01 · The Question
Are You Recording What a Paper Says, or What It Does for Your Review?
You finish reading a paper and carefully record its purpose, sample, method, major findings, and conclusion. Months later, you reopen your notes and find an accurate summary. There is only one problem: you can no longer remember why you saved the paper in the first place.
This happens because describing a paper and establishing its relevance are different intellectual tasks. A summary tells you what the authors did and found. A relevance note tells you what that information contributes to the question you are trying to answer.
That distinction becomes increasingly important as your literature collection grows. When dozens or hundreds of papers are involved, remembering why each source mattered at the moment you read it becomes unrealistic. Your notes need to preserve that reasoning for you.
03 · What You Need to Know
The Difference Between Recording Content and Recording Relevance
When researchers take notes, the most obvious information to capture is descriptive: What was the study about? What method did the researchers use? Who participated? What did they find?
Those details matter. Depending on your review, you may need much more precise information about design, measures, theoretical framework, context, limitations, or results. The problem arises when your notes stop there.
A summary answers, “What does this paper say?”
A descriptive note reconstructs the paper. For example:
“The authors surveyed university students about their use of generative AI and reported an association between frequency of use and several self-reported learning behaviors.”
That may be an accurate and useful record. But it does not tell your future self why this study belongs in your review.
A relevance note answers, “Why does this matter here?”
A relevance note connects the source to your own research problem:
“Useful for the section on student AI use because it provides evidence of an association between usage frequency and reported learning behaviors, but it cannot establish that AI use caused those behaviors.”
Now the note performs another function. It identifies where the paper may fit, what contribution it makes, and an interpretive qualification that may matter when the evidence is synthesized.
What the paper says
A record of the authors' question, methods, evidence, findings, arguments, or conclusions.
Why the paper matters
Your reasoned assessment of how the source contributes to the question, theme, comparison, argument, uncertainty, or gap you are examining.
Keeping these two layers distinguishable also helps you separate the authors' claims from your own interpretation. Otherwise, a comment you wrote six months ago can begin to look suspiciously like something the authors themselves concluded. Academic note-taking occasionally produces its own tiny archaeological problems.
“Why it matters” is not the same as “this is an important paper”
You do not need to assign every source a grand statement of significance. Relevance is relational. A paper matters to something.
It might matter because it provides evidence for a claim you are examining. Another paper may matter because it contradicts that evidence. A third might use a substantially different population or method and therefore complicate what initially looked like a consistent finding.
A source can also matter because it gives you a useful definition, introduces a theoretical distinction, documents the development of an idea, exposes a methodological limitation, or provides evidence about a context poorly represented elsewhere in your literature.
This is one reason synthesis matrices are commonly organized around themes or concepts rather than merely storing independent paper summaries. Stanford's guidance on synthesis matrices, for example, emphasizes relating sources to one another according to themes, topics, keywords, or codes so that agreement, disagreement, and divergence can be identified.
The same paper can matter for several different reasons
Do not assume that every paper needs one permanent relevance label. One study might provide theoretical background in one part of your review, empirical evidence in another, and a useful methodological contrast somewhere else.
This is particularly important when you have papers that belong to more than one theme. Relevance is often many-to-many: one source may contribute to several ideas, and one idea will usually depend on several sources.
The University of Sheffield's literature-review guidance similarly recommends examining whether sources build on one another, challenge one another, reveal debates, identify gaps, or collectively complete a larger picture. These relationships are part of synthesis rather than simple source description.
Record relevance while you still remember the reason
The best moment to make a brief relevance note is usually during or immediately after meaningful engagement with the paper. At that point, you still remember what prompted you to save it, what surprised you, and which part of your research problem it seemed to illuminate.
You can revise that judgment later. In fact, you probably will. Your interpretation of the literature changes as you encounter additional evidence.
The aim is not to predict exactly how the paper will appear in the final manuscript. It is to preserve your current reasoning so that you do not have to reconstruct it from scratch later.
Make the relevance note specific enough to be useful
Notes such as “important,” “useful,” “good paper,” or “cite this” tell you almost nothing when you return months later.
A useful relevance note should usually identify the connection:
- Supports the proposed relationship between X and Y in undergraduate samples.
- Contradicts the earlier studies in Theme B, possibly because this study measures X differently.
- Useful boundary case because the effect appears in experienced users but not novices.
- Provides the theoretical definition I am using for construct X.
- Shows that most existing evidence comes from cross-sectional designs.
- Potential evidence for the gap concerning rural institutions, although the sample is small.
Notice that these statements do not merely repeat findings. They position those findings relative to a larger intellectual task. This is closely related to tracking whether studies support, contradict, or complicate an argument.
Relevance should remain separate from extracted evidence
Your relevance note is an interpretation. Treat it accordingly.
If the authors report a particular result, preserve the result accurately and keep enough source information to locate it again. Then record your assessment of its significance separately. This makes it easier to distinguish evidence from interpretation and to reconsider your judgment as your review develops.
That separation becomes particularly useful when you later record your own interpretation alongside extracted facts. The paper itself should remain recoverable beneath the analytical layer you add to it.
Watch Out
“Why this paper matters” should never become permission to distort what the study actually found. A source may be convenient for your argument and still provide weak, indirect, qualified, or contradictory evidence. Preserve those qualifications when recording its relevance.
Relevance notes make later synthesis easier
The payoff becomes apparent when you stop reading and begin writing. A literature review normally needs to explain relationships among studies rather than simply reproduce one summary after another. Educational guidance on synthesis matrices therefore emphasizes comparison across sources and identification of agreements, disagreements, methodological differences, and gaps.
If your notes contain only descriptions, much of that analytical work still has to happen at the drafting stage. If they also preserve why sources matter, some of the intellectual groundwork for synthesis has already been recorded.
This does not mean you should draft your final argument while reading every article. Your interpretation remains provisional. It means your notes can capture the connections you notice rather than throwing those connections away.
06 · What This Means for You
Add One Analytical Question to Your Reading Routine
After recording the information you need from a paper, ask one additional question: Why might this source matter to the review I am actually trying to write?
Your answer does not need to be elaborate. It needs to be specific enough that your future self can recover the connection without rereading the entire article.
A simple decision framework
If the paper provides evidence directly relevant to your research question
Record which claim, relationship, theme, or uncertainty the evidence informs.
If the paper disagrees with other studies
Record the disagreement and any plausible methodological, conceptual, population, or contextual difference worth investigating later.
If the paper matters mainly for theory or definition
Record the specific concept it helps define, distinguish, or explain.
If the paper appears relevant but you are not yet sure how
Say exactly that and record the tentative connection rather than manufacturing certainty.
If you cannot identify any meaningful contribution after reading it
Do not force one. The paper may simply be peripheral to the review you are developing.
This practice also helps prevent a literature review from becoming a collection of papers you accumulate but never meaningfully revisit. The goal is not to make every saved paper useful. It is to make the usefulness, or lack of usefulness, visible.
07 · A Quick Checklist
What to Record After Reading a Relevant Paper
Before you consider your notes finished, check:
Can I tell what the authors actually found, argued, or concluded?
Have I preserved page, table, figure, section, or other location information when I may need to verify a specific claim later?
Have I written why this paper matters to my particular research question, theme, or argument?
Is my relevance note more specific than “important,” “useful,” or “cite later”?
Have I kept my interpretation distinguishable from the authors' own claims?
Have I noted whether the paper supports, contradicts, complicates, qualifies, or redirects an emerging idea when that relationship is already apparent?
Could I understand this note several months from now without rereading the whole paper?