Reviewers are not hostile readers. But they are fast ones. A literature review that runs three pages without establishing anything the study builds on will be skimmed, and a skimmed literature review produces comments like "the rationale for the study is not clearly established" and "the connection to existing literature is unclear" — both of which are requests to rewrite the section.
This post covers what a literature review is actually supposed to do, why most draft versions fail at it, and what to change.
Part of a series: This post is a companion to IMRaD Explained: The Structure Behind Every Scientific Paper, which covers the full paper structure including the Introduction, Methods, Results, and Discussion.
What the literature review is for
A literature review has one job: to create the intellectual space your study fills.
That is not the same as demonstrating that you've read widely. Reading widely is the prerequisite. The review itself is the argument that a gap exists, that the gap matters, and that your study addresses it.
When a reviewer moves quickly through a literature review, they are usually looking for two things: the gap statement and the justification for your methodology. Everything else is context. Context matters, but context is not the point.
Synthesis vs summarisation
The single most common problem in submitted literature reviews is summarisation where synthesis is required.
Summarisation describes what individual papers found. Synthesis draws a conclusion that couldn't come from any single paper alone. Side by side, the difference is immediate.
"Smith (2018) found that nitrogen volatilisation increased under high temperature conditions. Jones (2020) reported reduced NUE in flooded rice paddies in South Asia. Ahmed (2022) demonstrated that split application reduced nitrogen loss by 34%."
"Nitrogen loss under flooded conditions in South Asian rice systems is well documented (Smith 2018; Jones 2020), yet the relative contributions of temperature-driven volatilisation and anaerobic denitrification to overall NUE depression have not been disaggregated in field settings. Ahmed (2022) demonstrates that management interventions can reduce loss, but without mechanism-level explanation."
The summarisation version is an accurate account of three papers. It tells the reader nothing about the relationship between those findings, what they collectively establish, or what remains unresolved. The synthesis version uses the same three papers to make a point. The point is the gap.
The laundry-list problem
A literature review that runs through papers one at a time, in the order you read them, is a laundry list. Reviewers see a lot of laundry lists.
The fix is to organise by theme, not by paper. Each paragraph should have a topic — a claim about the state of knowledge — and the citations within it should be evidence for or against that claim. Papers that don't fit a theme either belong elsewhere or don't belong in the review at all.
The funnel structure
The literature review should narrow. It begins with the broad context of the field, moves to the specific problem your study addresses, and ends with the gap your study fills. Think of it as three concentric circles, each smaller than the last.
Most draft reviews spend too much time in the outer circle. Every introductory fact about soil nitrogen that every reader already knows does not need to be in the review. Get to the middle circle quickly. Get to the inner circle before the reviewer starts skimming.
Citation gaps
What you don't cite matters as much as what you do.
Reviewers who work in your area will notice the absence of key papers. A literature review on nitrogen use efficiency in South Asia that doesn't engage with the IRRI benchmarking work, or a review on PM₂.₅ in Bangladesh that skips the CAMS reanalysis data, signals unfamiliarity with the field rather than selective focus.
Recent work matters too. A review built primarily on papers from before 2018 in an active research area raises questions. One or two foundational older sources are expected. A review that treats them as the primary evidence base is not.
How to end the literature review
The literature review ends with the gap statement: the sentence or short paragraph that names precisely what is missing from existing knowledge and why your study addresses it.
Most researchers write this section last and treat it as a summary. It isn't a summary. It's a setup. The gap statement should make the reader feel that your study is not just interesting but necessary.
"Therefore, further research is needed to understand nitrogen use efficiency in Bangladesh."
"Existing NUE benchmarks for Bangladesh rice systems are derived from plot-scale trials under controlled irrigation. Whether those benchmarks hold under smallholder conditions with irregular water management has not been tested. This study addresses that gap by..."
The weak version states that more research is needed, which is always true of every topic. The strong version explains exactly what is missing and why this study is the logical response.
A note on tense
Tense inconsistency is one of the most common signals that a manuscript hasn't been revised carefully. The rule is simple.
What reviewers are actually checking
Reviewers use the literature review to answer a practical question: does this author understand the field well enough to have designed a credible study?
They check whether the gap statement follows logically from the literature presented. They check whether the methodology in the methods section is consistent with the approach implied by the gap. And they check whether key competing explanations have been acknowledged.
A literature review that answers all three checks quietly and efficiently, without demanding much attention, is doing its job. The sections reviewers remember are the methods, the results, and the discussion. The literature review's best outcome is that no one has much to say about it.