial and attribute it to material that does not contain one.

What you can safely take from the source is the diagnostic: if a search returns too many results, your operator use is one of the three things to inspect, alongside the database's own refinement options and publication-type filters. If you need the operators themselves explained, that is a second source you will have to go and find — and the same is true of proximity operators, field-specific searching and nested queries, none of which the source mentions at all.

Recording the search so it can be defended

The supplied source treats searching as a private activity — nothing in it asks you to record what you searched for. That is a real omission at thesis level, where an examiner may reasonably ask how you know your review is comprehensive.

What to capture as you search — a synthesis, not the source's instruction

  • The database or discovery layer used, and the date you searched it
  • The exact search string, including quotation marks and truncation characters
  • Any filters applied — date range, publication type, peer-reviewed only, language
  • The number of results returned, and the number you marked as relevant
  • Synonyms tried that returned nothing, so you do not repeat them in three weeks
  • The point at which new searches stopped surfacing anything you had not already seen
Note

Why the last item matters most

The moment when fresh search strings stop returning fresh material is the only practical evidence you have that your search was adequate. It is not proof, and no supplied source offers a threshold for it — but it is a defensible answer to "how do you know you found everything?", and "I searched until nothing new came back, and here is the record" is a far stronger answer than a number of papers.

Where this sits in the wider process

Search technique is the first half of the first of the two processes the source identifies. The nine-step routine for running and refining an individual database search is set out in Database search technique. What to do with the results once they arrive — the credibility tests — is in Assessing the academic value of sources, and the reading routine that follows is in Evaluating an article before you cite it.

  1. Concepts
  2. Keywords and synonyms
  3. Choose database
  4. Search
  5. Evaluate results
  6. Save and locate full text

What to carry forward

  1. Lower case, synonyms, most important word first, quotation marks, truncation, filters — six rules, and the second one does most of the work.
  2. Build the synonym set before the first search, from thesauri, keyword lists, subject headings and adjacent disciplines.
  3. A thin result set almost never means an empty field. It usually means you searched for your exact topic instead of related work.
  4. The source names AND and OR without defining them. If you need Boolean logic explained, that is a second source.
  5. Record the search string, filters, date and result count. Search until nothing new comes back, and keep the record that shows it.

Frequently asked questions

Which database should I start with?

The supplied source names five types and tells you to select the most relevant using your concepts and keywords, but gives no ranking. If you are writing a thesis, a thesis and dissertation database is a high-yield first stop because it shows you complete worked research designs at the length you are being asked to produce.

Does capitalisation really matter?

The source says most search engines ignore capitals or may only pick up the capitalised word. The risk is asymmetric — lower case costs you nothing and may avoid silently narrowing your results, so there is no reason to capitalise.

What does a wildcard actually do?

It stands in for one or more characters so a single search term covers a family of words. The source's own examples are globali*ation, which catches both the American and English spellings, and cook*, which returns cook, cooks, cooking and cookery.

How do I use AND and OR?

The supplied source names both operators as things to check when a search returns too many results, but never defines them. That is a genuine gap in the material and this page does not fill it with invented instruction — you will need a second source for Boolean logic.

How do I know when I have searched enough?

No supplied source gives a threshold. The most defensible practical signal is that new search strings stop surfacing material you have not already seen. Record your searches so you can show that point was reached.

References and source attribution

  1. Supplied teaching source, Week 4: Reviewing the Literature — study notes, slide deck and printed notes.
  2. O'Leary, Z. 2017, The Essential Guide to Doing Your Research Project, 3rd ed., Sage Publications, London.
  3. Bryman, A. 2016, Social Research Methods, 5th ed., Oxford University Press, Oxford.
  4. Trochim, W. M. K. 2006, Research Methods Knowledge Base.

Suggested questions for Ask KEVOS

  • Build me a synonym set for a research topic on risk management in R&D projects.
  • What is the difference between a discovery layer and a subject database?
  • How should I record my literature searches so an examiner can audit them?
  • Why does my search return thousands of results, and how do I cut it down without losing relevant work?
  • What publication-type filters should I apply when I only want peer-reviewed research?

Related KEVOS knowledge

Database Search Technique: The Nine StepsCore · literature reviewAssessing the Academic Value of SourcesCore · literature reviewStep 3: Checking the LiteratureCore · research processWhat Is a Literature Review?Foundation · literature reviewNote-Taking and Bibliographic RecordsCore · literature reviewEvaluating an Article Before You Cite ItCore · literature review
KEVOS® · Project Delivery · Research Projects Page KVS-PM-RES-0055 · v1.0.0 · content 2026.08 Last reviewed 2026-08-16

Continue learning

The Five-Step Literature Review Development ProcessGuide · Research ProjectsNEXT LESSON →Database Search Technique: The Nine StepsGuide · Research ProjectsWhy Review the Literature EarlyGuide · Research ProjectsAssessing the Academic Value of SourcesGuide · Research Projects
KEVOS · Engineering, manufacturing and project improvement
ArticlesServicesCase studiesAboutContact
© 2026 KEVOS®