Will this flag every instance of a repeated word, even intentional ones?

Yes — the tool has no way to distinguish an accidental repeat from a deliberate rhetorical one, so it flags every match meeting its rule (adjacent duplicate, or three-plus occurrences in a paragraph-sized window) and leaves the judgment call to you. Treat flags as candidates to review, not confirmed errors.

How is this different from the word frequency analyzer?

Word frequency analysis ranks every word by total count across the entire document, which is useful for spotting an overall pattern. This tool instead looks for two specific, narrower signals — immediately-adjacent duplicate words (a likely typo) and unusually clustered repetition within a short span (a likely wording tic) — which are easier to act on directly during a line-by-line edit.

Does it flag names that repeat a lot, like in a story?

It can, since character or company names naturally repeat often in narrative or business writing, and the tool doesn't distinguish a proper noun from any other content word. This check is most useful on descriptive prose (essays, articles, reports) where repeated adjectives or verbs are more often accidental than repeated names in a story would be.

← All Tools

Duplicate Word Finder

Flags accidental repeats like 'the the' and over-repeated words in the same passage.

This tool runs entirely in your browser. Nothing you type or paste here is ever sent to a server — see how the calculations work.

This scans for two specific kinds of accidental repetition: immediately-adjacent duplicate words ("the the cat sat"), which are almost always a typo left over from editing, and words that repeat unusually often within a short span of text (like the same word appearing three times in one paragraph), which usually signal a spot worth rewording even if each individual instance is grammatically fine. It's a distinct editing check from /word-frequency-analyzer/, which ranks every word across the whole document rather than flagging localized, suspicious repetition.

Adjacent repeats ("the the") and clustered word overuse appear here.

Adjacent duplicates vs. clustered repetition

Adjacent-duplicate detection compares each word to the one immediately before it (case-insensitively, so "The the" is still caught) and flags any exact match. This catches the extremely common editing artifact where a sentence gets rewritten and a leftover word from the old version accidentally stays next to the new one — a mistake that's notoriously easy to miss when proofreading your own writing, because your eye tends to read what you meant to write rather than what's literally on the page.

Clustered-repetition detection works differently: it looks at a sliding window of text (roughly one paragraph's worth) and flags any content word — excluding common function words like "the," "a," and "is," which naturally repeat constantly in normal prose — that appears three or more times within that window. This surfaces a word you've unconsciously leaned on in one section without you having specified it in advance, which is the same kind of insight /word-frequency-analyzer/ gives across a whole document, applied here at the more localized paragraph level where a repeated word is often more noticeable to a reader.

Both checks are pattern-based, not a grammar or meaning check — the tool has no way to know if a repeated word is intentional (deliberate repetition is a legitimate rhetorical device in some writing) versus accidental. It flags candidates for a human to review, not errors to auto-fix.

Who uses this

Catching an accidental "the the" or "a a" left over from an editing pass, before publishing.
Spotting a word you've unintentionally overused within a single paragraph or section during a self-edit.
A final proofreading pass on an important document (cover letter, essay, contract) to catch a class of error that's easy for the writer's own eye to skip over.

Edge cases to know about

Legitimate repeated words next to each other ("had had", "that that" in certain grammatical constructions, or a name that's genuinely repeated for emphasis, like "No, no, never") will still be flagged even though they're not errors — review each flagged instance rather than assuming every flag is a mistake.
Clustered-repetition flagging uses a fixed window size (roughly paragraph-length) rather than analyzing true semantic proximity, so a word repeated across two back-to-back short paragraphs might not be flagged even though a reader would notice the repetition, while the same word appearing three times within one long paragraph would be.
Proper nouns (a character's name in a story, a company name in a report) are expected to repeat often and will trigger clustered-repetition flags frequently in that kind of writing — this check is most useful for descriptive and evaluative words (adjectives, verbs) rather than names that are supposed to recur.

Related tools

FAQ

Will this flag every instance of a repeated word, even intentional ones?: Yes — the tool has no way to distinguish an accidental repeat from a deliberate rhetorical one, so it flags every match meeting its rule (adjacent duplicate, or three-plus occurrences in a paragraph-sized window) and leaves the judgment call to you. Treat flags as candidates to review, not confirmed errors.
How is this different from the word frequency analyzer?: Word frequency analysis ranks every word by total count across the entire document, which is useful for spotting an overall pattern. This tool instead looks for two specific, narrower signals — immediately-adjacent duplicate words (a likely typo) and unusually clustered repetition within a short span (a likely wording tic) — which are easier to act on directly during a line-by-line edit.
Does it flag names that repeat a lot, like in a story?: It can, since character or company names naturally repeat often in narrative or business writing, and the tool doesn't distinguish a proper noun from any other content word. This check is most useful on descriptive prose (essays, articles, reports) where repeated adjectives or verbs are more often accidental than repeated names in a story would be.