Avloryn LabsWhat is a literature review?
A literature review is a critical survey of what has already been published on a topic. Its job isn't to list studies — it's to organise them by theme, compare them, and reveal the gap your own work will address. It can stand alone or be Chapter 2 of a thesis.
For the full thesis context see the thesis format guide; for protocol-driven reviews see systematic review (PRISMA).
How a literature review is structured
- Introduction & ScopeTopic framing, boundaries and review questions.
- Methodology of the ReviewHow sources were found, screened and selected.
- Thematic AnalysisSources grouped into themes — the core of the review.
- Critical Synthesis & DebatesComparing studies; convergences and tensions.
- Identified Gaps & Future DirectionsThe evidenced gap your research addresses.
- ConclusionThe integrated picture and what follows.
Review vs summary
A summary describes each source on its own. A literature review synthesises — it groups sources by theme, compares and evaluates them, and builds one overall argument that ends in a gap. If your draft reads "Author A said… Author B said…", it's a summary, not a review.
The gap statement
Every strong review ends by naming a specific, evidenced gap — something existing work hasn't done, resolved, or studied in a particular context (e.g. the Indian setting). This gap is what justifies and frames the rest of the research.
Citations
Reviews are citation-dense, so every reference must be real and verifiable. LivoDraft builds the bibliography from DOI-verified papers in your chosen style (APA, MLA, IEEE, Harvard, Chicago, Vancouver). See the citation-styles guide.
How to search the literature properly
A review is only as good as the search behind it. Searching by typing your thesis title into Google is how reviews end up citing whatever happened to rank, rather than what the field actually says.
Build the search string first
Break your question into its two or three core concepts. For each, list the synonyms researchers actually use, join those with OR, then join the concept groups with AND:
(“soil stabilisation” OR “ground improvement”) AND (“fly ash” OR “GGBS” OR “industrial waste”) AND (“CBR” OR “bearing ratio”)
Run it in more than one place. Scopus and Web of Science for indexed coverage, Google Scholar for reach and for grey literature, PubMed for health, and a subject database for your own field. Coverage differs between them, so one database alone leaves gaps.
Then follow the citations
Take the most relevant papers you find and work in both directions: read their reference lists to go backwards in time, and check who has cited them to come forwards. This finds the seminal work a keyword search misses because it uses older terminology.
Know when to stop
Stop when new searches keep returning papers you have already seen. That point — saturation — is the honest signal that you have covered the field, not an arbitrary source count.
Record as you go
Save every source with its full citation and DOI at the moment you find it. Reconstructing references at the end is where fabricated and half-remembered citations enter a thesis. A reference manager, or our free Citation Studio, keeps the library and the bibliography in step.
How to actually synthesise
Everyone is told to synthesise rather than summarise. Almost nobody is told how. The practical method is a literature matrix: a table with one row per source and one column per thing you care about.
Once the matrix is filled, read down the columns instead of across the rows. Reading across gives you one paper at a time, which is a summary. Reading down shows you the patterns — where studies agree, where they contradict each other, which methods dominate, and which contexts nobody has looked at. Those patterns become your paragraphs.
What a synthesised paragraph looks like
A synthesised paragraph makes a claim about the literature and cites several studies as evidence for it: “Reported strength gains vary widely with curing period, from modest improvement at 7 days to substantial gains at 28 days (Sharma et al., 2021; Rao & Kumar, 2022), though almost all of this work is laboratory-based and field validation remains scarce (Menon, 2023).”
Compare that with “Sharma et al. (2021) studied… Rao and Kumar (2022) also studied…” — same sources, no argument. If your paragraphs each open with an author’s name, you are still summarising.
How many sources do you need?
There is no universal number, and any guide quoting one is guessing. What departments actually expect scales with the level of the degree:
Weight recent work heavily, but include the foundational papers whatever their date — a review of a mature field that cites nothing before 2020 signals that the search was shallow. And every source in the list must be genuinely cited in the text. An impressive bibliography full of works the author never engaged with is one of the easiest things for an examiner to test.
Common literature review mistakes
- Chronological order instead of thematic. Organising by publication date produces a timeline, not an argument.
- Purely descriptive. Reporting what each study found without ever evaluating how well it found it.
- No gap at the end. The review closes without naming what is missing, so the research that follows is unjustified.
- Citing only supportive work. Ignoring studies that contradict your position; a panel will raise them.
- Over-reliance on secondary sources. Citing a review that cited the original, rather than reading the original.
- Uncheckable citations. References with no DOI, wrong years or authors who never wrote the paper. Verify every one.
- The gap never returns. A gap named in Chapter 2 that the methodology does not actually address.
Draft a synthesised literature review
LivoDraft organises the literature by theme, synthesises critically (not source-by-source), ends with an evidenced gap, and cites real DOI-verified papers — print-ready Word.