Choosing Keywords That Get Your Paper Found

Why Keywords Matter

Keywords serve two purposes, though authors tend to think of just one. They control how readers locate your article in databases, and they help the editor send your manuscript to suitable reviewers.

Poorly chosen keywords therefore carry a double cost: you reach fewer of the right readers, and you are more likely to be assessed by someone whose expertise lies a little away from your topic.

How to Choose Them

1. Jot down the search-engine words you would use to track down your own paper if somebody else had authored it.
2. Also list standard vocabulary covering your method, the population or dataset studied, and where the work took place.
3. Look at which terms recent papers on the same topic use, and not the ones you would like to see become standard.
4. Add one broader term so that readers browsing the general area come across your work.
5. Add one specific term that sets your paper apart from neighbouring work.

Balance specificity on purpose: keywords that are all broad lose you among thousands of results, and keywords that are all narrow mean nobody beyond your immediate circle ever sees the paper.

How Many to Use

Four to eight is what most journals request. Look at the submission portal for the current requirement. Within that range, more does not help, since every extra keyword weakens the signal about what the paper is truly about.

Terms to Avoid

• Terms your title already uses — databases generally index the title anyway, so listing them again squanders slots
• Invented or in-house terms — if only your own research group uses the phrase, no one will search for it
• Very broad single words — "analysis", "system", "study" and "method" give no discriminating information
• Abbreviations without the expansion — give both forms when the abbreviation is really standard
• Aspirational terms — avoid tagging a paper with a fashionable technique that you did not actually use

Testing Your Choices

Try each candidate keyword in the database your readers rely on. If the results resemble your paper, the term works. If they come from an unrelated field, the term is ambiguous and will steer your paper to the wrong audience.

Next, try the reverse: search for the problem your paper solves, in the words a newcomer would use, and check whether papers like yours turn up. If they do not, you have spotted a gap that your keywords ought to fill.

Working With Title and Abstract

Keywords form just one piece of a discovery set that also contains the title and abstract. Terminology should be consistent across all three, because a paper whose title says "machine learning", whose abstract says "statistical learning" and whose keywords say "AI" is harder to find than one that chooses a term and sticks to it.

To place this in context, consult Abstract Writing: What Every Sentence Must Do along with Writing a Research Paper Reviewers Can Assess. Precise keywords further enrich the metadata deposited alongside your CrossRef DOI.

Checklist

• Within the journal's stated number
• Mix of broader and specific terms
• Terms that current papers in the field actually use
• No repetition of title words
• No invented or in-house vocabulary
• Each term tested in a real database search
• Consistent with the terminology in the title and abstract

At-a-Glance Checklist

1. Specific, not generic“Solar PV inverter efficiency” rather than “energy”
2. Include synonymsAdd the term a reader in another field would use
3. Spell out abbreviationsUnless the abbreviation is what people search
4. Test in a search engineResults should resemble your paper
5. Within the portal limitFollow the count the submission form asks for

Questions Researchers Often Ask

More Researcher Resources

This guide belongs to the JMVRA Researcher Resources collection, offering practical advice that authors and researchers can verify, free of promotional claims.

Browse all resources

Related Reading

Further reading on the same theme from the JMVRA blog.