There's no set number. The figures that exist come from individual programs, and in the handbooks below they run from 15 to 40 pages. Your program's handbook decides, and if it says nothing, your advisor does. Length is also a weak test: a review can reach its page count and still not support your research question.
Four examples, from a master's thesis to a doctoral project. They show the range, and none of them sets a standard for your program.
| Source | Level | Length of the literature review |
|---|---|---|
| University of Kansas, journalism | Master's | 20-25 pages in a thesis of about 50-75 pages |
| Bethel University, education | Master's | Minimum of 30-40 pages in one thesis format, 15-25 pages in the two formats that add an application chapter |
| Liberty University, music | Doctoral thesis | “The literature review should be at least twenty-five (25) pages in length” |
| UNSW Sydney, student guide | Thesis, general | “Around 15-30% of the whole thesis” |
Kansas adds a warning to its own figures: “Page numbers are rough estimates.” Bethel's two ranges point the same way: the longer one is for the format without an application chapter.
Your field matters too, and so does your question. A narrow question in a small field can run out of relevant studies quickly. A broad one can keep growing, which is often a sign that the question needs narrowing.
If your program gives no figure, you can start with a share of the whole. UNSW's rough guide of 15-30% would mean 9 to 18 pages in a 60-page thesis. That's simple math based on one university's guide, so use it as a starting point when you talk to your advisor.
UNSW also gives a warning sign at the high end: “If your literature review is more than 40% of your thesis, it’s probably too long.” Past that point, your own work is likely getting less room than the work of others.
Page counts won't tell you this. San José State University's Writing Center describes the job of a review this way: “It attempts to synthesize and evaluate the material and information according to the research question(s), thesis, and central theme(s).” So test it against your question:
| Too long | Too short |
|---|---|
| Sections on background no one in your field needs | A central concept is used later and never defined |
| Several studies making the same point, each with its own paragraph | One or two sources carry a whole argument |
| You can remove a section and nothing later breaks | Your method or measure appears in chapter 3 with no source behind it |
| The question is only mentioned in the last paragraph | No sign of disagreement in a field where people do disagree |
The same San José State handout is direct: “A literature review is not an annotated bibliography.” If each paragraph covers one article, try sorting the articles by what they claim, by method, or by where they disagree. The review often ends up shorter and more focused.
Esmond evaluates every assignment before you submit. It never writes it.
Esmond doesn't count your pages or check limits; your handbook and your advisor settle those. It doesn't open your sources either. It looks at how you use them in your draft and how well your chapters fit together, and it comments on passages where a claim clearly lacks support in your draft.
Ask your advisor. The handbooks above are mostly graduate level, so their figures may not apply to a senior thesis. A share of the whole is a better starting point than a page count from another program.
There's no set number. Liberty's doctoral handbook asks for a bibliography of at least 30 sources, but that applies only to that program. A better test: does each main claim in your review rest on sources you can defend?
Check your program's thesis guidelines. They say whether a limit covers the whole document and what it leaves out, such as references and appendices.
Stop when new sources stop changing what you can say about your question, and ask your advisor if you are unsure. Page counts are a poor stopping rule in either direction.
Read on November 3, 2026.
Esmond is not taking orders on this site yet.