The literature review is the section of an assignment where the most marks are lost for the least obvious reason. Most students who struggle with it are not short on sources. They have read plenty, sometimes more than the word count can even accommodate. The problem is almost always structural: the review reads as a list of what different authors said, one after another, rather than an argument about what the literature as a whole actually shows.
That distinction, between summarising sources and synthesising them, is what separates a pass-level literature review from a strong one. This guide walks through what a literature review is actually for, how to structure it, how to find and select sources efficiently, and how to synthesise rather than list, with a worked example showing the difference directly.
What a Literature Review Actually Is
A literature review is not a summary of everything you have read on a topic. It is an argument about the current state of knowledge on that topic, built from the sources you have read, and structured to lead somewhere: usually toward a gap in the existing research, or toward the position your own assignment or dissertation is going to take.
Think of it less like a book report and more like a lawyer building a case. Every source you include is there because it supports a specific point in your argument about what is known, what is contested, and what remains unclear. A source that does not clearly serve one of those functions probably does not belong in the review, no matter how relevant it seemed when you found it.
Three Ways to Structure a Literature Review
Before you start writing, decide how you are going to organise the review. Three structures cover the vast majority of assignments.
Thematic structure
Groups sources by sub-topic or theme rather than by author or date. This is the most common structure for MBA and business assignments, because it naturally supports synthesis: sources addressing the same theme sit next to each other, making it easier to compare and contrast their findings directly.
Chronological structure
Organises sources by when they were published, often used when a field has changed significantly over time and the change itself is part of your argument, for example, tracing how thinking on remote work has shifted before and after 2020.
Methodological structure
Groups sources by research method (quantitative studies, qualitative studies, mixed methods), useful when the methodology used strongly shapes the conclusions reached, which is common in social science and management research.
For most business and management assignments, a thematic structure is the strongest default choice. It is also the structure the worked example and template further down this guide are built around.
Finding and Selecting Sources Efficiently
A common trap is spending disproportionate time searching for sources and comparatively little time deciding which ones are actually worth including. A faster and more effective approach:
- Start with a recent systematic review or meta-analysis on your topic if one exists. Its reference list is often a faster route to key sources than a fresh database search.
- Use Google Scholar’s “cited by” feature on one strong foundational source to find who has built on or challenged it since.
- Prioritise peer-reviewed journal articles over blog posts, consultancy white papers, or unreferenced online content, particularly for anything beyond foundation-level assignments.
- Set a rough source count target before you start reading in depth. For a 2,000-word literature review, 12 to 18 well-chosen sources is a reasonable range, more than that and you likely will not have room to synthesise properly within the word count.
- Read the abstract, conclusion, and discussion section first for each candidate source before committing to a full read. This alone eliminates a large share of sources that looked relevant from the title but are not.
How to Synthesise, Not Just Summarise
This is the skill that actually determines the grade. Summarising restates what a source said. Synthesising puts multiple sources into conversation with each other and draws a conclusion from how they relate.
Here is the same underlying content written two ways, first as summary, then as synthesis, so the difference is visible directly rather than just described.
Summary version (weaker):
Smith (2021) found that remote work increased employee productivity by 13%. Jones (2022) found that remote work decreased team cohesion over time. Patel (2023) argued that hybrid models balance these two effects.
Synthesis version (stronger):
The productivity gains reported by Smith (2021) sit in tension with Jones’s (2022) finding of declining team cohesion under sustained remote arrangements, suggesting that the benefits of remote work may not be uniform across outcome types: individual output appears to improve while collective functioning may deteriorate over time. Patel’s (2023) argument for hybrid models can be read as a direct response to this tension, proposing a structural compromise rather than resolving the underlying trade-off Smith and Jones identify.
Notice what changed. The synthesis version does not just report what each source found. It explains how the three sources relate to each other, identifies a tension between two of them, and positions the third source as a response to that tension. That relational move, explaining how sources agree, disagree, or build on each other, is what synthesis actually means in practice.
A Practical Structure for a 2,000-Word Literature Review
The table below shows a working structure and approximate word allocation for a thematic literature review at roughly 2,000 words, a common length for an undergraduate or MBA module assignment.
| Section | Approx. length | What it does |
| Introduction | ~150 words | States the scope, the central research question or theme, and how the review is organised. |
| Theme 1 | ~400 words | Groups sources addressing the same sub-topic. Ends with a synthesis sentence, not a summary of the last source cited. |
| Theme 2 | ~400 words | Same approach, ideally showing where this theme connects to or conflicts with Theme 1. |
| Theme 3 | ~400 words | A third thematic cluster, often the one with the least consensus in the literature, useful for identifying your research gap. |
| Synthesis & gap | ~350 words | Pulls the themes together and explicitly states what the existing literature has not yet addressed, which is what your own research or argument responds to. |
| Conclusion | ~150 words | Summarises the state of the literature and restates the gap or position your assignment will now address. |
Adjust the number of themes and the word count per section proportionally for longer or shorter reviews, but keep the underlying shape: introduction, thematically organised body sections that each end in synthesis, an explicit synthesis and gap section, and a conclusion that restates the gap rather than introducing new content.
