A student's guide
For honours students, early-stage research students and anyone beginning academic research in a business school. Read it from start to finish, or jump to what you need and come back as your project grows.
A literature review is an intellectual process of developing, evaluating, organising and communicating knowledge, not simply collecting and summarising publications.
New to research?
A step-by-step guide for students who have never done research before. In plain English, it explains what research, a literature review and a research paper are, then takes you through six small steps: from finding your first 3 to 5 papers to writing your own research question.
Each step follows one rule: try it yourself first, then ask AI, then check the AI's answer against the paper. A glossary of research words and a checklist are at the end. When you are ready for more, the rest of this page goes further.
11 pages · 48 KB
1
A literature review is not simply a summary of papers that have already been published. It involves identifying, critically evaluating, comparing, organising and synthesising existing research to establish:
Literature reviews matter especially in business research, because organisational and economic problems usually involve several dimensions at once: people, technologies, institutions, markets and social contexts. A good review shows how these pieces of research fit together.
A strong literature review does not simply tell readers what previous researchers have said. It explains what we can learn from the research collectively, how that knowledge has developed, and what questions remain.
Descriptive writing
"Smith (2021) examined employee trust in AI. Jones (2022) investigated AI adoption. Lee (2023) studied organisational support."
Fictional teaching example: these are not real studies or findings.
Synthesised writing
"Research on organisational AI adoption highlights the importance of both individual perceptions and organisational conditions. While studies of trust emphasise employees' confidence in AI systems, research on organisational support suggests that adoption also depends on the wider environment in which these technologies are introduced."
Fictional teaching example: these are not real studies or findings.
The second example is stronger because it identifies relationships between ideas (individual perceptions and organisational conditions) instead of listing studies one after another. It tells the reader what the studies mean together.
Activity · Try it
Two new passages about remote work. Both are fictional: the authors and findings are made up for practice.
2
University libraries and academic skills teams offer practical guidance on understanding, planning, researching and writing literature reviews. Here are five free guides, open to anyone, in a suggested order. It is a recommendation, not a required path: start wherever you need to.
Understanding and writing a literature review
Academic skills guide
What a literature review is for, how to get started, ways to structure it (for example by theme, from broad to specific, or chronologically) and how to evaluate and write about the literature critically.
Why start here: a clear first overview of the purpose, structure and writing of a review.
Exploring literature review approaches
Library guide
An introduction to the different types of review, including traditional, systematic, scoping and rapid reviews.
Why it helps: different research purposes call for different methods. This shows you the range.
Understanding traditional literature reviews
Library guide
Guidance on traditional, narrative reviews: the kind of review often written as part of a research thesis.
Why it helps: if your review sets up your own study, this is probably the kind you are writing.
Planning a literature search
Library guide
How to turn your topic into key concepts and search terms, and combine them: alternative terms with OR, different concepts with AND. See also Brainstorming keywords and Building a search.
Why it helps: a planned search finds more of the right research and is easier to describe in your methods.
Conducting a comprehensive search
Library guide
A step-by-step approach for thorough searching: a first general scan of the literature and its terminology, checking reference lists and later citing papers, building your search, and choosing databases (you may need more than one).
Why it helps: for reviews that need to be thorough, such as a standalone or systematic review.
Each link opens in a new tab. These guides are published by UNSW Sydney and are free for anyone to read. Your own university's library and academic skills team will have similar guides, and can help you in person.
3
There are many kinds of literature review. Three broad approaches are common in business research. Choose one to see its purpose, typical questions, activities, strengths and limitations.
These reviews critically examine and synthesise literature to develop understanding: to explain how theories and ideas relate, to interpret a body of knowledge, or to suggest new perspectives. They are especially useful for developing an argument or setting out the theoretical foundations of a research project.
Narrative and integrative reviews share a lot, but they are not identical in every classification. In Snyder's (2019) overview, for example, narrative reviews are grouped with "semi-systematic" reviews, while integrative reviews are a separate approach aimed at new frameworks and perspectives.
Systematic reviews use explicit, transparent and structured procedures to identify, select, evaluate and synthesise the relevant research. They suit questions that are defined clearly enough to support a structured way of gathering evidence.
Documentation is central: record your search strings, databases and dates, your inclusion and exclusion criteria, and how many studies you kept or left out at each stage (often shown in a PRISMA flow diagram, a standard chart of these numbers).
Not every literature review has to be systematic, and a systematic review is not automatically better. It suits some questions and not others.
Library guide: Systematic reviews.
Bibliometric reviews analyse publication and citation data to examine patterns in academic research. They can help identify influential publications, the intellectual structure of a field, collaboration networks and emerging research themes.
Bibliometric analysis does not show whether research findings are valid, and it does not replace critical interpretation of the literature. A citation shows attention, not quality.
