Count case-insensitive literal word or phrase matches in pasted text. Review exact spacing and word-boundary behavior; this page reports occurrences, not a density percentage.
The keyword frequency checker searches pasted text for a supplied literal keyword and reports how many matches it finds. Paste the passage, enter the word or phrase and select Check Frequency. Unlike the token-density table on another page, this tool can search a multi-word string. It does not produce a percentage, extract a vocabulary list or calculate a recommended number of mentions for an article. Its output is a count under a specific matching rule.
Use it when you want to inspect repeated wording, verify that a phrase appears in a draft or compare the same expression between two versions. Select the relevant article body rather than including navigation and unrelated page components. The function cannot decide which copied words belong to the main content. Keep the sample boundaries consistent if you intend to compare counts; otherwise a new heading or footer may account for the difference you attribute to an edit.
The function lowercases both the passage and keyword, trims the keyword's outer whitespace and escapes characters that would otherwise have special meaning in a regular expression. It then wraps the escaped keyword with word-boundary checks and searches globally. This means you are supplying literal text, not a custom pattern language. A dot or plus sign in the input is not deliberately interpreted as a wildcard or repetition command by the search.
Escaping is only one part of the rule. The added word boundaries still affect whether a literal phrase can match at its edges, especially when it begins or ends with punctuation. Do not assume that every visible substring is counted simply because special characters are treated literally. Test a representative term containing symbols before using the result as a definitive count, and distinguish the matched text from a dictionary definition or a language-aware word analysis.
For the keyword “cat”, “Cat cat catalog” gives two matches after lowercasing. The longer word “catalog” does not satisfy the end boundary after the requested letters. “cats” and “catlike” are also different forms, while a hyphen can provide a boundary around a part of a compound. These examples explain why this count is not the same as a plain substring search or an automatic grouping of all related vocabulary.
The checker does not stem words, recognize synonyms or connect grammatical variations. If your editorial question includes several forms, inspect them separately or use another analysis method. A zero for “run” does not imply that “running” is absent, and a count for one phrase cannot establish complete coverage of its topic. Use the output to locate exact wording, then read the surrounding passage to assess what those occurrences actually contribute to the explanation.
Internal whitespace in the keyword is literal. The phrase “seo tools” matches that one-space sequence in the lowercased text, but does not automatically match two spaces, a tab or a newline between the words. Trimming removes only the keyword's outer whitespace; it does not normalize its internal spacing or rebuild the passage. A phrase that looks continuous after visual wrapping can still contain a real newline copied from another source.
For a small test, enter a passage with “seo tools”, “seo tools” and the same words separated by a newline. Searching the one-space phrase counts only the matching form under this rule. If your task should consider those representations equivalent, normalize a working copy deliberately and document that preprocessing. Do not silently change a quoted source and present the result as a count of its untouched original string.
Word boundaries in this regular-expression approach are based on transitions involving basic word characters. A term such as “C++” can behave differently from an ordinary alphabetic word because its final character is punctuation. At a normal word ending followed by a space, the desired boundary may be absent after that punctuation. This is a current matching limitation, not evidence that the term never appears in the passage you supplied.
Accented initials and non-Latin scripts need similar care because a basic word-boundary rule does not represent every language's word edges. Hindi text can therefore return unexpected results. Select a language-appropriate method when accurate multilingual counting is required. For technical terms with symbols, test the exact spelling and context rather than changing the keyword until you obtain a convenient number. Keep a note of the edge rule when explaining differences from another search tool.
The global search returns non-overlapping matches. In “go go go”, searching “go go” counts the first two-word occurrence, then continues after that match rather than starting again inside it. The result is one, not two overlapping phrase starts. This distinction matters for repeated short phrases, linguistic examples and test data where neighboring occurrences can share words. A count is meaningful only when the overlap convention fits the question you are asking.
For ordinary editorial checks, non-overlapping occurrences may be sufficient, but document the convention if you compare with another method. This page has no overlap switch. It also does not highlight each match or list its position, so inspect the source manually when the total surprises you. Testing a three- or four-word example is often enough to reveal whether the disagreement comes from overlap, internal whitespace or the boundaries around punctuation rather than a calculation error.
A repeated phrase can be necessary terminology, a quotation or redundant copy. Read its context before removing occurrences. A clear instruction may repeat the exact object name to avoid ambiguity, while a promotional sentence might repeat the same point without adding information. The useful edit is the one that improves the explanation. A smaller frequency count is not itself proof that the passage became clearer, more accurate or more useful to its intended audience.
Likewise, do not add exact phrases just to reach an arbitrary mention count. Google's spam guidance cautions against unnatural keyword stuffing, and this occurrence checker cannot evaluate search quality from a number. Use relevant vocabulary where it belongs, then check facts, examples and missing instructions separately. If the goal is investigating lost search traffic, an exact phrase count is only a narrow inspection of one sample, not a diagnosis of indexing, competition or a ranking-system change.
Select Check Frequency again after editing either field because the displayed sentence represents the last explicit run. The tool requires a non-empty passage and keyword, but it does not validate whether the selected term is linguistically appropriate. Check a known short example first, then save the real sample and exact keyword with any reported count. Include spacing and symbol details if those are central to the matching question you need to reproduce later.
When comparing revisions, note whether you changed the article boundaries, normalized punctuation or added a section that includes the phrase. Those changes affect occurrences without necessarily altering repetition in the original paragraph. Keep the checked draft in your normal editor and use this page to answer the exact-text question it implements. For density, vocabulary tables or multilingual word segmentation, choose the appropriate separate method and compare its definitions before combining the results into a broader content assessment.
Yes, when the literal phrase satisfies the boundary rule. Internal spaces must match the passage exactly.
No. The passage and keyword are lowercased before searching.
No in ordinary alphabetic contexts. The boundary check excludes that longer word from the exact “cat” result.
No. The keyword is escaped and treated as literal text before the boundary checks are added.
No. Internal whitespace is literal rather than normalized automatically.
No. The global search returns non-overlapping matches. “go go” in “go go go” therefore gives one under this rule.
Punctuation at an edge can conflict with the added word-boundary requirement. Test symbol-heavy terms before relying on the count.
No. This page reports occurrences only. The density and ratio pages use their own denominator rules.
No. Its basic boundary approach can behave unexpectedly with accented or non-Latin text. Use an appropriate language-aware method for exact requirements.
No. Select Check Frequency again after editing the passage or keyword.