Buy this report only
A one-time payment unlocks the full results of this check. No subscription required.
Buy this reportAnswers about preparing your data, results and tool limitations.
The tool processes a list of keyword phrases and counts normalized tokens across the list. Select the input language independently of the interface. Russian uses morphological lemmatization; English currently uses lowercase token normalization and English stopwords, keeping word forms distinct. The result helps identify repeated vocabulary and topics. Keep the original phrases as well: token counts do not preserve their complete meaning or search intent.
For example, start with a small set of phrases from one service or product category. Review the output alongside your existing pages and business scope, then expand the workflow to the rest of the list. Changing the interface language does not translate saved source phrases or replace an existing report.
Paste one phrase per line and remove irrelevant data such as URLs, spreadsheet headings or unrelated notes. The limit is 2,000 phrases. Choose the correct input language and decide whether stopwords should be included. Blank lines do not contribute to the phrase count.
Normalization removes case differences, and Russian morphology additionally groups supported grammatical forms. This makes recurring vocabulary easier to inspect. English word forms currently remain separate. The output is a vocabulary summary, not a substitute for grouping complete search phrases by intent.
By default the tool hides common function words and excluded tokens to make the useful vocabulary easier to read. Enable the option to inspect what was filtered. The hidden counts describe the excluded tokens; they are not missing source phrases or an indication of lost project data.
Use frequent content words to spot topics, attributes and irrelevant directions. Return to the original phrases before choosing page titles or groups: one normalized token can appear in several different intentions. Do not turn the frequency list into a requirement to repeat words a fixed number of times.
It counts appearances of the normalized token in the submitted phrases, including repeated occurrences within the list. It is not Wordstat search volume, unique-user demand or recommended keyword density. Compare counts only when the input and processing settings are comparable.
Short, ambiguous, specialized or misspelled words can have several possible interpretations. Russian morphology chooses a dictionary analysis without the full meaning of every phrase. English currently preserves word forms after lowercasing. Inspect important terms in their original context before making content decisions.
Copy the normalized-token list using the result controls. Keep a separate copy of the original phrases for filtering, clustering and content planning. The token table is a derived summary and cannot reconstruct the original list or its grouping on its own.