How to Write a Literature Review Without Drowning in Sources

A literature review isn't a list of everything you've read. It's an argument about what a field knows, where it disagrees and what's still missing, built from the sources that matter to your question.

Most people drown because they do those steps in the wrong order. They read first and look for a question later. The fix is to scope tightly, screen quickly, take notes that record exact claims with page numbers, and then organise by theme rather than by paper.

This guide takes you through that process step by step, with a synthesis matrix you can copy.

Key points:
1. A narrow, answerable question decides what you read and what you skip.
2. Good reading notes capture a claim, the evidence behind it and the page it came from.
3. Organise by theme, not by paper, and your review starts to write itself.

Scope the Question First

Before you open a database, write down the question your review answers. One sentence. If it takes a paragraph, it's too broad.

Compare these two:

Too broad: How does social media affect teenagers?

Scoped: What do longitudinal studies since 2015 say about the link between passive social media use and anxiety in 13 to 17 year olds?

The scoped version tells you which study designs count, which years to search, which age group to include and which outcome to track. It also tells you what to leave out, which matters just as much.

Then set your boundaries on paper:

• The years you'll cover.
• The kinds of sources you'll include (peer-reviewed studies, reviews, reports, books).
• The populations, settings or methods that are in or out.
• The languages you can actually read.

These are your inclusion criteria: the rules that decide whether a source belongs in your review. Writing them down stops you second-guessing every paper.

Search and Screen Without Getting Lost

Searching is easy. Screening is where the hours go. Work in passes so you never read a whole paper you'll end up discarding.

1. Build a search string from your question's key terms and their synonyms, and run it in the databases your field uses.
2. Record each search: where, when, what terms, how many results. Your supervisor or reviewer may ask.
3. First pass: screen titles only. Remove anything clearly off topic.
4. Second pass: read abstracts against your inclusion criteria. Keep, reject or mark as maybe.
5. Third pass: skim the full text of the keepers and maybes. Decide.
6. Follow the reference lists of your best sources, and check who has cited them since. This often finds studies your search missed.

Keep a simple log of rejected papers and the reason. When you're writing up your method, you'll be glad you did.

Take Reading Notes That Capture Claims and Pages

Here's where most reviews go wrong. People highlight, or write a general summary of each paper, and three weeks later they can't find the sentence they need or remember which page it came from.

Instead, take notes as claims. For each important point, record:

• What the source claims, in your own words.
• The evidence behind it (design, sample, key finding).
• Any limitation the authors admit, or one you spot.
• The exact page number.
• A short quote only if the wording itself matters.

A good note looks like this:

Source: Author, year, p. 214. Claim: Passive scrolling predicted higher anxiety scores a year later; active posting didn't. Evidence: longitudinal survey across two school years. Limitation: self-reported screen time. My note: contradicts the cross-sectional studies in theme 2.

That last line is where synthesis starts. You're already connecting sources while you read.

If your sources are PDFs, a research notebook that keeps page boundaries and answers questions with page citations removes a lot of the hunting. In Strut, clipping a passage into a note carries the file name and page number with it, and you can ask a source a question and see which pages support the answer.

Organise by Theme, Not by Paper

The classic weak review reads like this: "Smith (2019) found X. Jones (2020) found Y. Lee (2021) found Z." That's an annotated bibliography, not a review.

A synthesis matrix fixes it. It's a table where rows are your themes and columns are your sources, so you can see agreement and disagreement at a glance.

Theme	Study A	Study B	Study C	What this means
Passive vs active use	Passive use linked to anxiety (p. 214)	No difference found (p. 9)	Passive use linked, small effect (p. 57)	Mostly supported; B's sample was younger
Direction of effect	Use predicts anxiety later (p. 216)	Not tested	Anxiety predicts later use too (p. 60)	Likely runs both ways
Measurement	Self-report	Device logs	Self-report	Self-report may inflate the link

Fill it from your notes, not from memory. Empty cells are useful. They show gaps, and gaps in the literature are often the most interesting thing your review can point out.

Once the matrix is full, your themes become your sections. The right-hand column becomes the topic sentence of each paragraph.

Write the Review

With a scoped question and a filled matrix, the writing is mostly assembly. Follow a simple structure.

1. Introduction. State the question, why it matters and the scope you set.
2. Method (if required). Explain your search, databases and inclusion criteria.
3. Thematic sections. One per theme. Lead with the synthesis, then bring in sources as evidence.
4. Gaps and tensions. Where studies disagree, and what nobody has tested yet.
5. Conclusion. What the field knows, what it doesn't, and how your own work fits.

Here's the before and after in practice.

Before: Smith (2019) studied 500 teenagers and found passive use was linked to anxiety. Jones (2020) studied 300 and found no difference.

After: Evidence that passive use predicts anxiety is fairly consistent in older teenagers (Smith, 2019; Lee, 2021), but the one study with a younger sample found no effect (Jones, 2020), which suggests age may shape the link.

The second version makes a claim, then uses sources to support it. That's synthesis.

If you're drafting several themes at once, a research memo template is a handy way to write up one theme at a time before merging them into the full review.

Keep Citations Exact

Citation mistakes are easy to make and hard to find later. A few habits help:

• Cite from your notes, which already carry the page number, never from memory.
• Add the citation the moment you write the sentence, not in a cleanup pass.
• Check every direct quote against the original before you submit.
• Pick one reference manager or system and stay with it for the whole project.

For a deeper look at the options, our guide to citation management for writers compares approaches and where each one fits.

Where Strut Fits

Strut is a browser workspace where your notes, imported sources and draft sit in one project. For students, that means a thesis chapter and its reading notes stay together; see how Strut works for students writing essays and dissertations. For academics running bigger reviews, Strut for researchers covers source imports, opt-in OCR for scanned pages and page-cited answers.

The free plan includes one project, up to three documents, unlimited notes and source imports. If you need more, compare the options on the Strut pricing page.

The Short Version

Scope your question to one sentence and write down inclusion criteria before you search. Screen in passes, titles then abstracts then full text. Take notes as claims with page numbers, fill a synthesis matrix by theme, and let the themes become your sections. Cite from your notes as you write, and your review will be an argument, not a pile of summaries.