These approaches overlap, and many reviews combine them. The most appropriate one depends on your research question, the contribution you intend to make, the conventions of your discipline and your supervisor's guidance.
A review within an empirical thesis sets up your own study. It shows what is known, builds the theoretical foundation and makes the case for your research question. It is usually selective and argued.
A standalone review is the study itself, often published as a review article. It usually needs a more explicit method, broader coverage and its own contribution, such as a new framework or research agenda.
Not sure which you are writing? Ask your supervisor early: it shapes everything else.
Activity · Reflect
A traditional or integrative review is a common choice. Organising the literature around concepts (Webster & Watson, 2002) can help you build a clear foundation for your study.
A systematic review may fit, if your question is defined clearly enough. Tranfield et al. (2003) and Okoli (2015) explain how; your university library can help with searching.
An iterative, hermeneutic approach may help while your understanding is still forming: search, read and refine in repeated rounds (Boell & Cecez-Kecmanovic, 2014).
An integrative review may fit. Snyder (2019) describes integrative reviews, and a concept matrix (Webster & Watson, 2002) helps you compare perspectives.
A bibliometric review may fit, ideally combined with close reading of the key work. Start with Donthu et al. (2021).
That's normal at the start. Snyder (2019) gives a short overview of the options. Then talk them through with your supervisor.
This is a starting point for a conversation with your supervisor, not a definitive recommendation. Your research question, intended contribution and discipline come first.
4
Eight main activities. They are presented in order, but in practice they are iterative: reading will send you back to searching, and writing will sharpen your question. Open each step for details, a short example and a quick check.
Write your topic as a question, list two to four central concepts, and set boundaries: the time period, the contexts (industries, countries, types of organisation) and the kinds of study you will consider. Expect your question to change as you read. That is a sign of learning, not a failure.
Example Topic: AI at work. Initial question: "What influences employees' willingness to use generative AI tools?" Concepts: employees, generative AI, adoption. Boundaries: workplace studies from 2018 onwards.
Quick check: Can I state my focus in one sentence? Do I know what is out of scope?
Useful places to search include Scopus, Web of Science, Business Source Complete and Google Scholar, plus databases for your discipline (for example EconLit in economics, or the AIS eLibrary in information systems).
"generative AI" OR "large language model*".AND narrows a search (every concept must appear), OR widens it (any of the alternatives), NOT excludes words (use it carefully).Journal Basket writes a Scopus or Web of Science search limited to the ABDC-rated journals of your discipline: a focused place to start, not the whole search.
Quick check: Have I recorded each search, the date and the number of results?
For each publication, consider:
Journal ratings (such as the ABDC ratings this site uses) and citation counts can help you decide where to look first. They are not a substitute for judging each paper yourself.
Quick check: Could I explain why each key paper is in my review?
Skim the abstract, introduction and conclusion first to orient yourself, then read the most relevant papers closely. For each, identify:
Take notes in your own words, including your own thoughts. A reference manager such as EndNote or Zotero keeps track of your sources (see Citing and Referencing).
Quick check: For my key papers, can I say what they found and what they could not show?
Organising the literature around concepts, theoretical perspectives, debates, methodological approaches and research streams helps you move from individual papers to the bigger picture.
A concept matrix is a simple table with one row per publication and one column per concept. Ticking which concepts each study examines shows patterns at a glance: which ideas are studied often, which rarely, and which together. Build one in Example 1.
Quick check: Are my headings ideas or debates, rather than authors' names?
Compare findings, identify relationships between them, and evaluate competing explanations. When studies disagree, ask whether different contexts, theories or methods explain the difference. Aim for an integrated understanding of what the literature shows as a whole.
Useful phrases "Taken together, these studies suggest…" · "In contrast to…, … found…" · "These findings diverge when…" · "Building on…"
Quick check: Does each paragraph make a point that draws on several studies?
A review can reveal unanswered questions, unresolved contradictions, theoretical limitations, overlooked contexts, or assumptions worth challenging. These point to where your research can contribute.
Finding a topic with few publications does not, on its own, establish a meaningful research gap. Explain why the missing or inadequate knowledge matters, and for whom.
Quick check: Can I explain why answering my question would matter to researchers or practitioners?
Structure your review as an argument: tell the reader how it is organised, open paragraphs with a claim, and connect each section to your research question. Expect to revise it several times as your understanding grows, ask your supervisor for feedback, and run your searches again before you submit.
Quick check: Would a reader know, from my review alone, why my study is needed?
5
Most literature searches in business start in Scopus, Web of Science or Google Scholar. They look alike, but they work differently. Knowing the difference helps you search well and describe your search accurately in your methods.
Subscription database · Elsevier
A large database of journal articles, books and conference papers. An independent board chooses which journals it includes. For each publication it records the title, abstract, keywords, authors, journal and references, and it shows which later publications cite it.
