What a title has to do
A title works in three places. In a database search or a table of contents, it is usually the only text a reader sees before deciding whether to open the abstract. For an editor screening submissions, it is the first signal of what the paper reports and whether it fits the journal. For search engines and indexing services, its words help determine whether the paper is found at all. An effective title serves all three: it is accurate, specific and findable.
Ten recommendations are common: be clear and concise, capture the essence of the study, use keywords, avoid jargon, create intrigue, highlight outcomes, be original, consider length, test the title with colleagues and revise it. Most of these hold up well. Two need care, as the evidence below shows: intrigue can cost citations, and outcomes must be stated no more strongly than the results allow.
What to include
Most effective titles combine some of the following elements. Not every title needs all of them, but leaving out the population or design is a common reason a title is too vague to be useful.
| Element | Example wording | When it matters |
|---|---|---|
| Topic or main variables | Digital payment adoption | Always |
| Population or setting | among small independent retailers in a mid-sized city | When results may not generalise |
| Design or method | a mixed methods study | When a reporting guideline requires it, or when the design is a selling point |
| Main finding (optional) | Perceived fraud risk outweighs transaction cost | In fields where results-stating titles are accepted |
| Time period (optional) | 2015 to 2025 | For historical, trend or policy studies |
Use the words your readers would search for, ideally the terms used in the abstract and keywords, and spell out abbreviations unless they are better known than the full form. The APA manual advises titles that summarise the main idea simply and stay focused and succinct (American Psychological Association, 2020), and the APA reporting standards for quantitative studies ask that the title identify the main variables and the relationships studied (Appelbaum et al., 2018).
Reporting-guideline requirements
For some study types the title is not left to taste. Major reporting guidelines include a title item, and many journals require authors to follow them.
| Study type | Guideline | Title requirement |
|---|---|---|
| Randomised trial | CONSORT | Identify the report as a randomised trial in the title |
| Systematic review | PRISMA 2020 | Identify the report as a systematic review |
| Observational study | STROBE | Indicate the design with a commonly used term in the title or abstract |
CONSORT 2010 asked authors to identify a randomised trial in the title (Schulz et al., 2010), and the guideline was updated as CONSORT 2025 (Hopewell et al., 2025); check the current checklist when you submit. PRISMA 2020 asks that systematic reviews be identified as such (Page et al., 2021), and STROBE asks authors of cohort, case-control and cross-sectional studies to name the design in the title or abstract (von Elm et al., 2007). Naming the design helps readers and indexers classify the paper correctly.
What the evidence says about length and title type
Several studies have looked for associations between title features and how often papers are read or cited. They are observational, so they show associations rather than effects, but they point in a consistent direction.
- Letchford et al. (2015) found that journals publishing papers with shorter titles received more citations per paper, consistent with the idea that shorter titles are easier to understand.
- In 423 research articles from PLoS and BMC journals, Paiva et al. (2012) found that short titles had higher viewing and citation rates, that titles describing results were cited more often than titles describing methods, and that question marks, geographical references and colons were associated with fewer citations; after adjustment, only length and type remained independent predictors.
- In 2,172 articles from six PLoS journals, Jamali and Nikzad (2011) found that articles with question titles tended to be downloaded more but cited less, that longer titles were associated with slightly fewer downloads, and that titles with colons were associated with fewer downloads and citations.
The practical message is modest: keep the title as short as accuracy allows, prefer a clear statement to a question unless the field expects questions, and do not rely on a clever subtitle to carry the meaning. Field conventions matter too. Results-stating titles are common in some biomedical journals but rare in much of the humanities and social sciences, and some journals discourage them. Read recent titles in your target journal, and see how to select a journal.
Titles across fields and article types
Conventions differ more than general advice suggests, so treat your target journal as the main guide.
- Humanities and much of the social sciences. Two-part titles joined by a colon are common, often with a phrase or quotation followed by a descriptive subtitle. If you use this form, make sure the subtitle alone tells a searcher what the paper is about, because the first part is rarely searchable.
