Experienced reviewers and editors can spot AI-assisted prose within the first paragraph. The tells are not grammatical errors. They are patterns of diction and structure that no experienced researcher would produce.

The most common offender is “delve into.” No management scholar writing a methods section has ever chosen that phrase unassisted. Similarly, “it is worth noting” does nothing that a period and a new sentence cannot do better. “Paradigm shift” belongs in conversations about Kuhn and nowhere else. “Robust” has a specific meaning in statistics; using it as a general intensifier marks the writer as someone who has not read enough empirical work to know the difference.

“Leverage” as a verb meaning “use” is corporate jargon, not academic prose. “Shed light on” is a metaphor that has been dead so long it has decomposed. “Navigate the complexities” is the kind of phrase that makes editors reach for the reject button before finishing the abstract.

The fix is not to memorise a banned list. The fix is to read enough published work in your target journal that you absorb its register. AI tools are useful for generating first drafts. They are not useful for knowing what good academic prose sounds like. That knowledge comes from reading, not prompting.

Welcome to WordPress. This is your first post. Edit or delete it, then start writing!

The introduction is the most rewritten section of every published paper. AMJ Best Article Award winners report spending 24 percent of their total writing time on the introduction and rewriting it an average of ten times. Most submitted manuscripts rewrite it zero times after the first draft.

Grant and Pollock (2011) identified three questions every effective introduction must answer. Who cares? What do we know, what do we not know, and so what? What will we learn? Most desk-rejected introductions fail on the first question. They assume the reader already cares about their topic and skip directly to the gap.

A gap is not a contribution. “No one has studied X in the context of Y” is not a reason to study it. The question is whether studying X in Y changes what we understand about X. If the answer is no, the study is descriptive, and descriptive studies face an uphill battle at top journals.

An effective introduction is short. Three to four double-spaced pages. It does not describe the structure of the paper at length. It does not overpromise. And it does not write cheques the rest of the article cannot cash.

Zhang and Shaw (2012) reduced the requirements for a publishable methods section to three words: completeness, clarity, and credibility. Most rejected methods sections fail on the first criterion.

Completeness means explaining not just what you did but why you made each decision. Why this sample? Why these measures? Why this analytic approach? A methods section that reports procedures without justifying them leaves the reviewer to guess whether the choices were principled or convenient. Reviewers do not guess charitably.

Clarity means a reader can follow your procedures without rereading. Variables should appear in the same order in the methods, the results tables, and the discussion. Construct definitions should precede operationalisations. If you use a scale, name it and cite it before reporting its properties in your sample.

Credibility means demonstrating that your design matches your question. Cross-sectional data cannot test causal arguments. Samples of 47 cannot detect moderate effects. Control variables must meet three conditions: correlated with the dependent variable, correlated with the independent variable, and not a more theoretically central variable than what you are studying.