You can search exact fields (for example title, abstract and keywords, journal, ISSN or year), keep a search history, and export results to EndNote or Zotero.
Subscription database · Clarivate
A citation database whose Core Collection includes journals chosen by Clarivate's editors, with citation records going back many decades. Its citation data are the basis of the Journal Impact Factor.
Like Scopus, it allows exact, repeatable searches and exports. The two cover many of the same journals, but not all, so many reviews search both.
Free search engine · Google
Finds scholarly writing across the web: journal articles, but also theses, books, preprints, conference papers and reports. It is free, very broad and good at finding full text, and its Cited by links make it easy to follow later research.
But it does not say exactly what it covers, its results change over time, it cannot limit a search to a list of journals, and its details are sometimes wrong. Use it alongside a database, not as the only search for a formal review.
Its AI search mode, Google Scholar Labs, is described in the guide to using AI tools.
| Question | Scopus | Web of Science | Google Scholar |
|---|---|---|---|
| What does it include? | Selected journals, books and conference papers | Selected journals, books and conference papers | Almost anything scholarly it can find online |
| Who pays? | Your university, through a subscription | Your university, through a subscription | Free |
| Can you limit a search to a list of journals? | Yes | Yes | No |
| Can someone repeat your search? | Yes, from your search text and date | Yes, from your search text and date | Not reliably |
| Can you export many results at once? | Yes | Yes | Only a few at a time |
That is why Journal Basket writes its searches for Scopus and Web of Science: they can search by ISSN (the number that identifies each journal), so a search can be limited exactly to the ABDC-rated journals of your discipline.
Scopus and Web of Science are not free. Most universities pay each year for licences that let their students and staff use them, and the companies have to check that you are one of them.
A free personal account in Scopus or Web of Science (made while signed in through your university) also lets you save searches and set up alerts for new articles.
6
Two influential methodological publications from information systems research offer distinct and useful guidance. They are not competing methods: each emphasises different things, and many researchers draw on both.
Approach A · Concept-centric
Webster and Watson provide influential guidance for writing concept-centric literature reviews. They emphasise organising research around important concepts, theoretical relationships and research themes, rather than presenting one summary after another.
Their best-known technique is the concept matrix: a table that organises publications by the concepts they investigate.
This approach can be particularly helpful when you are reviewing an established research area with identifiable concepts and theoretical traditions. It is guidance for organising and writing, not a rigid, linear procedure.
Diagram A · Concept-centric synthesis
Approach B · Hermeneutic
Boell and Cecez-Kecmanovic describe literature reviewing as an iterative process of developing understanding. Searching, reading, interpretation, critical assessment and argument development are connected activities. As you engage with the literature, your understanding changes. That may lead you to refine your searches, reconsider how relevant earlier publications are, challenge assumptions or reformulate your research question.
Their framework has two intertwined hermeneutic circles (hermeneutics is the study of interpretation):
Search and acquisition
Searching, sorting, selecting, acquiring and reading publications, then identifying new terms and sources and refining the search.
Analysis and interpretation
Reading, mapping and classifying the literature, critically assessing it, developing an argument, and (re)formulating the research problem and questions, which leads back to searching.
Reading connects the two circles. In the paper, the search circle sits within the wider circle of analysis and interpretation.
Understanding individual papers helps us understand the wider research field. As our understanding of the field develops, we may interpret those same papers differently.
The authors also warn about balance: too much searching without interpretation leads to growing confusion, while too much interpretation without searching risks missing important work (Boell & Cecez-Kecmanovic, 2014, p. 263).
Diagram B · Explore the hermeneutic circles
Select an activity to see what it involves, or follow a worked example through both circles.
Shared by both circles
Search and acquisition
Analysis and interpretation
Select an activity
Each activity is explained in plain words, based on the descriptions in Boell and Cecez-Kecmanovic (2014).
Diagram drawn for this guide, inspired by Figure 1 in Boell and Cecez-Kecmanovic (2014). The worked example is hypothetical.
| Dimension | Webster & Watson | Boell & Cecez-Kecmanovic |
|---|---|---|
| Main emphasis | Concept-centric synthesis | Iterative interpretation and understanding |
| Central concern | Organising and communicating existing knowledge | Explaining how understanding develops through engagement with literature |
| Searching | Identifying relevant research, including citation-based approaches | Searching and interpretation continually inform one another |
| Organisation | Concepts, themes, relationships, concept matrices | Evolving classifications, conceptual maps, critical interpretations |
| Strength | Developing coherent conceptual synthesis | Exploring complexity, refining understanding, challenging assumptions |
| Particularly useful when | Important concepts and research streams can be identified | The domain is unfamiliar, evolving, contested or conceptually complex |
| Potential challenge | Risk of premature or mechanical classification | Risk of insufficiently documented decisions or an expanding scope |
These are differences in emphasis, not exclusive features.
