Most advice on literature reviews describes what one should contain. Far less explains how to get from an unsorted stack of PDFs to a document with a spine — a literature review that argues something, rather than simply reporting that things exist. The difference between the two is structure, and structure is decided before drafting begins.
Decide what the review is arguing
A literature review is not a list of summaries. It has a claim: the field agrees on X but disputes Y; most work assumes Z, and that assumption is rarely tested; the evidence clusters around one method and says almost nothing about another. Write that claim as a single sentence before you write anything else. Every source you keep should either support it, complicate it, or explain why it matters.
Group by idea, not by author or date
A common failure mode is the “and then” review — Smith found this, then Jones found that, then Lee found something else — which reads as a list because it is organized like one. Group sources by the position they take or the variable they isolate, then order those groups so the argument builds. Chronology can be a subplot inside a thematic group; it should rarely be the whole structure.
Write the synthesis sentence first, evidence second
For each section, draft the sentence that states what the cluster of studies collectively shows before you draft the summaries that support it. This reverses the usual drafting order and prevents a review from turning into parallel single-study paragraphs with a vague transition between them. If you cannot write that sentence, the section is not ready — you likely need to re-read for the pattern rather than write around its absence.
Leave the gap exposed
The strongest literature reviews end sections by naming what is still unclear, not by summarizing what was just said. State the gap plainly: what population hasn’t been studied, what method hasn’t been tried, what claim rests on evidence that is thinner than it looks. That gap is usually the seed of your own research question.
Before you begin drafting
Make a one-page map of the themes, the studies that belong in each, and the point of tension or uncertainty that each section must address. A writing workspace can help turn a selected source set into a provisional thematic outline, but the useful test is still yours: can you explain why each study belongs in that section? When you are ready to move from selection to a traceable first draft, the research workspace keeps that source set and its citations together.