- Biomedical and health research. Titles usually name the condition, population, intervention or exposure and design. Results-stating titles are accepted by some journals and discouraged by others.
- Engineering and computer science. Titles often name the method or system and the problem it addresses. Avoid naming a system with an acronym alone; readers will not search for a name they have never seen.
- Reviews. Say what kind of review it is, for example systematic, scoping or narrative, and what it covers.
- Conference papers and preprints. If the paper later appears in a journal, the title may need to change to fit the journal's style; keep the key search terms so the versions can be linked.
Whatever the field, the title, abstract and keywords should use the same core terms, so that a reader who finds the paper through one meets the same vocabulary in the others.
Some words are best avoided in most fields. "Novel", "first" and "unique" claim priority that reviewers may dispute and that the title cannot prove. Words such as "effect" or "impact" imply causation, which an observational design may not support; "association" or "relationship" is more accurate in that case. Many journals also ask for a short running title, often with a character limit; write it at the same time so that it is not improvised at submission.
If you write in English as an additional language, compare your title with five recent titles in the target journal on similar topics. Patterns of word order, use of articles and the choice between noun phrases and full sentences become clear quickly, and matching them makes the paper look at home in the journal.
Worked example: revising a title
The following example is illustrative and hypothetical. It shows one title revised in four steps.
| Version | Title | What changed |
|---|---|---|
| 1 | A study on the impact of various factors on digital payments | Starting point: vague, no population, no design |
| 2 | Factors influencing digital payment adoption among small retailers | Topic and population added; filler removed |
| 3 | Digital payment adoption among small independent retailers: a mixed methods study | Design named; more specific population |
| 4 | Perceived fraud risk outweighs transaction cost in small retailers' adoption of digital payments: a mixed methods study | Main finding stated, for a journal that accepts results-stating titles |
Version 3 suits most journals. Version 4 is appropriate only if the finding is robust and the journal's recent titles use this style. Phrases such as "a study on", "an investigation of" and "the impact of various factors" add length without information and can usually be cut.
A drafting and testing process
- Write down the paper's main question and main finding in one sentence each.
- List the four or five words a reader would type to find this paper.
- Draft three titles: one descriptive, one stating the finding, one naming the design.
- Check the journal's author instructions for limits on length or characters, rules on abbreviations and any required reporting guideline.
- Cut filler words and check that every remaining word carries information.
- Ask two colleagues, one in your specialism and one outside it, to say what they expect the paper to report from the title alone.
- Revise the title after the paper is finished, so that it matches what the results show.
Keep the title consistent with the abstract and keywords; see how to write an abstract and how to write a journal manuscript section by section. If the paper comes from your thesis, the thesis title will usually be too broad for a single article; see how to convert a PhD thesis into a journal manuscript.
Common mistakes
- Titles that name a broad topic but not the population, setting or design.
- Filler phrases such as "a study of" or "some aspects of".
- Unexplained abbreviations or in-group jargon.
- Question or teaser titles in fields where readers expect statements.
- Stating a finding more strongly than the results support, or implying causation from an observational design.
- Ignoring the title item of the relevant reporting guideline.
- Reusing the thesis title for a single article.
For annotated manuscripts, browse sample work; for what reviewers look for, see how to peer review a manuscript.
Key takeaways
- A good title is accurate, specific and findable: topic, population or setting and, where relevant, design.
- Reporting guidelines require trials, systematic reviews and observational studies to be identified by design (Schulz et al., 2010; Page et al., 2021; von Elm et al., 2007).
- Shorter titles and results-describing titles are associated with more citations, but the evidence is correlational (Letchford et al., 2015; Paiva et al., 2012).
- Question titles may attract downloads but were cited less in a large study of PLoS articles (Jamali & Nikzad, 2011).
- Check the journal's instructions and current guideline versions, and revise the title after the paper is finished (Hopewell et al., 2025; American Psychological Association, 2020).