The two approaches can complement each other. You might use hermeneutic principles to explore an unfamiliar topic, building your understanding through repeated searching and reading. As the important concepts become clearer, a concept matrix inspired by Webster and Watson can help you organise the literature and develop a coherent synthesis. The two can also run side by side: new insights may prompt you to revise the matrix and search again.
Webster and Watson help us understand how to organise and synthesise research around meaningful concepts. Boell and Cecez-Kecmanovic help us understand how knowledge develops through searching, reading, interpretation, and reflection. Both offer valuable and complementary guidance.
7
Different business disciplines have developed their own research traditions and expectations for literature reviews. The publications below are influential methodological guides that can help students understand different approaches to reviewing scholarly literature. Some are particularly associated with specific disciplines, while others provide broader guidance applicable across business research.
These readings are influential methodological starting points, not an exhaustive list or a definitive ranking. Research traditions differ across disciplines, and students should consult their supervisors about the approaches most appropriate to their research.
Discipline-associated particularly associated with that discipline. Cross-disciplinary guidance broader guidance that works well there; it is not that discipline's accepted standard.
| Business discipline | Recommended methodological reading | Why read it? |
|---|
The readings could not be loaded. Please refresh the page.
If you read nothing else to begin with, these three give a strong foundation. They are suggestions, not compulsory readings, and not more important than every other contribution.
Webster & Watson (2002)
For understanding how to organise literature around concepts and develop a coherent synthesis.
Read the paperTranfield et al. (2003)
For understanding structured and transparent approaches to reviewing evidence.
Read the paperSnyder (2019)
For understanding the differences between review methods and how to choose an appropriate approach.
Read the paperComplementary reading
Boell & Cecez-Kecmanovic (2014)
Particularly valuable for understanding the iterative and interpretive nature of literature reviewing.
Read the paper8
Three hypothetical scenarios show how the ideas in this guide come together. The questions, studies and findings are invented for teaching.
Hypothetical example 1
"What factors influence employees' willingness to adopt generative AI tools in organisations?"
A student might organise the research around five concepts: trust, perceived usefulness, organisational support, social influence and risk. A concept matrix shows which studies examine which concepts, so similarities and differences stand out.
Activity · Build a concept matrix
Read each fictional study and tick the concepts it examines. Then check your matrix.
| Study (fictional) | Trust | Perceived usefulness | Organisational support | Social influence | Risk |
|---|
A concept matrix is an analytical aid, not a substitute for critical reading. It shows where concepts appear, not how well each study examined them.
Hypothetical example 2
"How do organisations understand and respond to concerns about AI sovereignty?"
A broad, unfamiliar topic like this suits a hermeneutic approach: understanding grows through repeated rounds of searching and reading.
Follow this example step by step on the hermeneutic diagram (choose "Follow an example").
Hypothetical example 3
"How does customer trust influence engagement with digital platforms?"
Relevant research may come from marketing, information systems, psychology and management. The challenge is that the same word can be defined and measured differently. Four imaginary studies might frame "trust" like this:
Before combining findings, compare how each study defines and measures trust, perhaps in a small definitions table. Then decide on a working definition for your review and explain why you chose it.
Writing one paragraph per paper.
Organise the discussion around concepts, arguments and relationships across studies.
Assuming highly cited papers are automatically high-quality evidence.
Evaluate relevance, research design, theoretical contribution and limitations.
Claiming a research gap simply because few papers exist on a topic.
Explain why the missing or inadequate knowledge matters.
Ignoring contradictory findings.
Examine disagreements, and consider whether different contexts, theories or methods explain them.
Treating literature searching as a one-time activity.
Revisit searches as your understanding develops, while documenting important decisions.
Assuming that all literature reviews must follow one prescribed method.
Select an approach that suits your research question and disciplinary context.
9
Activity · Check your own review
Tick each question you can answer "yes" to. Your ticks are not saved; copy the checklist if you want to keep it.
0 of 10 ticked
AI tools can help with parts of a review, such as finding search words, exploring a topic and checking references, but they cannot do the scholarly work for you. AI-generated citations may be fabricated, summaries can leave out important qualifications, and you remain responsible for reading and interpreting the original research. Rules differ between courses, so check what is allowed and acknowledge any use.
The guide to using AI tools shows how to use Claude, ChatGPT and Google Scholar Labs well in a literature search, step by step.
Your university will have its own rules on academic integrity; for example, see UNSW's guide to academic integrity and plagiarism.
About this guide: publication details were checked against Crossref and the publishers' pages on 17 October 2026. All examples, studies and findings marked as fictional or hypothetical are invented for teaching.