Literature Review Samples
Reading a template is useful, but seeing what a properly synthesised paragraph actually looks like in a live literature review is more useful. Below are three short sample excerpts, each written the way a strong thematic literature review section should read: sources put in conversation with each other, ending in a synthesis statement rather than a list of findings.
| Sample 1: Leadership styles and organisational performance
The relationship between transformational leadership and organisational performance has been consistently supported in Western contexts (Bass and Riggio, 2018), yet its application in GCC business environments remains contested. Al-Malki and Juan (2018) found that transformational leadership behaviours were positively associated with employee commitment among UAE public sector employees, though the effect was moderated by perceived organisational support, a variable largely absent from the original Western models. This suggests that transformational leadership theory may require contextual adaptation rather than direct transplantation into GCC organisational settings, a gap that subsequent regional studies, including this research, attempt to address more systematically. |
| Sample 2: Digital transformation and SME competitiveness
Early digital transformation literature framed technology adoption primarily as a cost-efficiency exercise (Vial, 2019), a view increasingly challenged by more recent work positioning digital capability as a source of competitive differentiation rather than operational saving (Verhoef et al., 2021). This shift in framing has particular relevance for small and medium enterprises, where Chatterjee et al. (2022) found that digital adoption driven by efficiency motives alone showed weaker performance outcomes than adoption driven by strategic repositioning. Read together, these studies suggest that the motivation behind digital transformation, not merely its presence, may be a stronger predictor of competitive outcome than the technology adopted. |
| Sample 3: Employee engagement and remote work arrangements
Kahn’s (1990) foundational conceptualisation of engagement as psychological presence at work predates remote and hybrid working arrangements entirely, raising the question of whether the construct translates cleanly into distributed work contexts. Wang et al. (2021) found that engagement levels among remote workers were statistically comparable to in-office employees, but were driven by different antecedents, autonomy and trust rather than physical presence and informal interaction. This divergence in antecedents, rather than in outcome levels, may explain why organisational engagement strategies designed for co-located teams have shown inconsistent results when applied unchanged to hybrid workforces (Chanana and Sangeeta, 2021). |
Notice the pattern across all three samples: each one opens with an established position, introduces a source that complicates or extends it, and closes with an interpretive sentence explaining what the relationship between the sources actually means. That closing interpretive sentence is doing the analytical work. Remove it from any of the three samples above and what remains is a competent summary, accurate, but not yet a literature review.
AI Prompts for Writing a Literature Review
AI tools can genuinely speed up the slower parts of literature review writing, organising themes, checking whether a paragraph actually synthesises or just summarises, and spotting where your argument relies too heavily on one source. Used this way, as a structural aid rather than a replacement for reading and thinking, AI fits comfortably within most institutions’ academic integrity policies. Used to generate the review itself from source titles alone, it does not, and the result usually reads exactly like what it is.
| A note on responsible use: Always check your specific institution’s AI use policy before using any of the prompts below, and treat AI output as a first draft to verify and rewrite in your own words, not a final answer. AI models can also fabricate citations that look real but do not exist, so verify every source AI suggests or references before using it. |
| Prompt 1: Identifying themes across your sources
I am writing a literature review on [your topic]. Here are the key findings from my sources: [paste 5 to 10 short bullet points, one per source, with author and year]. Based on these, suggest 3 to 4 possible themes I could use to structure my literature review thematically, and briefly explain what would go under each theme. |
| Prompt 2: Checking if a paragraph synthesises or summarises
Here is a paragraph from my literature review: [paste your paragraph]. Does this paragraph synthesise the sources together into a connected argument, or does it just summarise each source separately one after another? Be specific about which sentences are doing which, and suggest how I could rewrite it to synthesise more effectively if needed. |
| Prompt 3: Finding the gap in your literature
Based on this summary of my literature review findings: [paste your themes and key points], help me identify what appears to be missing or under-researched in this body of literature. What questions does the existing research not yet answer, and how might my own assignment or research address one of these gaps? |
| Prompt 4: Stress-testing your argument
I am arguing in my literature review that [state your central claim or synthesis point]. Play devil’s advocate: what is the strongest counterargument or contradicting evidence someone could raise against this claim? I want to make sure I have addressed it or acknowledged its limitations. |
| Prompt 5: Checking for over-reliance on one source
Here is my literature review draft: [paste your draft]. Review it and tell me if any single source appears to be doing too much of the argumentative work, or if my analysis is too dependent on one or two sources rather than drawing on the full range I have cited. |
None of these prompts ask AI to write the review itself. Each one asks it to check, organise, or challenge work you have already done, which keeps you doing the actual thinking while using AI for what it is genuinely good at: pattern-spotting across a large amount of text faster than a first read-through usually allows.
Common Mistakes That Cost Marks
- Ending each paragraph with the last source cited rather than a synthesis sentence pulling the paragraph’s sources together.
- Organising by author instead of by theme, which almost guarantees a summary-style review rather than a synthesised one.
- Including sources with no clear function in the argument, simply because they were found during research.
- Never stating the gap explicitly, leaving the reader to infer why the review exists at all.
- Over-reliance on very recent sources at the expense of foundational work that the recent literature is itself responding to.
Need Help With Your Literature Review?
If you would rather have an expert structure and write your literature review directly, our team covers literature reviews for MBA, undergraduate, and dissertation-level assignments across all subjects.
